Page 1

First published in the UK by Gambit Publications Ltd 2003 Copyright © Steve Giddins 2003 The right of Steve Giddins to be identitied as the author ofthis work has been as­ serted in accordance with the Copyright, Designs and Patent� Act 1988. All r ights reserved. This book is sold subject to the condition that i t shall not, by way of trade or otherwise, be lent, fe-sold, hired out or otherwise circulated in any form of binding orcoverotherthan that in which it is published and without a similar condition including this condition being imposed on the sub� eque nt purchaser. A copy of the British Librmy Cataloguing in Publication data is avai lable from the British Library. ISBN I 901983 897 DISTRrBUll0N: Worldwide (except USA): Central Books Ltd, 99 Wallis Rd, London E9 5LN. Tel +44 (0)20 8986 4854 Fax +44 (0)20 8533 5821. E-mail: USA: BHB International, Inc., 302 West North 2nd Street, Seneca, SC 29678, USA. For aJl other enquiries (including a full list of all Gambitchess titles) please contact the publishers, Gambit Publications Ltd, PO. Box 32640, London WI4 OJN. E-mail: Or visit the GAMBIT web site at Edited by Graham Burgess Typeset by Petra Nunn Printed in Great Britain by The Cromwell Press, Trowbridge, Wilts.

10 9 8 7 6 5 4 3 2 I

Gambit Publications Ltd Managing Director: GM Murray Chandler Chess Director: GM John Nunn Editorial Director: FM Graham Burgess Gennan Editor: WPM Petra Nunn





5 8


The Keys to Successful Opening Play


Variety - the Spice of Life?






Main Roads or Side-Streets?



Move-Orders and Transpositions



Use and Abuse of Computers






Infidelity and Divorce



Some Players' Repertoires Analysed


Index of Players


Index of Openings




++ # "

!? '!!





t l'


-+ Ch Chi Web

check double check checkmate brilliant move good move interesting move dubious move had move blunder White is winning White is much better White is slightly better equal position Black is slightly better Black is much better Black is winning championship team championship world championship

Weht world team champi(ln�hip Ech European championship Echt European team championship Eee European Clubs Cup Ct candidates event IZ interLonal event Z zona] event olympiad OL Jt junior event WOIll women's event tpd rapidplay game It team tournament COIT. correspondence game \-0 the game ends in a win for White liz_IiI the game ends in a dr.IW 0- I the game ends in a win for Black nth match game In) see next diagram (D)


Openings: Can't Live with 'em, Can't live without 'em

ending" or "If only I could play fixed pawn-structures I)CHer", or something similar? If your team-mates and friends are anything like mine, the answer

Openings are an area of the game with

will be "Hardly ever". Instead, what

which most chess-players have a love­

OTle hears in the great majority of cases

hate relati onship On the one band, al­

is someth i ng like "He knew the open­


most all serious chess-players lake a

ing better than me", or "It's that open­

great interest in the opening phase of

ing. I'll have to give it up; I always

Ine game. We spend a very large pro­

lose with iL".

portion of our chess sludy time


The truth of the mattcr is that the

npenings. and an equally large propor­

great majority of players below master

tion of our chess book moncy goes on

level spend a disproportionate amount

opening books. Although many of us

of their chess time on openings, yet

would claim that we regret the large

achieve very little in the way of con­

role played by opening theory in mod­

crete benefits. There arc a number of


chess, the hard facts tend to give

rcason!.; for this - changing openings

'nlC failure of Fi�chcr­

too often, over-concentration on rote

the lie to this.

random chess to attract much interest

learning of variations, at the expense

rmln the great IllaJority of us attests to

of undcr>landing the po sitio ns

the fact that, whatever we might like to

great a willingness

tell ourselves, most of us have little

rather than using their own judgement,



to tru�{ authority

real desire to escape from the vast edi­

etc. Ahove all, too few players under­

rice that is modem opening theory.

stand how to study openings, and how

However, on the other hand, we are

to form an opening repertoire. The

almost all sure that openings are to

negative results of this extend to the

Illame for the majority of our defeats,

;I[J(1 that we would be much stronger

whole of one's game. Because we spend

liI,lyers "jf only I knew my openings properly". Just think of all the times

neglect other areas of the game. [n ad­


have heard team-mates and friends

,ompiaining after losing a game. How "lien have you heard them say "Well, I lost hecause I misplayed the rook


much time on openings, we

dition, when we lose a game, we fre­ quenlly blame the opening, decide to learn something else, and so spend yet more time on openings, oblivious to the fact that the real reason we lost the



game in question was because we mis­ played the ending! The aim oflhis book is to try 10 help readers rectify this sorry state of af­ fairs, by ensuring that their opening studies are conducted efficiently and effectively. With properly-directed ef­ forts, it i� not �o difficult to have a reli­ able and effective opening repertoire, in much less time than most of u� ha­ bitually spend on openings. Of course, having such a repertoire is not going to stop us losing games, nor will it ensure that we never emerge from the open­ ing with a bad position. However, it will reduce the number of times these things happen. and it should also help us to focus our attention on the real cause of our defeats, which in practice is rarely the opening. One thing the book does not attempt to do is to recommend any spccific opening rep­ ertoire per se, still less one that guar­ antees you a win in under 20 moves against any defence there are plenty of other authors out there who offer such miracle cures, if that is what you are seeking. Instead, what I have tried to do is to help you decide for yourself which openings to play, but in so do­ ing. I have discussed a large number of specific opening systems, and my opinions on these will be apparent from the text. You may legitimately be wondering what qualifie� me to give advice on such a subject. I would freely admit that for much of my chess career, I had no idea at all how to approach the problem of openings. Up until 1993, by which time I was in my 30s and

ntted about 2250, I had an extremely weak opening repertoire. In fact, I had almost no repertoire at all. because I chopped and changed openings virtu­ ally on a monthly basis. 'W'hat I did have was a very good all-round knowl­ edge of lots of openings, chiefly he­ cause I happen to be blessed with an excellent memory. However, there was no single opening which I had played more than a handful of times. or of which [ could claim to have any real understanding. And, inevitahly, every time I lost a game, I put il down to the opening, made a mental note not to play that line again. and consigned my scoresheet to the dustbin. My first steps on the path to righ­ teousness came when I started living and working in Moscow in late 1992, and was exposed for the tirst lime i n my life to an experienced chess trainer. I still recall my embarra<:sment when he asked me to write out my whole opening repertoire, including which lines I played against each main black defence, etc., and I was forced to con­ fess that in all but one or two cases, I simply could not say! I would just make my mind up at the board, proba­ bly picking some line which I hap­ pened to have seen in it game in a recent magazine. After rolling his eyes in disbelief for a minute or two. he gave me a severe dose of the "every Russian schoolboy knows better than that" routine, and then we started some serious opening work for the firs! time in my life. Under his guidance, I soon began to develop a proper, cohesive repertoire, and within a couple of


y.:ars I already had a whole series of lines which I bad studied, played and analysed enougb to have developed at kast a modicum of understanding of what I was doing. Needless to say, my cunfidence soared, and my results fol­ luwed suit Of course, not many players ha ve Ihe opportunity to gain access to an experienced Soviet-era chess trainer. My aim in Ihis book is to try at least partially to bridge the gap, by passing 011 10 my readers some of the things I have learned about opening prepara­ I il HI, and so to enable them to develop :1 wel l balanced and efe f ctive I""l )(:rtoire with a minimum of effort. WI' all enjoy chess more when we are w i nning, and while having a decent "I't�ning repertoire is not in itself a ";Isl-iron guarantee of success, it is ( I"lainly a very good step in the right ,IJr,'dion. .





Acknowledgements Thanks are due to Graham Burgess, for editorial assistance, and for the drunken evening in Gausdal which spawned the idea for this book. Many players, inadvertently or otherwise, gave invaluable in sights into their own approach to openings, and their con­ tribution� are acknowledged in the ap­ propriate place in the text. By far my greatest debt is to 1M I gor Belov, my chess iI'ainer when I lived in Moscow and the man who first taught me the difference between a .�erious chess­ player and a player who works seri­ ously at chess. Finally, this book is dedicated to my friends Dave, Roger, and the late John; without their generous encour­ agement of an enthusiastic kid al the local chess club, tbis book would ncver have found its author. ,

1 The Keys to Successful Opening Play

The Slav, GM Matthew

him to play well in one specific posi­

Sadler enumel1lted three key elements

lion. Teach him why the moves are

in successful opening play:

played, and you enable him to play

In his book

1) Knowing the aim� of the open­ ing; 2) Knowing the value of move­


well in lot� of positions. The player who depends entirely on his memory, with little real understanding of the ideas behind the muves, will be com­

3) Understanding typical positions.

pletely at sea the moment his 0ppo­

In this excellent advice, with which

nenl deviates from the 'book', or he

I concur 100%, there are a number of

reaches the end of his memorized

things I should like to comment on in

line. By contrast, the player who un­

more detail.

derstands the opening will be able to

Fishing for Compliments

sonahle move, even if he is blissfully

find the best move. or at least a rea­ unaware of what Grandmaster Anono­

The tlrst thing to note is that memoriza­

vich recommended in his lalest open­

tion of opening lines is not mentioned

ings hook.

at aIL Contrary to what one may think,

The following game is an excellent

memorizing variations is a relatively

iHustration of the value of understand­

small factor in enabling one to play an

ing an opening, rather than merely

opening successfully. It is far more

learning lines by heart It also proves

important to understand the positional

that even at the highest level,

and tactical themes ()f the opening,

standing is more important than mere

and to appreciate what you should be



aiming for in the given position. It is rather like the popular aphorism, be­ loved of humanitarian relief agencies: "Give a man a fish, and you feed him for a day_ Teach a man how to fish, and you feed him for life." Chess openings are rather similar. Teaeh a player a se­ ries of opening moves, and you enable




Pamp!ona /999/00

1 d4 e6 2 <'i.'Jf3 fLlf6 3 c4 d5 4 4k3 i.e7 5 ;tg5 h6 6 :Jih4 0-0 7 e3 b6 8 Jid3 �b7 9 0-0 ti'lbd7 10 �g3 c5 11 'l!i'e2 fLle4 12 cxd5 exd5 13 :tadl (D)



certain squares, how didShort'sapplication of general principlesleadto his choice of move? Well. itis obviousthat the opposition


ofthe white rook on d1and the black queenon dB is awkward for Black. As we saw fromtneVyzhmanavin exam­ ples ahove, th e pin on the

d5-pllwn means that Blackmust take on c3before recaptur­ ing. byI3... soShort came upwith the ideaof sim­ This posilion had arisen a number "I

t imes prior to this game, with

While's13thmove having emerged as II While.

pl ymoving his queen off the df- ile.

this respect. hisqueen' srook 011 dI, ratherthan his ki ng 's rook, helps Short's idea; theking's rook on dI,

VY/,hmanavin had played 1 3 :a.adl \,'wral tim es, with excellent results.

I 'or example, Illn� advantage f o llo wing

13 ...!bd6 f

II dxc5lbxc315 bxc3.i,xe516 CDd4.

I\uolher move which had been tried

in vit e 14 nac I. Another factorin fa­ vour of 13..:t!fc8 is that once the knight leaves d7, ableto go to e6, atypical blackqueenin structures where he has hangingpawDs on d5 and c5. Short's move hadan immediate ef­

W:IS 1 3... dxr5hxc51 6�hl


h;ll I ging

long time andfaikdtocome upwith

lia hi ilt y in Kramnik-¥usupov,

an effective

1l11llld I99S. Instead,

14 j.bI

Short pointed out that if White

plays" /(1 V yzhm an avin wilh 14 d xcS ,

live newmove forBlack:

1 3. :i!Vc8!

Black has1 4 ...QJxc3 1 5bxc3 and now


l'Of ourpurposes, nis commentsin

Nnl' in Chess (\12000, p.

IS ...tLlxc5 ispossible.Bla ck will fol· low up w it h...tDe4

and ... �tf6,

excellent play.

llIore significant "'Although I had seen [tbe Kramnk i ­

Ylisu pov game J, I hadforgotten every­ l I lillg.

14 ..'bdf6 15 �c2 .••

Thehattery al ong the bl-h7 diago­

nal is not very eifective here,

Illlfinally try to work o ut !"'llcral

because Lautier thought for a

prin ci pl e s. If I

things by

can't under­

slalld whyI am putting my pieces on

Black is neverlikely to bein real dan­ ger of beingmat

15 Jk816 ..ie51lt'e617 dxc5 bxc5 ••



It is clear that Black has solved his development problems very well and has no difficulties at all here. In fact, Short was already feeling contiden t enough to turn down a draw offer at this point, and he went on to win a highly impressive game. To my mind, thi� is a really excellent ex.ample of the value of understand­ ing one'� chosen opening, rather than merely learning variations. Thanks to his many years of playing the black side of the QGD, Short was able, ovcr the board, to find an effective answer to a line which had caused significant trouble for Black. His words in thc note to Black's 13th move are espe­ cially noteworthy.

you decide that you would like to play this line. You therefore spend several weeks studying the position aftcr 1 d4 d5 2 c4 e6 3.'be3 'Df6 4 cx.d5 exd5 5 iLg5 :i.e? 6 e3 0-0 7 i.d3 lLlbd? 8 It)ge2 c6 9 'ti'c2 �e8 10 0-0 'DfS I [ f3 (D).


Order out of Chaos The second of Sadler's three elements is knowing the value of move-orders. This is so important that I shall devote a whole chapter to it later in thc book. For the present, suffice it to say that it is no good knowing an opening really Well, if you then play a movc-order which allows your opponent to trick yOll into some other line that you know nothing about. In preparing our openings, therefore, we must take ac­ count of move-order �uhtletics and transpositions. As an example, let'� suppose tha t you are starting your opening reper toire from scratch, and decide to open 1 d4. Against the QGD, you have been very impressed by a couple of games you have seen where White played the Exchange Variation with tbge2, and ­

By the time you have studied such classics as Botvinnik-Keres, USSR Ch (Moscow) 1952, and Kasparov­ Short, London PCA Wch (15) 1993, and others, you are all ready to crush any un�uspecting QGD player with a mighty pawn avalanche c4-e5-f4-f5, etc. You then turn to the line [ d4iDf6 2 c4 c6. After some consideration. you decide that you reatly like some of Kasparov's wins in th e line 3.!tJrJ b 6 4 a3, so you decide to avoid the Nimzo­ Indian and play 3 !DB instead. Being a thorough chap you do not forget to prepare something against the £ogo­ Indian. 3. . .Jtb4+, as well. Eventually, the great day comes when you wheel out your d-pawn in a tournament game. The gamc starts I .



t!4 �f6 2 c4 e6 3lLl13, but now the un­ speakable rogue plays not 3...b6, nor \ . �b4+, but 3...dS. Suddenly the aw1"111 truth dawns. If you go back to a <)GD with 4 lbc3, you cannot reach your beloved lLlge2 Exchange V aria­ lion, because your knight is already ,·ommitted to f3. You could play 4 g3, ..

Iransposing into a Catalan, but this is an opening you have never studied,

and some of the lines· can be preHY ,·omplicated, especially those where Black takes on c4 and tries to cling


you have chosen. This is probably the main area where players below master level fall down. It is relatively easy to understand the main points of an open­ ing, to master jts move-orders and to memorize a few key lines. It is far more difficult to develop a really good understanding of the typical middle­ game and endgame positions. As a first example of the impor­ tance of really understanding typical positions, 1 should like to consider the following position.

onto the pawn - hardly the sort of Ihing you want to play without prepa­ ration. So, the only alternative is the ultra-tame 4 e3, immediately forego­ IIlg any pretensions to an opening ad­ vantage, and certainly not what you had in mind when you were analysing ;111 those Kasparov crushes. In short. you have been 'move­ urdered'. When you decided on your various opening choices, you failed to l·heck whether any of them were in­ nmsistent with another, and now you have paid the price. Don't let it happen to you! Hope­ flllly, after reading Chapter 5, you will

Kramnik - Timman 8elfl,mde 1995

he on the business end of any moV('­

order trickery which happens to be ("Ioating about.

Typical, Just Typical Ille third element is understanding typ­ Il·al positions. This is something which involves going well beyond opening sludy per se, and instead learning to play the types of middlegame and end­ game which arise from the opening

The position is easily recognizable as having arisen from a Queen's Gam­ bit Declined, Exchange Varialion. Most strong players will be aware thai White's principal plan in this struc­ ture is the so-called Minority Attack, whereby the pawn is advanced to b4 and bS, with the aim of capturing on c6 and leaving Black with a backward pawn. In the diagram position, there­ fore, White of course played 16 b5,



didn't he? Well, no, actually h e didn't.

the dark squares, as would have been

Kramnik played...

the case in the line after 16 b5, he now threatens to get a bind himself on

I., bxa5! ...and wrote "A typical deci�ion,

those very same dark squares. A few

since 16 h5 c5 doe� not promi:{(! White

moves later, having used the queen­

as much as he would like"'. This is an

side, threats to disrupt Black's forces,

example of really understanding typi­

Kramnik switched plans again and

cal positions. The point is that after 16

broke through in the centre with the

b5 c5 White can isolate the black d­

pawn by 17 dxc5. However. it is one 0['

advance e4, soon achie,ving a winning attack.

the more subtle aspects of these struc­ tures that IQP positions are usually not

The next example features a little­

so good for White when his h-pawn

known subtlety in a typical IQP posi­

ha� advanced to b5. This is because he


ha,> weakened his c3- and c4-squares, which Black can frequently exploit for counterplay. For example, a knight can come to c4, supported by the IQP. or Black can put his rooks on the c-fiJe and pressurize White in that way. An­ other point is that White is weak on the dark squares on the queenside, particularly

b4, thanks to the fact that

he ha� no dark-squared bishop and the black a5-pawn controls h4. An incur­ sion by the black queen on

b4 could he

rather awkward for White, especially if the white rook has left the a-file, with the result that an exchange on b4 leaves his a-pawn undefended.

Stanec Graz



For all of these reasons, Kramnik instead switched plans.

16 J:txa5 ..

Kmmnik now revealed the point of his alternative plan:

17 liJd2

Unlike most IQP positions. where White seeks to turn his extra space into a kingside initiative, here he has played mainly on the queenside. How­ ever, such a plan is rarely juslified.

The intention is to seize the c5- and

unless Black has weakened his queen··

b6-squares by playing by thb3. as and

side structure in some way. Here he

lha4. One can see that White has com­

has not, and already White's hopes

pletely reveI"St':d the trend on the queen­

should b e connected with equalizing

side - instead of White being weak on

by a timely d5 advance. However,



Ildi"vsky forestalled this plan with an <"\('dlent sequence: I H...J.c6!

19 .txc6 bxc6!

11 is the last move which is the key uk;!. At first sight, it looks crazy for 1l1;u.:k voluntarily to accept a weak, Isolated pawn on an open file. Tn rcal­ Ily, however, the d)-pawn is much less weak than it appears, and it is the

while pawns on d4 and b2 whjch are Ill,' more vulnerable. Meanwhile, the d'-pawn emmres that White will never 111' able to liquidate his IQP by advanc­ IIIg d5. Such a structural transfomla­ lIOn is a known idea in such IQP

present book, so I give lhe conduding moves without further comment.

positions, although it is not dealt with III most books (Alexander Baburin's lVillfling Pawn StruUures and Drazcn

23 tbg5 l!d5 24 �f3 �d6 25 h3 .!ib8 26 .tr.d2 �b4 27 'ti'xb4 l1xb4 28 l';tc2 f6 29 a6 Wf7 30 :lel h5 31 h4 l1d6 32 g3 !iJf5 33 �g2 'i¥e7 3411ec1 �d7 35 na1l)L xd4 37 �a5 �d5 38 .i:la3 e5 39 �.r31ib6 40 lla3 r3;c7 41 <;t>f3 Udb5 42 �a211b3+ 43 We2 :6b4 0·1

Marovic's Understanding Pawn Play ill Chess being notable exceptions). The first high-level example of which I am aware is Lm:ker-Capablanca, Ha­ vana Weh (10) 1921, although in that !'''Ime, the full didactic value of Black's uk... was lost after a fcw moves, when White exchanged'pieces on d5, per­ milting Black to eliminate the extra p;lwn-island by ...cxd5. In the present J'.ame, the c6-pawn survives for the hlllg term, and Beliavsky gives a text­ Ik)tlk demonstration of the advantage� "r lhe black structure.

20 lhe4 �d7 21 as'?! 'Ibis weakens the a-pawn and drives lhl' knight where, it is going anyway.

2l,.,liJc8 22 'fI'a4 0:\e7 (l» The knight is ideally placed on e7, <ldending the c6-pawn and preparing ioJter jump to f5, attacking the d4 weakness. The remainder of the game IS not really within the scope of the


These two examples both iUustrate subtle aspects of a typical pawn-struc­ ture. If you play openings which lead to these structures (for example, if you play the Queen's Gambit as either White or Black), your success rate will improve greatly if you develop a fa­ miliarity with such typical plans and positional ideas as those shown above. The depth of some grandmasters' understanding of typical middlegamc and endgame positions from their fa­ vourite openings is quite remarkable. In this respect, I am reminded of a story quoted by Alexander Kotov, concerning former world champion Mikhail Botvinnik.



Botvinnik - Matulovic Palma de Mallon·u 1967

This was the adjourned position. Writing in his classic Think Like a Gmndnwster, Kotov told the follow­ ing tale: "DUling the break from play, Botvinnik remarked to SmysJov and me, 'The position is an easy win for White. At the appropriate moment, there is a decisive knight sacrifice on g6 or h5. I analysed similar endings when I was preparing for my match with Tal'''. Sure enough. when play resumed, Botvinnik qnickly wrapped up the full point. thanks to a sacrifice on g6: 43 �f1 ltc8 44 4:Jg3 Jid7 45 �e2 'lWr'a4 46 >tf2 �e8 47 .!bxg61 <LJxd5

Or: 47. ..lilxg6 48 iLlxfS; 47 ...�xg6 484JxhS. 48 4JxhS tfJde7 491M1·g 7+

This is a petfect illustration of the depth of OM preparation. The ending above had arisen from a Benoni, a fa­ vourite opening of Tal's. Clearly, when preparing to face Tal in their world championship matches in 1960 and

196J, Botvinnik's study of the Benoni had gone way beyond simply analys­ ing the opening sequences themselves. Instead, he had analysed characteristic middlegames and endgames arising from the Benoni, and had worked out the typical methods of play in such positions. A small piece of that knowl­ edge was demonstrated in the Matu10vic game. So how docs one develop such knowledge? Well, firstly, not by me.m­ orizing variations - no amount of rote leaming will enable you to find moves like Kramnik's 16 bxa5, or Beliav­ �ky's 19 ...hxc6, let alone Botvinllik's endgame plan against Matulovic. Sim­ ilarly, most opening books will not teach you such tbings. It is true that many opening monographs nowadays include an introductory chapter on po­ sitional themes of the opening, and typical do's and don'ts, but these are of necessity brief and usually only scratch the surface of the subject In­ stead, the main meth(xt of acquiring sucb knowledge consists in studying well-annotated games by players who specialize in the opening in which you are interested. I should like to present an example from some work I once did along these lines. Ever since I stu.1ed to have a con­ sistent and identifiable opening reper­ toire, my defence to 1 e4 bas been the French, and, more specifically, the Winawer Variation: 1e4e 6 2 d4d53lDc3�b4 4 eS cS 5 a3 j,xc3+ 6 bllc3 CiJe7 (D)

In this opening, Black takes some clear positional lisks. He places his



particular, played almost nothing but the French throughout a forty-year ca­ reer. As luck would have it, some ten years ago he published a book of his best French Defence games, under the modest title Ein Lehen lanK Franzii­ si.l"ch - richtif.: gespielt! ("A lifelong French - correctly played!"). An ex­ panded English translation of the book is available, entitled Winning With the French. I naturally used this book as the hasis for my work. and one of the games I studied closely was his fa­ mous victory over Bobby Fischer: -

central pawns on light squares and then gives up his dark-squared bishop, thus leaving himself with severely weakened dark squares. In return, however, he douhles the white pawns and induces White partially 10 block Ihe position with e5, thereby robbing Ihe bishop-pair of much of their effec­ liveness. In lhe ensuing middlegame, Ihe strategic battle usually revolves ;u·ound White's attempts to open the position for his bishops, and, in partic­ ular, to get his dark-squared bishop Into the game effectively, often via Ihe a3-fS diag,!ual. Black. meanwhile, gl'nerally struggles to keep the pos.i­ lion closed, often playing ...e4, and Iryingto pressurize White's weakened Ilawn-structure. Sometime around 1994. after los­ Ilig a couple of na�ty games on the 1,1;lck side of this position. f concluded 111;11 I needed to improve my under­ slanding of these typical Winawer III;ddlegames. In order to do so, I de­ mkd to anaJyse some games played hv Ihe main masters of this opening. Ihllv;nnik and Uhlmann. The latter, in

Fischer - Uhlmann Buenos Aires 1960

7 l{jrJ .ltd7 8 a4 �a5 9"i!Vd2liJbc610 Jl.d3 c4 11 .te2 f6 12 .§La3 liJg6 13 0-00-0-014�d61ilce7151ilh4 !IdeS 16 tUxg6 hxg6 17 exf6 gxf6 18 h3 Iilf5 19�h2 g5 20f4 1ild6 21 �f3 (D)

Here we see a typical Winawer struggle. Black has played ...c4, castled into (relative) safety on the queenside. and has broken up White's centre with



...f6. With his central pawn-mass and excellently placed knight on d6, Black has very good prospects, if he can keep White's bishops from getting ac­ tive. This, however, is looking diffi­ cult, as it appeal>: impossible to stop the dark-squared bishop on h2 becom­ ing tremendously active after either ..gxf4 by Black, or fxg5 by White. .

This is clearly a critical moment in the game, and Uhlmann rose to the occasion, with a move which every French player should rememher:

21...g4! !

White has a protected passed pawn on g5, it is clear that the pawn is going nowhere, Jacking as it does any sup­ POit from the white pieces. White's position is now totally passive, and he can only watch as Black strengthens his grip and prepares the fi.nal assault. 23 g5 !!e7 24 -ig3 <�,e8 25 '8Vd

lbe4 26 �.xe4 dxe4 (D) Not, of course, the awful positional blunder 26 . .fxc4??, which would ne­ gate the whole of Black's previous play by allowing White to free his en­ tombed bishop with 27 t5. .

With this wonderu f l sacritice, Uhlmann underlines the main theme of Black's play in such Wina­ wer po�itions - the fight to tame the while bishops. White cannot take with the bishop, because after 22. ..t.!Je4 his queenside would collapse. He is there­ fore forced to take with the pawn:

22hxg4 f5! (D)

27 wt'2 �eh7 281ltbl �dS 29 tlVe1 UhI 30 \¥kxhl e3+ 31 <Stgl lIxh1+ 32 <;t>xbl c2 33 nbS iJ.xb5 34 axb5 �xb5 35 nel as 36 .lhc2 a4 3711xc6 a.J 38 g6 1Wd7 39l1.e5 b6 40 ..ItM a2 4l 11cl 'fiJg7 42 &tal �xg6 0-1

to the spot. and in so doing condemns the h2-bishop to a life of monastic in­

This is a magnificent example of one of Black's most important strate­ gic goals in the Winawer .. white bishop-pair. Another typical Winawer theme on which I found some tremendously instructive material is

activity behind the cloisters. Although

the positional exchange sacrifice. The

The point. Black nails .


following example b y Botvinnik i s a dassic.


meanwhile, has no active plan. The game continued:

23 h3 '>itd7 24llel �h4 White's only remaining active piece is his queen, so Botvinnik seeks to ex­ change that as well. If White acqui­


esces, Black will have a completely free hand.

25 �e5 '&f6 26 'iWg3 llh4! This stops .tr.e3-f3.

27 lIe3 llf4

Black has a tot"l grip on the posi­ tion, and now plans .. :·l*'h4 to force off the queens. He followed up with ...M and won easily.

Tolush - Botvinnik USSR Ch (Moscow) 1945 Here White has managed to get his dark-squared bishop into Black's posi­ tion, but while be has been doing so, Black has captured the white a-pawn, and has a potentially dangerous pawn phalanx on the queenside. Botvinnik now removed White's most active piece with a thematic exchange sacrii, fice:

The following is another instruc­ tive example of Black's positional ex­ change sacrifice in such positions, this time with some slightly different themes:

21. l!xd61 22 exd6 :;tc6 ••

For his exchange, Black already has one extra pawn, and the d6-pawn is !;,:oing to drop S(Xmer or later. In addi­ tion, the blocked nature of the position means that Ihe whitc rooks have no open tiles along which to penetrate, whereas Black's pieces (especially his knight) have far greater activity. Fi­ nally, Black also has the long··term Illan of advancing either his queenside majority, or playing . f6 and ...cS and .

lIsing his central pawn-mass. White,

Byvshev - Geller USSR Ch (MoscoW) 1952 Once again, we have many of the typical features of the Winawer mid­ dlegame. On this occasion, however, it



is White's light-squared bishop which is the more active, whereas its fellow

the positional themes underlying this

middlegame structure. By performi ng

prelate on c l presents a sorry specta­

such work, I was able to learn a great

cle behind the fixed white pawns.

deal about how to handle these posi

Gellerdealt with the light square pres­ -

sure by offering the exchange:


tions, and my results with the Wina­ wer improved markedly If you wish .

34 .litef7! 35 i.xf5 r1xf5

to improve your play in your favourite

As in Tolush-Botvinnik above, the

opening this i s the sort of work you

main factor justifying Black's sacri­

should try to do. I dentify some leading


fice is the blocked nature of the posi­ tion, which denies the white rooks any

scope. An additional factor tbis time is the weakness of the light squares in the white

position. Once again, there

is little White can do, and his thrash­


players who specialize in that open­

ing, and study their best games, pref­

erably with their own notes. When you come across an interesting or in­ structive moment, make a note of the position and the key moves, particu··

ing about soon led to a lost position:

larly empbasizing the ideas and themes

36 i..a3 ii..f 7 37 UcO �xa4 38 i.e7 �c6 39 g6 !i'Jxc7 40 gxf7+ Wxf7 41 ltel '/;/,6 42 J:tg2 g6 (D)

doing, you will gradually build up

which support the move played. By so

your understanding of the typical


sitions reached from your opening, their characteristic plans and ideas, tactical devices, etc. Such knowledge does not go out of date, nor is its value restricted to only one given position. [f you continue to work in this way on your opening systems, you will soon find that your results will improve significantly and your all-round mid­ ,

dlegame and endgame play will bene­


All Openings are Sound ,

Before leaving the subject of success­

an iron grip on the l ight squares, and

ful opening play. there is one other

queenside. He now prepared ... b5-b4,

plementing Sadler's three principles.

Black has a pawn for the exchange

the now-familiar pawn pbalanx 011 thc and went on to win comfortably.


I wish to make, by way of sup­

This is to say that by and large, it is a mistake to think that one opening is

I bope you can see bow studying

these and other examples brings out

objectively better than another. At first sight, this may seem a radical claim,



but it is one supported by most strong

popularity mainly fQf subjective rea­

players. For example, consider the fol�

sons. Wbat matlers is how comfortable

lowing exchange between interviewer

you feel with an opening, and how

Bachar Kouatly and Garry Kasparov,

well you "know' (in the broadest sense

1991 OM Video production con�

cerning the final Kasparov�Karpov

of know and understand) it. It follows from this that one should


be very careful about �witching open�

on a

Kouatly: "You once said that the

ings, just because a certain variation

Grunfeld Defence is 100% sound"

looks unpleasant or a certain

(emphatic nod from Kasparov). "So


OM has

IO�1 a game with it. All openings

why then, did you also play the King's

are sound, and all openings have tbeir

Indian Defence in tbis match?"

critical variations. If you give up an


"J think all openings are

opening otvery time you encounter a

1 00% sound - all normal openings, lilat is! It is just a question of your

problem with it, you will never have a

mood and your preparation."

playing a reputable main�line open�

decent opening repertoire. If you are

The queslioll put by Kouatly sounds

ing, it is essential to understand that

pretty nai"ve, of course, but the answer

there cannot be anything much objec­

it received is actually very valuable.

tively wrong with it. Consequently,

Contrary to what many players be�

when you encounter a problem in a

lieve, there is no objective basis for

given variation, you need to study the

Ihinking that the Sicilian is better than

line concerned and find an improve­

Ihe French, or the Nimzo�lndian better

ment. Switching instead to a new

Ihan the Slav, etc. Tn reality, all 'nor­

opening will simply bring new pmh­

rnal' main-line openings are perfectly

lems, as well

sound, and there is no good reason for

chance to utilize your experience and


depriving you of the

preferring one opening over another.

understanding of your previous open­

( lpening theory is largdy a malter of

ing. This is a subject 10 which we will

iashion, and lines go in and out of

return in more detail in Chapter 8.

2 Variety - the Spice of Life?

In approaching the question of open­ ings. a player has essentially two main options - either to stick resolutely to a narrow sdection of openings, Of play a wide variety of systems. Both these approaches have their pros and cons.

Playing Against the Pieces Over the years, the majority of lll<lster players have tended to have a rela­ tively nrumw opening repertoire. They choose a certain .set of lines for White, and as Black they have one main de­ fence against I c4, and one against 1 d4. With only a few, relatively rare, c)(ccplions, they stick to these lines through thick and thin. This approach has the key advantage that the player has plenty of opportunity to build up his knowledge and understanding of the lines he plays. There is nothing like practical over-the-ooard experi­ ence to develop one's understanding of an opening, or a certain type of middlegame or endgame structure. As we saw in the previous chapter, having a deep understanding of the typical middlegame and endgame JX)sition� to which an opening leads is one of the most important aspects of successful opening play. By devoting all of one's efforts to a narrow range of openings,

one has the chance to gain the neces­ sary experlence. We have already cited above the ex­ ample of Wolfgang Uhlmann, who showed an almost religious devotion to the French Defence throughout his entire career. Another leading non-So­ viet GM of the l YSOs and 196()s was Svetozar GligoriC_ Although not quite as monomaniacal1y focussed as Uhl­ mann, Gligoric too was a player who stuck to a narrow opening repert.oire. Against 1 e4, he usually defended a Closed Lopez, although he occasion­ ally switched to the Sicilian, particu­ larly in his younger years. Against 1 d4, he remained faithful to the King's Indian for his entire career, moulding the opening into a formidable weapon Ihat brought him numerous points. As White, he was similarly well-focused, opening I d4 in the great majority of his games, and having a series of fa­ vourite opening line." from which he rarely varied. As an example of the way Gli­ goriC's approach can hring great divi­ denels, l sh{)uld like to look brictly at his handling of one particular line, namely the Exchange Variation of the GrUnfe-.ld, GligoriC was a firm be­ liever in this line as the best reply for White. In the Russian colleclion of his best games, 19rayu protiv Figur,



he wrote on this subject: "My fond­ ness for the Exchange Variation has lasted for more than 11 quarter of a century, Maybe fondness is not even theright word - rather, it seems to me a matter of pri nciple, thllt if B lack of­ fers his opponen t the chance to seize the centre, White is almost obliged to lake up the chaJ lenge, notwithstand­ ing Black's rapid development and queenside counterplay."The following game is a character­ I stic example of GligoriCs expertise

with his favourite line.

Tn his notes in the above-mentioned book, GligoriC points (luI an amusing

Gligoric - Tukmakov USSR - Yugoslavia, Odessa


I d4 4:Jf6 2 e4 g6 3 €lC3 dS 4 exdS fi \xd5 5 e4 4:Jxc3 6 bxc3 J..g7 7 .tc4

Although cuO'ent theol)' concen­ Irates on the lines with 7 tLlf3 as White's most dangerous try, the text­ nu ,ve was considered virtually obliga­ lory until the 1980s, nnd was the line ;dways preferred by Gligoric. 7 cS •••

( ll\c of GligoriC's cllrliest experi­ ,'!In'S with the Exchange Variation saw

.1 I ypically vigorous demolition of the 1,11l' with ... b6: 7. .Jl-fl H !De2 b6 9 0-0 �- 1"7 1 0 f3 c5 1 1 �.c3 cxd4 1 2 cxd4 . Il'h 13 l:tcl e6?! 14 d5! exdS I S �xd5 " 1;1') 1(1 tbf4 11c8 17 llxc8 l!Hxc8 1 8 ';' d· 1 0,xd4+ 1 9 WHxd4 1Dc6 20 �b2 " 1<" 1 2 1 ltcl 'iWb8 22 lhe6 �xd5? , " l'x.c6 23 i.xe6+ 1(n 24 '/Wf6 \\W'fS " , TV? +--) 23 lJ!'g7# ( 1-0) Gligoric­

I'." I,m;!n, Buenos Aires 1 955.

X '�' \d 4\c6 9 �e3 0-0 100-0 (D) 1tI 0'\a5 .••

paradox. In the 1 973 Leningrad Inter­ zonal tournament, he had lost two games on the white side of the Ex­ change Griinfeld, against Smejkal and Tukmakov. Subsequently, Smejkal had

repeated the Grtinfe1d against him, this time trying out the line that Tuk­ makov had used at Leningrad. Armed with an improvement for White, 01i­ goric had won in the following crush­ ing style: 1 O .. .'llWc7 1 1 lIc l Itd8 1 2

�d2 �a5 1 3 nfd l b6 1 4 �h6 Sta6 1 5 J..xa6 �.xh6?! 1 6 '&xh6 '&xa6 1 7 �f4 cxd4? 1 8 li.d3 ! �h5 1 9 bth3 '/We5 20 �xh7+ �f8 2 1 tbe6+! fxc6 22 �xg6 �f6 23 11f3 'i'xf3 24 gxf3 d3 25 �h l 1-0 Glignric-Smejkal, Milan 1975 . Now, in the present game, it is Tuk­ makov's turn to repeat the Grtinfe1d, and he Ch(XlSeS the line with which Smejkal had beaten GHgorlC at Lenin­ grad 1973! This little tale illustrates something very typical of Gligoric's approach to openings - a refusal to be cowed into changing his line ju�t be­ cause of a defeat.



1 1 id3 cxd4 12 cxd4 b6 13 .tt.c1 The first sign that Gligoric has an improvement ready. In the aforemen­ tioned game against Smejkal, where Black had prefelTed the immediale 1 1 ...b6, he had adopted the plan of 'tid2 and .tho. Here he haf: a very dif­ ferent idea in mind.

13 e6 (D) ...

1 3. .. i.b7 would be met by 1 4 dS.

GligOlic used the same strategic idea to beat Vaganian in different variation of the Exchange Grunfeld: 7...0-0 8 !be2 �d7 90-0b6 1 0 'lWd3 i.b7 1 1 e5! 'tk6 1 2 'tJf4 e6 1 3 'iVh3 llJaS 1 4 £e2 c5 1 5 i.e3 cxd4 16 cxd4 l1fd8 1 7 .:tad1 Vie7 1 8 't!Ig3 11ac8 19 h4! witb initiative, Gligoric-Vaganian, USSR­ Yugoslavia, Odessa 1975 ( 1 -0, 56). The present game continued:

14 ..ii.b7 15 Lt)f4! W"e7 1.6 'i¥g4 l,lJc6 17 h4 Lt)b4 IS i(.c4 b5 19 i.b3 h5 20 i\lg3 .

20 �h3! is stronger, planning later g4.



20...ltfc8 21 ii.d2! llxc1?! 22 ttxc1 (D)



This excellent move is the key to GligoriC's new plan. Although it looks odd to concede the d5-square and give up the chance of creating a passed pawn with d5, the move's merits out­ weigh the�e factor�. Black's Grtinfeld bishop on g7, usually the pride of his position, lose>: it� effectiveness at a stroke. Meanwhile, the diagonal of White's d3-hishop is opened towards the black king, and White will follow lip with lbf4 and "lUfg4, with the mak­ ings of a dangerous kingside attack. Meanwhile, (he black knight on a5 i � a long way from the key d5-square. I n another game played at the same event,

At this point, Black blundered:

22 hlcS?? •••

However, even after the superior 22 .. .lbd5 23 ft'ld3! While would have had the advantage.

23 Uxc8+ AxeS 24 'iVc3 1-0 White is winning a piece. Gligoric had clearly won the theoretical debate, and avenged his previous defeat in the variation.





The lesson to be drawn from these

past. In this respect, it is perhaps no

",ames is that the player who sticks to

coincidence that Karpov is notorious

particular opening is able to de­

amongst elite GMs for being a distinct


vdop a much deeper understanding of


his syHtems than somebody who plays

Nevertheless, there are a few very

lois of differcnt lines. However, in or­

notable examples of players who have

lief to apply the OligoriC approach,

an almost mystic belief in the correct­

you need to be the sort of player who

ness of their pet lines, and hardly ever

has great self-confidence and belief in

vary from them. The most obvious

his own judgement. In effect you are

ca<;e which comes to mind is that of

saying "I believe this line is the abso­

Russian OM Evgeny Sveshnikov. His

lute best way to play against this open­

repertoire is well-known - 4...e5 in the

ing, and I don't care who knows I play


2 c3 against the Sicilian, etc.

it, or what prepared li.nes they may

He has written extcmively on both of

have waiting for me, I amjustgoing to

these lines, and even advances a the­

play my line and Jet them do their

ory which purports to prove on princi­

worst" Gligoric's helief in the correct­

ple that 2 c3 is the best move against

ncss of thc Exchange Griinfeld is clear

the Sicilian!

from the quote given above, and it runs

Another, less well-known GM who

through his entire approach to open­

adopts the GligoriC-style approach is

ings. Indeed, the title of best game col­

the young Russian, Alexei Alexan­

lection quoted above, /grayu protiv

drov. My attention was first drawn to

Figur, actually means "I play agmnst

him in the mid-1 990s, because one of

the pieces", or to put it more colloqui­

the lines he plays was my choice


against the Meran Variation.

"1 play the board. not the man" .

In contemporary chess, there are !'ewer leading players who follow the Gligoric approach, Karpov is one who still does so, sticking very predictably

10 a narrow set of openings


I d4 with

1 d4 d5 2 c4 c6 3 4lf3 ltJf6 4 lbc3 e6 5 e3 ltJbd7 6 �d3 dxc4 7 .i.xc4 bS The usual move is now 8 .td3. However, I became interested in:

8 �e2 (D)

White, Caro-Kann and Nirnzo/Queen's

This was used successfully by Mat­

Indian with Black. The main reason

thew Sadler in the early 1990s, and was

why he has few followers amongst to·

also a favourite of my Russian trainer.

Jay's elite is that the increasing use of

Aftcr a brief spell of popularity, 8

I;omputer databases makes it so much

Jle2 fell out of fashion following the

casier to locate the weak spots ill a

game Seirawan-Kramnik, Manila OL

player's repertoire, and this has tended

1992. After 8 ...a6 9 e4 b4 10 e5 bxc3

10 encourage a policy of greater diver­

I I exf6, Kramnik avoided the crazy­

sity_ In addition, databases also make

looking 1 i...cxb2 12 fxg7 bxal'i!N 1 3

the task of mastering a new opening a

gxh8\1W which favours White because

lot quicker and easier than it wa� in the

of his safer king position, and instead




preferred the simple 1 l ._.4Jxf6 (al­ though hailed as a novelty hy many commentators, the move had in fact heen played before, including one game as far hack a� 1949, hut it was Kramnik's use of it which led to its popularity) 1 2 bxc3 .i..d6 (D).

one of those who stuck to his favourite line. After a number of attempts to prove something for White after 1 3 �a4, his perseverance finally paid off. Facing Bareev in the FIDE knockout in 2000 at New Delhi, Alexandrov reached the point of no return. After losing the first game of the mini­ match, he was forced to win with White. Bareev is a Semi--Slav special­ ist, and so it was not hard to predict that the position after 12....Jtd6 would arise on the board. At this point, how­ ever, Alexandrov produced the strong new move 1 3 c4!, which immediately puts Black in some trouble. The dif­ ference from Seirawun--Kramnik ahove is that now 1 3 .. _c5 can he met by 14 dxc5 �a5+ (or 14...i.xc5 1 5 �xd8+ �xd8 16 .'De5 with a clear advantage) 1 5 .id2 'liYxc5 1 6 0-0 0-0 1 7 .i.e3 Wic7 1 8 c5! with advantage. The attempt to exploit White's uncastled king by [] . ,�J)'j-+ fails to 14 j.d2 ,i',xd2+ 1 5 �xd2 fbe4 1 6 '&c3 �a5+ [7 '.ttf l!, when White is again better_ Bareev chose 1 3 .. -<.)-0 but after 14 c5 .�c7 15 0--0 White wa<; somewhat better, thanks to his space advantage and the bad bishop on c8. Tragically, Alexandrov later hl.ew his advantage and lost, but he had proved his point theoretically, and at the time of writing, no top player has repeated the line beginning K ..a6 since_ Bareev himself wrote that the next time he needed a draw against 8 .i.c2, he would choose 8...04. However, even here Alexandrov has repeatedly shown that \\'hite can try to squeeze some­ thing out of the position, thanks to ..

Now Seirawan tried 1 3 0--0 0--0 14 (;4?! c5 and soon slOod worse. Subse­ quent glllnes concentrated on J 3 lbd2 or 1 3 0-00-0 1 4 �g5, but White failed to achieve anything. As a result, mosl top OMs stopped playing 8 j,c2, but Alexandrov was


Black's potentially weak b4-pawn. In following game, I witnessed him do just this against WGM Nino Khurt­ �jdze: II",

Khurtsidze Bad WiJrishoien 2001



S M 9 tZ\a4 :�.b7 9...1Le7 10 0-00-0 I I a3 (Of 1 1 �c2 ':"'h7 12 a3 bxa3'!! 1 3 b3!} 1 L.bxa3?! 1 2 b3! �b7 1 3 �c2 and now two ex­ amples illustrate the point made in the Hcxt note: a) 1 3 .. :&a5 1 4 0.d2 ttfcS 15 �f3 �.a6 16 lLic4 �xc4 1 7 bxe4 ktah8 1 8 l·S. Black has failed to achieve the ...c5 break and still has a weak pawn on e6, Aleksandrov-Svirin, Smolensk 1997. b) I3...c5 14 dxeS h4 15 �e3 fie7 1 6 .ixa3 lbg4 1 7 g3 lbdf6 18 i-b2 .l!I..b7 19 b4 and White was simply a strong pawn up in Alexandrov-J1les­ ��as, Hatumi Echl 1999. ...


The natural 1 1 ...bxa3?! is an error because of 1 2 b3!, when Black has trouble achieving the ... c5 break with­ out allowing White a favourable ex­ change of dark.-squared bishops. This idea was demonstrated in the two Alcxandrov games in the previous note. 12 'i'c2 0-0 13 lIdl c5 14 Ctlxc5 €\xe5 15 dxc5 �e7 16 axb4 axb4 17 _8

TIlis is the first deviation from two earlier Alexandrov games, in which he had prefened 1 7 .'td2, and each time went on 10 win in very similar fashion to the present game: 17 .. .'ihc5 1 8 �xc5 i.xe5 1 9 tDe5 fie7 20 f3 (D).

100·0 �e7 (D)

ll a3 aS

20...!hal (20...fifc821 :'xaS IhaS 22 Wf2 :a2 23 QJd3 h6 24 .ubI 'Dds 25 e4 4:lc3 26 ,ixc3 bxc3 27 bxc3 i.a6 28 ttb8+ "ph? 29 lta8 f5 30 e5 j.c4 3 1 :Cxa2 <itxa2 32 'It>e3 g5 33 'it'd4'it'g6 34 tbcs f4 35 pid3+ �g7 36 c4 1 -0 Alexandrov-Kuporosov, Mos­ cow 19(5) 2 1 :axal .nd8 22liJd3 b3 23 :ta?lL:I.7 24 'lti>f2ibdS 25 lh5 1Dc7 26 'io'el .l.1dS 27 e4 Si.a6 2s lbcs ..Ilxe2 29



<;&>xe2 .liLf6 30 �3 i.d4 3 1 �a3 lIhS 32 i.f4 e5 33 .ixe5 .axeS 34 qJxeS tZJb5 35 l1a5 f6 36 lOd7 1-0 Alexan­ drov-S.Pedersen, Minsk 1994. 17..�xa8 18 lbel 'ii'xc5 19 'flYxc5 i.xc5 (D)


At first sight it is hard to believe ,

that Black should have much trouble

holding this, but in fact the b4-pawn drops off in surprisingly short order.

20 lbd3 �f8 21 .:td2 b3 22 f3 .!lJdS 23 'Del The pawn is already indefensihle, and Alexandrov went on to win, de­ spite determined resistance. The most striking thing anout the above examples after 8.. b4 is the dose similarity between the way in which Alexandrov won all three games. His strategy was the same in each case, namely to exchan ge into an ending where Black's advanced b4 paWtl was a weakness Even though Black should be able to hold an ending of the type reached above, it is clearly not easy for him, and to defend such a position against a player who has extensive .



experience in just such endings is even more difficult. None of Alex3ndrov's opponents in the above games can be classed as weak, yet none was able to hold the ending a gainst him. I am not for one moment suggest­ ing th at the 8 .te2 line is any kind of refutation of the Meran. nor even that it is a sure route to a white advantage. What I am saying, however. is that these games are another very good il­ lustration of the plus side of stickin g faithfully to one's chosen lines, and continually polishing and refining them in the light of one's experience. Rather than continuaJly changing one's system, in pursuit of the very latest fashion amongst elite GMs. it can in practice be much more effective to de­ vote one's time to enhancin g one's knowledge of a particular variation, even if it does not lead to a 'theoreti­ cal' advantage. Whatever the objective merits of the line concerned, most op­ ponents are going to find it difficult to defend against a player who has played the position numerous times before and built up a substantial levc1 of ex­ perience with it.

Keeping on the Move lbe foregoing examples have demon­ strated the advantages of the narrow repertoire approach, nut there are sev­ eral drawbacks as well. Firstly. the danger of walking into opponents' preparation is much greater, and, as pointed out above. the widespread use of computer databases has increased this risk. However. alth ough this is a '



legitimate concern at master level, it

which means that one can develop a

seems to me that for the average club

very good understanding of that par­

;md league playee, such a danger is

ticular structure, without that knowl­

relatively remote and should not be over-estimated. In local leagues or

edge being so useful in any other type

weekend tournaments, one rarely has

cialize in an opening which typically

of pusiliun . By contrast, if you spe­

much advance warning of who one i�

produces an IQP structure for­

I() play, and even if one did know, very

pie, the lessons you learn in that open­

few of one's opponents will have

ing will be of relevance to a host of


)!ame� on a database. As a result, there

other openings, many of them quite

is much less chance of running into

different, but which happen also to

�pecific preparation at the average

lead to IQP structures.

amateur level, whereas the advantages

Another example of an opening

of playing an opening that one has

which produce s rather unique middle­

played many times before and under­

game and endgame p ositions is the Si­

stands well are likely to be of rather

cilian Dragon. This line has been very

greater significance

popular at junior levels in the UK for


The second drawback of the narrow

several decades, with Chris Ward an

repertoire is rather more subtle, and

especially influential advocate. How­

relates more to one's all-round devel­

ever, the types of rniddlegame which

opment as a player. It is the danger that

arise from this opening, particularly

player who sticks exclusively to cer­

the Yugoslav Attack lines, do not re­

tain opening lines will become too

semble very closely those of any other


limited in hi� play. It is noteworthy

opening. This has the great practical

that when outlining the main princi­

advantage that it is relatively easy to

ples on which his chess school was

gain a good understanding of such a

hased, Mark Dvoretsky mentioned "a

narrow set of typical positions, but it

rejection of the concentration of one's

can also limit the development of a

dforts on only one opening which is,

player s all-round positional ability. In

unfortunately, typical of our days", It

this respect, I remember an intere st ing

seems to me that this risk is particu..

comment by Scottish no.

lady great for young players, whose

Rowson, to the effect. that he has a ten­

style and all-round chess education is as yet relatively undeveloped. It i f> also

dency to over-estimate the strength of

a particularly serious problem when

probably a consequence of playing too

the opening concerned leads to posi­

many Dragon s in his youth.


I, Jonathan

exchange sacrifices, and that this is

lions of a very specific lype. The

There is no doubt at all that at

French Winawer examined i n the pre­

grandmaster level the pendulum has

vious chapter is one example. The

swung markedly in recent years to­

types of positions reached do not re­

wards the policy of having a wide rep­

semble thOf>e o f any other opening,

ertoire. If one goes back to the 1970s



and earlier, there were hardly any

of most players at that level is a lack of

leading players who adopted a wide

understanding of the typical middle­

variety of different openings. Proba­

games and endgames reached from

bly the fin;t w()rld-da�� player to do so

the opening. This knowledge is best

was Jan Timman, who has

improved by concentrating on a rela­



played a colossal range of openings

tively small number of openings, and

of computer datahases became com­

knowledge and experience of them.


both as White and Black. Until the use


working to

improve one's

monplaco: over the past decade or so,

2) Having said that, such an ap­

he WiLS about the only example. but

proach does req uire a certain strength

now mosl of the top players have a

of character and belief in one's own

fairly flexible repertoire In Chapter 6,

judgement. [f you are the sort of

we will consider in more detail the im­

player who is easily swayed by other


pact of computers on opening prepara­

people s opinions, or by the outcome

tion. For the present, I would just like

of a specitic game, then you are likely

to emphasize that the main reason for


to find it ditlicult to stick with one

the move towards variety at the top

opening although you should gener­

GM level is the fear of opponents'

ally try to do so.

preparation, which is made much eas­ ier by the use of databases_ For the reasons discussed above, this consid­


3) For j uniors and others with a se­

rious ambition to improve their


significantly, it is probably better to

eration is much less important at club

adopt a hroader repertoire, in order to

and weekend level.

improve their all round positional un­

By way of conclusion, I would say


derstanding By concentrating on only .

that each player must make up his or

one or Iwo openings they can often

her own mind about which approach

achieve better results and more rapid

to prefer, but


would emphasize the

following point�:


apparent improvement in the short term, but this often comes at the ex­

I ) I am firmly convinced that, at

pense of an excessively narrow chess

the club and weekend level, a player

outlook. This in turn can become a

will usually get better results by stick­

barrier later on. when they find that the

ing to a narrow mnge of openi ngs.

lack of an all-round chess culture op­

This is hecause the biggest weakness

erates to check further improvement.

3 Stylistics

Another factor

weakness was t"cti cnl ability. One of

which looms large in I!lOst people' s perception of how one should choose openings is playing

the things which soon became appar­

style. It is generally accepted that one

an experienced trainer was that this

choose openings which in wi th one's style, and I would cer­ ,ainly not argue with this. However, it d'll�s seem to me that a certain amount .. I nmfusi on and misconception SUf­ I< Hmds this topic, and J therefore pro­ I "lSI.' to go into it in some detail.

picture of my play was seriously wide of the mark. In fact, my best results in

should try to IiI

Who am I? 'I 'Ill' first thing to say is that, in order to . huose openi ngs to fil your style, you

I!lTd [0 know what your style is. This " , 1101 as easy to achieve as you might 111I1lk. If you have never had your

a trainer or strong play.T, you may well be harbouring ,!lHll' notable jlJusiol1s or m isconcep­ (''''is ahout your chess style. It is cx-

V;IIIll"S studied by

1 1 " lIIl:1y difficult to be objective about "'H"'S

own play, and most players' per­

their own style is much it to be, III_lu lu what i t actually is. t>.1 y ('wn case is typical. Before my 11(1 \1;111 experience, I had always be­ ion",1 myself to be predominantly a 1 " " . l l lon:!1 player, whose strengths " ' - I , · Illy positional understandin g and ,-ud),alllL' technique, and whose main

" -I,llon of

, I"\n 10 what they would like


ent when my games were dissected by

individual games were almost always

the resul t of tactical opportunism, usu­

outplayed po­ sitionally. Although I had won many ' smooth' pos ition al games, these were generally in positions where I h ap­ pened to have seen the appropriate plan in another game and was up against an opponent who was not strong enough to prevent me from re­ peating it. Whenever J faced an unfa­ miliar positional problem, against a strong player, I generally showed a complete lack of any idea how to play the position. And as for my endgame technique . The reasons for this are not hard to see. I first became seriously i nterested in chess around 1973-4. This was (he period when Anatoly Karpov was tak­ ing the ehess world by storm. Natu� rally enough, Karpov became my first chess hero, and has remained a favour­ ite to this day. Subconsciousl y, he was the player I wanted to be like, and I grew lip believing that my playing style was similar to his. Another early hero was Ulf Andersson, whose legendary ally after having been

. _



ability to squeeze out wins from ap­ parently equal-looking endings had a significant influence on my play as a junior. I went in for a lot of eady queen exchanges in those days, and it is true that I didwin a lot ofendings. In fact, at the age of 15, I probably did have pretty good endgame teclmique by the standards of 2000-ratOO, Eng­ lish IS-year-olds, that is. The problem is that while other aspects of my play improved subsequently, the endgame technique was neglected. By the time I was 30-somelhing and an FM, the technical side of my game had fallen well behind certain other aspects. Thc conclusion to be drawn from this is that one should be very ean:ful when considering one's own style. Firstly, you should try to analyse your games properly, especially the losses. These will frequently tell you far more about your strengths and weaknesses than your wins. Secondly, ifat all pos­ sible, get an experienced trainer. or at least a !\trong player, to have a look at a representative selection of your games (not just the good ones!) and say what he thinks your strengths and weak­ nesses are. Indeed, if you are able to, get a second, third and fourth opinion. The more different, objective views you get, the more likely you are to ar­ rive at a coherent and reasonably accu­ rate conclusion. �

No Style without Strength Another point to bear in mind i s that playiug style is generally of more sig­ nificance, the stronger a player is. At

the level of the typical club and week­ cnd player, differences in style are not usually as great as most people imag­ ine. Even at top grandmaster level, a surprisingly large proportion ofgames are decided by tactical oversights, deep in the middlegame and endgame. In­ evitably, therefore, this will be true to an even greater extt:nt, the weaker the pl<lyers are. As a result, players at an average amateur level should not take too mw.:h notice of small differences in playing style. It i� far more impor­ tant to pick. openings which you like and feel comfortable with, than to WOfl"Y about whether a particular line really 'suits' your style.

The Opening - a Servant, not a Master Once you have decided what sort of style you think you have. you can start to think about spccit"ic opening choices. Just as most writers on the game like to divide players into 'strat­ egists' and 'tat:ticians', so they do thc same with openings. Typically, you will see something such as the follow­ ing:

OpeninK.\" for posiliofWl players: White: 1 d4. t c4, 1 fi:lf3 Black against 1 e4: French, Caro­ Kann, Petroff Black a�nst 1 d4: QGD, QGA. Nimzo-Indian, Queen's Indian

Opening.rfor tactical players: White: L e4


Rlack against ] e4: Sicilian



h3 i.c5 31 tZlli2 i.1'2 32 c4 dxc4 33 bxc4 h5 0-1

Modem, Alekhine

Rlack against ] d4:


King's Indian,

Alexander - Burger

(JriinfeJd, Benoni, Benko, etc. In reality, just as it is misleading III describe Capablanca as a 'posi­ liona]' player and Alekhinc as a 'tacti­ cal' player, so the above classification of openings can be equally mislead­ ing. Consider the following couple of

Margate 1937

1 e4e5 2 f4 d5 3 cxd5 e4 4d3 tDf6 5 dxe4 ti:ixe4 6 ltJrJ �cS 7 �e2 .il.fS 8 4:Jc3lWe7 9 iLe3 ti:ixc3 10 i.xcS o.7\xe2 11 it_xe7 ttJxf4 12 .ta3 ftJd7 13 J..bS 0-0-0 14 0-0 (D)


Em. Lasker - Pillsbury St Petershurg ]895/6


I e4 e5 2 lbf3 ttJf6 3 tDxe5 d6 4 �3 !i\xe4 5 d4 d5 6 .i..d3 ii..e7 7 0-OQk68 ttcl .i.g4 9 c3 f5 10 'i'b3 0-0 11 iU'4 lio.xf3 12 gxf3 lt\g5 13 q;,g2 'tiJId7 14 "tWc2 tiJe6 15 J.el j,d6 ]6 .'bd2 1itae8 17 /ilfl (D)

14...ttJxdS 15 ti:ig5 .tg6 16 lLlxf7 i.xf7 17 :txf7 tthg8 18 11el h6 19 c4 lbsf6 20 .t1l4 c6 21 .tc2 'it'b8 22 nee7 J:tge8 23 .i.fS .u.xe7 24 .\lixe7 ttJe5 25 lil.xg7 litd1+ 26 Wr2lid2+ 27 <Jo>e1 lfJxc4 28 Jixf6 1-0 In the first, Pillsbury chooses that well-known

'drawing weapon', the

Petrotf Defence, and proceeds to tear the then reigning world champion's

17 .!iJexd4! 18 'WId] J:l:xel 19 �xe1 \,D! 20 'lttxf3 f4 21 "I'd! Q.JeS+ 22 ·i.-do �g4+ 23 'iitd2 'iYxdl+ 24 <;!;>xd1 mL\ 25 'it>e2 4leS 26 f3 lle8 27 b3 � \�..+ 28 Wd2 lLIe3 29 .i.b2lbg2 30 ...

head off_ In the second example, that


ferocious attacking weapon the King's


endgame, with White having two bish­

Gambit results inside


moves in an

ops and a powerlul initiative. The



point is that most openings are not as one-dimensional as people would have us believe. Around the tum of the cen­ tury, the Petroff was widely employed by players such as Pillsbury and Mar­ shall, hoth of whom are regarded as vigorous attacking players. Althougb nowadays the line i� used at super­ tournament level by players looking merely to make a draw as Black, there is no objective reason why it should not be played by more ambitious players, especially at Jess exalted levels. There is really only one variation which is of a drawish character, that being the line where White plays 5 1We2. Even here, the disappearance of the queens does not automatically mean that hands must be shaken, as is amply demon­ strated by the fact that no less a player than Timman once lost the white side of this position in a .�erious tourna­ mentgame:

Karpov Tilbtlrg J 980



1 e4 eS 2 4.Jf3 iDf6 3 4.JxeS d6 4 4.Jf3 lbxe4 5 �el �e7 6 d3 euf6 7 �g5 'tWxe2+ 8 i.xe2 i.e7 9 <'be3 c6 10 0-0-0 4Ja6 1 1 ttJd4 4Jc7 12 llde1 lbe6 13 lDxe6 ..Itxe6 It is clear that Black has no prob­ lems, but that is no reason to agree a draw. Of course, the position should be drawn with correct play, but then again, the same could be said about the initial position of the game. In this case, neither player was inclined to of­ fer a draw, and play continued. 14 f4 d5 15 .>I.f3 \1;d8 (D)

Preparing to bring his king's rook into play. Of course not 15 0-0-0??, which would lose immediately to 1 6 f5. ...

16 <1:Jel ne8 17 'bg3 g6 18 !be2 JLd7 19 c4 dxc4 20 dxc4 �g8 21 lbc3?! Thus far, \Vhite's manoeuvres have brought him no advantage, but from here on, he begins to get the worse of it. This move is the start of hi.'; trou­ bles, because he allows his quecn's bishop to be shut out of play. Correct was first to exchange on e7, and only then to bring the knight to c3.

21 f6 22 .th4 i.d6 23 Uxe8+ <&->xe8 24 ttJe2 .i.e6! ••.

This nice move provokes a signiti­ cant weakening of the while king posi­ tion.

2S b3 ..af5! Now ... ..ta3+ is a lhreal, and the white king fatls under the crossfire of the two black bishops. 26 \1;b2 !!d8 27 M2? (D) A tactical error in what has now be­ come a rather tricky position. Correc1


was 27 .tel, but Black is still rather hetter.


he can follow up with such moves as g3,Jtg2,q]c3,e3, �ge2,O-O, b3, jtb2, etc. On the other hand, the player who is looking for more immediate hand­ to-hand lighting can develop more ag­ gressively with tZ:lc3, tDt3, and an early


d4 or e4, for example. The same is true of almost any main-line opening. In the Caro-Kann, for example, after I e4 c6 2 d4 dS 3 �)c3 dxe4 4 tDxe4, the player who is looking for a more adventurous game than that offered by either 4....ifS or

.. 4. .QJd7 can play the line 4. bf6 5 .


liJxf6+ gxf6 (D).

27 i.xf4! 28 g4 �.eS+ 29 <Bel :i.:.c6 30iLxa7?! :eta831 �d4 t!xa232 :t1c1 rtJf7 33 i!.xc5 fxeS 34 h3 �f6 3S '''k3 J:th2 36 gS liJd7 0-1 •.•

He is losing another pawn, with no nllnpensation. Although this game is hardly likely to make its way into any hllure edition ofTimman's best games, Ii Ie fact that a player who was by then

hnuly among the world elite can lose til is position so easily is a clear illus11:llion that an equal position is by no IIIl'ans the same thing as a drawn one. The key point about all this is that

This variation had a spell of popu­

IIlost major openings are sufficiently

larity in the 1970s and 1980s, but has

I"'�ihle to be played in different styles,

never been fully tlUsted and then went

,Iqx:nding on the taste and ability of

almost totally out of fashion. How­

thl" player concerned. The English

ever, in previous years it was em­


tpcning (1

c4), for ex:ample, is auto­

ployed very successfully by creative

III:,\'i..:ally classified in the 'positional'

players �uch as Larsen and Bronstein.

1,;iSkct, yet it has been used regularly by

Ine Russian master and trainer Kon­

1I1:I1lY very vigorous attacking players,

stantinopobky played the line formany

IIIduding Tal and Kasparov. If White

years, in both over-the-hoard play and

\\';l11ls a quieter positional game with

correspondence chess, with enonnous

I.'ss dependence on COIK'l'ete variations,

success. Consider, for example, the



following crush against one of the best Soviet players of the first half of the 20th century:

Levenfish Konstantinopolsky LeninXmd 1947 -

sharp position typical of the variation had arisen, and Black eventually won after considerable complications.

8..,lbd7 9 \lWa4 e6 10 .id2 !'tgS 11 0-0-0 iLf51 12 .l1de1 Itg41 (D)

. illfJ The position after 6 i.c4 Itg8 7

Qif3 arose by transposition in Sok­


olsky-Konstantinopolsky, USSR eh (Moscow) 1950, but after 7 . jH5 8 .tf4 e6 9 0-0 Ji,d6 10 .t xd6 '&xd6 I I ..

lLlh4 Ji.g6 1 2 t4 f5 1 3 c3 4.Jd7 14 a4 0-0-0 1 5 as <i:lf6 16 a6 h6 17 �b3 i.h5! 1 8 g3 c5! the white kingsidewas falling aparL

6..�g4 7 .\te2 'illYc7 (D) The strut of a deeply-calculated and original plan of attack, involving pIaeing the rook in a dangerous position in the centre. I3 h3 1Ie4 14 \\Wb3 0-0-0 15 i.e3


cS! 16 ttJd2

If instead 1 6 j_d3, there follows

J 6 Jlxe3 1 7 I1xe3 Axd3 1 3 lhd3 c4. After the text-move, Hlack exe· ..

cutes a long-planned exchange sacri· fice to destroy the white king position.

16...cxd4! 17 lLlxe4 .lfLxe4 18 il.d2 8 c3 Levenfish plans to castle queen­ side. hut his king proves surprisingly vulnerable on that side of the board. In Abroshin-Konstantinopolsky. USSR corr. eh 1952-5, White prefelTed to castle kingside. After 8 i.e3 l,1j,j7 9 c4 e6 J 0 �a4 �g8 I I h3 i..h5 1 2 0-0 Jid6 1 3 b4 i(.f4 14 '§'b3 0-0-0 15 a4 f5, a

After J 8 J.xd4 tbcs 19 i..xeS 'WxcS there is no defence to a lethal check on the h6-c 1 diagonal.

18, tbcS 19 �b4 lbd3+ 20 .txd3 1l.xb4 21 Itxe4 dxc3 22 .:bc3 llxd3 23 :xb4 l1xc3+ 0-1 •.

Konstantinopolsky himself demon­ strated, in prototype fonn at least, the best response to Black's line. which is



10f White to fianchetto his king's

reasonable prospects for Black. White

I,ishop, This has the effect of blunting

can claim a small advantage, with his

IHack's cOlmterplay along the g-file,

bishop-pair and slightly sounder pawn

while the bishop can exert useful pres­

formation, but probably no more so

sure down the long diagonal (in con­

than in mo�t openings.

jUllction with a pawn advance b4-b5, lor example), in the event of Black

Even an opening like the Sicilian of­

<',lstling queenside. It is this plan which is largely responsible for the decline ill

fers great �cope for a variety of ap­ proaches. If you want real blood and

popularity of 5... gxf6.

thunder, you can choose a line such as the


but those who prefer a

Konstantinopolsky - Flohr

quieter life can do very well with some

USSR Cit (Moscow) J 945

of the other Sicilian lines, such as the

I c4 c6 2 ('iJf3 d5 3lbc3 dxe4 4 Qjxe4 t, 11"6 5 COxf6+ gxf6 6 g3 ..tg4 7 .i.g2 e6 K d4 lbd7 9 0-0 !#...g7

have fewer violent forcing lines, and

Accelerated Dragon, which tends


in which positional understanding is more important. Another Sicilian vari­

With White's pawn on g3, the bishop

ation where po�itional understanding

has little to do on the b8-h2 diagonal,

is more important than knowledge of

it makes some sense to place it in­

concrete tactical variations is the so­

skad on g7, where it also defends the

called Kalasbnikov Variation, perhaps

sligh.t1y weakened black king position. 10 l:tel 0-0 1 1 i.e3 lles 12 'i¥c1

better described as the 'Son of Svesh­


/'11"8 13 c3 (D)

nikov' variation:

1 e4 c5 2 li:lf3 ' Cbc6 3 d6 (D).

d4 cxd4 4 ('iJxd4 e5 S LUbS


.\low, instead of Flohr's positionally

In this variation, the early fixing of

lii�ly 1 3... e5?, Konstantinopolsky nx­

the black centre pawns on e5 and d6

,'Ilullended 13 ...J..xf3 14 �xf3 f5, with.

rules out any quick e5 breaks by



White, such as occur frequently in other Sicilian lines where Black plays ...e6 rather than ...e5 This in tum has the effect of slowing down the pace of the game, and rendering it more posi­ tional. If you look through a represen­ tative selection of games in Ihis line, you will tlnd few examples of a quick tactical bhtz deciding the same in 25 moves or so. This is especially true in the main line, where White clamps down on the d5-sl/uare with 6 c4. The subsequent play revolves around each side's attempts to develop their pieces satisfactorily and (especially in Blru:k's case) to effect strategically-desirable piece exchanges. Thus, a key idea for Black. is to delay devdopment of his king's knight, so as instead to play ...ke7 and ....fI.g5, exchanging off his bad bisholJ and removing the main defender of White's weakened dark squares. You can find wme examples of how play develops after 6 c4 in the final chapter of the book, when we take a closer look at Sveshnikov's opening repertoire. It will be clear from those examples that this particu­ lar variation is much less tactical and viol�nt than is generally the case with the Sicilian. So. in conclusion, don't assume that a certain opening is only suitable for one particular �tyle of play. If you like an opening, play it, and do not worry if a celtain beok dismisses it as only suitable for 'positional' players, or 'tactical' players, and you don't: feel that you belong to the relevant cate­ gory.

He's a Lumberjack and he's OK I mentioned above that in my youth, 1 had a certain fondness for queenless tniddlegames. By playing for an early queen exchange, one is not generally seeking a significant opening advan­ tage, but rather trying to obtain a posi­ tion in which one hopes to feel more COlllt�)rtable than one's opponent If you are a player who enjoys endgames, there a number of lines which resull in an early queen exchange, and it makes sense to consider building your reper­ loire around them. With the white pieces, mo�t of the variations concerned arise from dosed openings. We will look at a few exam­ ples in more detail helow. Against the King's Indian Defence. there are various ways for White to ex­ chlUlge pawns on e5, the main one be­ mg:

1 d4 liJf6 2 c4 g6 3 <'bc3 Ji.g7 4 e4 d6 5 llJf3 0-0 6 i.e2 e5 7 dxc5 dxe5 8 W'xd8 :l.xd8 9 ..\tg5 (D)



This line was played �uccessfully in 1970s by Ulf Andersson. 9...,lle8 This is the natural reply. 1 0 I1ldS [0 (}.(}.O i� also po�sible. 1 0 l2JxdS 11 cxdS c6 12 .tc4 Ander��on won several impressive 1',;lIlle� from this position, a characterIStiC example being:



14 1LJd2 "Db6?! 14...lbcs is the critical line. IS .Jib3 .i.e6 16 �e2 idS 17 nhc1 �d6 18 i.xe6 l1xw 19 i.e3 (D)



Andersson - Byrne sao Paulo 1979

1 2...cxd5 13 J.xd5 lbd7 1 3...<'ba6 also led to an instructive ;ld vantage for White in Andersson­ IIlyillsis, Hastings 1979/80 after 14 J" '2toc7 15 i.b3 i.e6 1 6 .ilhdl �xb3 I I axb3 <'be6 18 .ie3 CDd4+?! 19 '';':u14! exd4 20 '>iid3 (D).


rhe passed pawn on d4 is more of a ';Ikness than a strength, and all of lilt' white pieces are better than their I";wl; counterparts. Andersson acti­ \ "h'd his kingside majority by f4 and .''>, and went on to win. II,

To the uninitiated, this position may look dead drawn, but in fact White has a number of small advantages: the more centralized king, control of the c-file, the slightly better bishop. the misplaced black rook on e6, etc. It may not look like a lot, but it proves too much for Rohert Byme. 19...'liff8 20 ltd nc8 21 :fJ.ac1 lIxc3 22 Uxc3 '.t>e8 (D)



23 g4! A gorgeous move, very typical of such po�itions. The threat is artifi­ cially to isolate the e5-pawn, by ad­ vancing g5. White wauld then also have the plan of h4-h5, opening the h­ file for his rook to penetrate. Black can stop this by playing .. .f6 himf:Clf, hut the threat of undermining the e5-pawn hangs forever over his head.

23 f6 24 :b3 .�c7 25 a4 'gc7 26 as QjeS 27 fLlc4 .lld8 28 llbS �c7 29 b3 a6 30 J::tdS �d7 31 gS! ..•

See the previou� note.

in this line should check out the game Larsen-HUbner. Leningrad IZ 1973, in which White won in instructive fash·· Ion. Another opening which offers White the chance to head for a quiet te-.:hnical position is the main-line Queen's In­ dian. beginning: 1 d4 ltJf6 2 c4 e6 3 fLlf3 b6 4 g3

i.b7 5 i.�2 i.e7 6 O�O 0-0 7 1Oc3 ti:le4 Now, imtead of the mosl common mave 8 irc2, White can force some simplification by:

31..JhdS 32 exdS fxgS 33 'i.'d3 g4 34 'ft>e4 .i.e7 35 ltJxe5 �b4 36 tLlxg4 ..txa5 37 We5 '>ite7 38 i.cS+ �f7 39 d6 �:Jb6 40 ..txb6 .ixb6 41 Wd5 .idS 42 tbc5+ 1-0

S ttJxe4 Xt.xc4 9 lbel hg2 10 i1lxg2 (D)

Black's best defence against this line may be 9 ... c6 (stopping White's f2:ld5 move) 10 fLlxe5 .l:I:e8 1 1 (}-O-O! fLla6!. when theory considers Black's position equal. Even here White can keep trying with 1 2 11d6, although he should not achieve anything against accurate defence·. There are other ways to reach a simi­ lar endgame against the King's Indian. .I One such is 6 le3, instead of 6 .ae.2. This line was played successfully for a number of years hy Larsen. Against the most natural responsc, 6...e5, White's idea is to simplify with 7 dxe5 dxeS 8 �xd8 l1xd8 and now 9 m<i. Once again, this does not objectively promi�e White anything significant against best defence, but it leads to the kind of tech­ nical position we are looking far. and in which many King's Indian playeN do not fed comfartable. Players interested


Once again, this position is less easy for Black than it appean;.

10...dS Else White can take full control of the centre with I I e4,

11 'm'a4 The slight vulnerability of Black's queenside light squares now means that he is probably best advised ta ex­ change queens.

11 �e8 ••.


Or 1 L�d7, which is similar,

12 'iWxe8 Ilxe8

Now there is a further pawn ex­ ,"!lange: 13 cxd5 exd5 (D)


fairly unpleasant for a long time to come. with very few chances of losing. One characteristic high-level exam­ ple was Petrosian-Botvinnik, Moscow Weh (3) 1963, in which the queen ex­ change occurred on d7 rather than e8: l 1 ...c5 [2 i.e3 �d7 1 3 �xd7 �xd7 1 4 cxd5 exd5 I S lbf4 llJf6 1 6 dxcS bxc5 [7 :tad d4 1 8 ..td2 a5 1 9 tikl3 lild7 (DJ.


Although the position may look lairly hannless for Black, practice sug­ ).:csts that White has very good chances "I' an initiative. The queenside pawn­ structure is such that Black will almost certainly have to play ...c5, since oth­ l"I"wise his c-pawn will remaln back­ ward. Once he plays ... c5, an exchange of pawns will leave Black with hang­ ing pawns on d5 and c5, and in this lalher simplified position, hanging pawns are certainly something of a lia­ hility. White's subsequent plan will be to pressurize the pawns, and try to force one of them to advance, where­ upon til!' pawn duo can be hroken up hy a timely b3 (if Black has played .(4) or e3 (if Black has played ...d4). White's practical results from this po­ sition have been very good, with the I'('sults suggesting that even if White ,Ioes not win, he can make Black's life .

White has achieved his optimal piece �et-up, and now the thematic 20 e3! dxe3 21 J.xe3 consolidated his ad­ vantage. After a gruelling 86-move defence, Hotvinnik finally managed to salvage half a point. but it is hard to see the black side of this variation ap­ pealing to anyone but masochists. Both of the above variations can be reached via I d4, but the next line, the Anti-Grtinfeld, is only available to those who open 1 lLlf3 or 1 c4.

I llJf3 lLlf6 2 c4 g6 3 tiJc3 dS 4 cxd5

lllxd5 In this position, White has a choice ofattractive ways to avoid the Grtinfeld



proper. Those looking for a sharper middlegame should investigate 5 �b3 or 5 �a4+ (or even 4 'iWa4+on the previous move), but those with a fondness for wood-chopping have the alternative..


5 e4ibxc3 6 dxe3 �xdl+ Black has no good way to avoid the queen exchange. After 6 ... lbd7, 7 $.c4 forces the ungainly 7 ...e6. since the natural 7 . .i.g7? loses immediately to 8 JiLxf7+, while in Andersson-Kouatly, Malta OL 1980 Black tried the ar­ tificiaI 6..."S'd6?! but lost quickly after 7 �b3! Ji.g7 8 .lte3 0-0 <) ttdl �c6 1 0 �c4lDa6? (D). .


actively than their black counterparts with .ic4 and i.t4 or i.e3. Black, on the other hand, has to find a way to ac­ tivate his own bishops and connect his rooks in time to contest the d-fiJe. Above all, the psychological battle is likely to favour White, since he has managed to force his opponent into the kind of quiet endgame position which White is looking for, and which will not usually appeal to most Griin­ feld players.

Once again, Ulf Andersson is one of the high priests of this line, thc fol­ lowing being a model of White's play:

Andersson - Franco Now 1 1 ,txf7+! nxf7 1 2 lid8+ .li.f8 1 3 lbes 't'kf6 1 4 liJxf7 �xf7 1 5 .lth6 �.e6 1 6 lha8! .txb3 1 7 axb3 'Llc5 1 8 f:.;dB+ VWxf8 19 i.xf8 'itxf8 20 We2 gave White a decisive advan­ tage.

7 'otxdl (D) Once again, White has hopes of a small advantage in this quiet position. His king has a nice centra1ized post on c2, and he can post his bishops more

Buenos Aires ] 979 7 f6! ...

Generally regarded as hest. Instead, 7. ..cS?! 8 .i.e3 b6 9 a4 lbc6 10 �bSI i.d7 t 1 r,t>c2 Ji.g7 1 2 Uhd1 a6 1 3 ii.c4 �g4 1 4 h3 �xr3 I S gx.O 0-0 1 6 f4! Ita7 17 eS left White clearly better in Andersson-Tempone, Buenos Aires 1979, while 7 ...!bd7 8 .ic4 �g7 9 llel c6 10 �c2 0-0 1 1 i.e3 h6 1 2 a4



,,) 13 e5 e6 1 4 Itadl 'ue8 15 �d4 b6

Ih .�e3 .if8 17 i.e2 g5 1 8 g4 kg7 1 9

h,1 �xe5 20 lDxe5 .Jtxe5 21 hxg5

hAg5 22 .iD �b7 23 .ixg5 i.g7 24 'J/.e3 :a6 25 i.e2 .:taa8 26 hb6 : +­ was Romanishin-Grigorian, USSR Ch (Moscow) 1976.

" i.e3 e5 9 liJd2! �e6 (D) A key theme of the position is that

White seeks to prevent Black from " xchanging dark-squared bishops hy ..�.c5, which would rid Black of his IIl;Idive bishop. Thus, 9...lbd7 would I...· met by 10 <!flb3! stopping ...i.c5.

This move weakens tbe e5-pawn.

which Andersson exploits immedi­ ately.

16 i.xb6! axb6 .t7 lbc4 Now White has a dear advantage, with the superior minor piece, a target on e5, and the chance of breaking in down the a-file. As always, the little man converts his superiority in impec­ cable style.

I O Jlc4.Yl.xc4 11 lDxc44Jd7 12b4! Again stopping ....ic5.

12...€'lb6?! Slightly better is 1 2. ..0-0-0, with " nly a small edge for White.

I3 liJa5! Tying Black to the defence of the 101 pawn.

13 0-0-0+ 14 <>&c2 i.e7 .••

1 4...h5!, planning ....lih6. would be 111<II"C consistent with Black's positl<lllal aims.

I S a3 fS? (D)

17...SU6 18 a4 ii.g7 19 !tbe! llbeS 20 b5! f4 21 as bxa5 22 nxaS b6 23 Ya7 kf6 24 !'!.eal l::te6 2S �la6 -'!de8 26 ..t>b3 .idS 27 lla8+ 'it'd7 28 :i:l.a2 Jtf6 29 lld2+ <tIe7 30 Aa7 1:c8 31 ltdS <t>e8 32 b3 We7 33 lbb2 WeS 34 lbd3 ii...g7 35 c4 <tr6 36 cS bxcS 37 llJxcs �e7 38 11a6 �h8 39 �c4 i.g7 40 f3 'u'b8 41 lbe6 :if6 42 llc6 I-0 Thus, we can see from the above that players who are happy as White to play for a small edge in a queenless mid­ dlegame have a number of lines where they can achieve the sort of position they want. Even in other variations. the willingness to settle for a near­ equal endgame, rather than trying to



obtain an objective opening advan­ tage, makes one's whole job of open­ ing repertoire management very much ca�ieL It reminds me of .�omething Petrosian wrote about his work with baak BolesJavsky, who became Pet­ rosinn's second during the .I 960."1. Boleslav.�ky was a leading theoreti­ cian, whereas Petrosian's approach 10 openings had always been much more laid-ba<.:k, With his superb intuition and depth of positional understanding, he was accustomed to treating the opening relatively flippantly, alld did not normally strive very hard to gain a theoretical advantage. He wrote that, when anaLysing with Bole>:lavsky, whenever he suggested a relatively tame or second-rate move, Boleslav­ sky would not even try to refute it, but WQu[d simply say "To play like that, there is no need to prepare". Although Boles\avsky's words were clearly meant critically, it seems to me that for many players below master level, having a repertoire where there is minimal need to prepare could in fact be quite altractive. It must be re­ membered that, despite its �hortcom­ ings, Petrosian's approach proved good enough to wrest the world tiUe out of the hands of Botvinnik, one of thc best-prep,lred players ever. We saw above the quiet approach Petrosian used against the Queen's Indian in the third game of their match. Two games later, his play was just as tame-looking against the GrUnfeld: 1 d4 liJf6 2 c4 g6 3 ibc3 d5 4 CZJf3 iLg7 5 e3 0-0 6 2e2 dxc4 7 i.xc4 c5 8 d5 e6 9 dxe6 '8Vxd I + LO Wxdl he6 1 l iLxe6 fxe6 (D).

No doubt, during their preparations for the match, Boleslavsky dismissed this position as not being worthy or preparation, but this did not prevent Petrosian from going on to win a text­ hook ending. A similar example to the above oc­ curred in the game Ander�son-Xie Jun, Belgrade 2000, which opened thus: 1 "ZJf3 tLlf6 2 c4 g6 3 g3 iLg7 4 iLg2 0-0 5 0-0 d5 6 c�d5 1ZJxd5 7 lbc3 4Jc6 8 d4 4Jb6 () e3 IteR. Proving an advantage for White in this line has taxed the ingenuity of many top OMs_ Even Karpov. who has played thl" white side against Kasparov, was Ull­ able to show anything for White. Be­ cause of this, attention has recently switched to lines wherc White avoids the capture on d5, and instead allows Black to play ...dxc4 himself. How ever, if you are a player like Ull Andersson, YOIl do not need to spend weeks poring over the subtleties oJ such matters, and trying to find a way of � out an edge for Whitv Instead, you just take a dead-eqlllli ending with 10 b3 e5 1 1 4Jxe5 4Jxe5


1 ) ,lxeS 'lixdl 1 3 11xdl il.xe5 1 4 i.b2

, II. lurn down all your opponent' s ,haw offers, and eventually grind him

in this case, her) down in 83 lIloves! ()f course, Andersson is an extreme ' dse of a player who revels in queen­ hs middlegames, and plays them magnificently. I remember once hearI'...


their world championship match in London 2(x)(): 1 e4 eS 2 ibf3 ti)C6 3 ..lthS ibf6 4 0-0 iDxe4 5 d4 tiJd6 6 i.xc6 dxc6 7 dxe5 l1Jf5 S tlt'xd8+ �xdS (D).

1111' ahout his attitude to skittles games, \\'IK'II he played at the Hastings tour11;lllIcnt several times in the late I 970s . ..h usual in such tournaments, the

,,1;(yefS could often he found playing hlil/. in the bar in the evenings. In AIKlersson's case, however, he would I" '(IUently remove the queens from the '''",rd, and sometimes the rooks as wdl, before starting play! Needless to ',;IV, he was unbeatable in such games. ' I'hus far, we have looked at the end1'," lIlc-oriented approach from White's

I believe that this line is quite a go(xi practical choice for a player who has little inclination to devote a lot of


time to the study of opening theory,

1';1I'lly because White has the first Ill<'Ye and can dictate the pace of the 1';(IIIl' more readily than Black. In ao­

and who enjoys playing endgames and technical positions. Despite the fact that it has become very popular at top GM level, it is a line which can be played at less exahed levels with rela­

.. k Naturally, it i:o. much easier to hllllg about :o.uch queenJess middle­ I'<IIIIC positions with the white pieces,

,lll l<lIl, players who are Black are gen­ ,'(ally less averse to a draw, and thus 1I" "'l' likely to allow an early queen ex­ , h;mgc, in the oftcn mistaken belief 11i;11 it promises easy equality. Never­ I I wkss, players who are happy with a .Jlu"'·1\less middlegm.y.e have one very " I'\,1I1US choice against 1 e4. and given

,I ', l" "'sent popularity, J cannot pass it I,v wilhout some brief comments. I re-

10-,. Ill' course, to the infamous Berlin

I J" "'lIn�, which Kramnik: u:o.ed so SilC­ " ...�rllily to blunt Kasparov's 1 e4 in

tively little preparation, because the lines involved are almost all non­ forcing, and rely much more on under­ standing than on knowledge of spe­ cific tactical sequences. In addition, Black has many different ways to play, which makes specific preparation by White rather difficult. After the main move 9 lbc3, for example, Black has at least five playable replies : 9.. .WeS, 9 ...h6, 9 .. ,.td7, 9.JDe7 and 9 . ..aS, all of which are similar, yet also subtly different.



Without in any way attempting to write a textbook on the line, it is worth outlining its main positional consider­ ations. As in the Exchange Lopez, Black has doubled queenside pawns, but ha'> the two b ishops. An added fac­ tor here is that he has lost the right to castle. Although the absence of queens means that he is unlikely to fall victim to a dired attack on his king, the main drawback of this is thaI he is likely to have some difficulties bringing both his rooks into play. So, one may ask, why hasn't White simply got a superior version of the Exchange Lopez? After all, he has the same pawn-structure as in that line, plus he has excnanged queens (usually to White's benefit) and he ha<; mis­ placed the black king. The answer is that, unlike the Exchange Lopez, here White's e-pawn has been .lured for­ ward to e5. This apparently insignifi­ caut factor makes all the differeuce, because it means that Black's minor pieces have the use of the excellent squares d5 and f5, and in addition, the eS-pawn obstructs White's remaining bishop. Indeed, one of Black's prin­ cipal strategic aims in this line is to ex­ change his fS-bishop for White's c3knight, thus producing an opposite­ coloured bishop position. By threaten­ ing ...a4 or ...c4, White can usually be induced to place his queens ide pawns on the light squares, where they are vulnerable to Black's bishop. coming round the back via the f5-square. By contrast, White's bishop is usually un­ able to get round the back of Black's queenside pawns, because tne white

pawn-Shllcture on e5 and f4 bishop's path. Another major theme of the line is the attempt by White to mobilize his kingside pawn-majority by f4 and f5. Black in turn must fight against this, trying to engineer a blockade by moves such as ... h5. Often, White will ad· vance g4 to expel a black knight from fS, but in this case, he must b e on the alert for the subsequent counterblow . . .hS, which may undermine his king­ side pawn phalanx. There arc numerous GM games be· ing played with this line at present, so there is no shortage of study material for players who are interested in it. As well as Kramnik and the late Tony Miles, another player who specializes in the line is Alexandrov, some oj whose games on the white side of the Semi-Slav we studied earlier. I f you are interested in the Berlin Defence, a study of the games by those players will teach you a great deal. Naturally. White is not obliged to answer l ... c) with the Lopez, and indeed, at cluh and league level, the non-Lopez Opell games are quite popular. Nevertheless. with the possible exception of th,' Scotch Game, none of these poses to., formidable a threat to a reasonabh well-prepared player, so if you an' comf0l1ahie with the Berlin, you wi II be well on the way to a very sol id and dependable repertoire against I e4. What I have sought to achieve in tl1<' foregoing is to give an exampleofhllw one's opening repertoire can be buill around one's stylistic preference. Thll� a player whose style tends toward" central

obstructs his



nulgames and technical play can build

result that relatively little time is re­

,III opening repertoire around this, by

quired to keep one's theoretical knowl­

, llollsing various lines which produce

edge up to date. Naturally, one cannot

' I1U'l�nless middlegames of the type

always achieve the type of position one

',Udl a player should enjoy. Because

wants, even with White, hut having

IIB'SC lines, by [heir very nature, are

repertoire based around the lines con­

II, HI-rorcing, their theory is unlikely to

sidered above will enable one to do so

, h;lIlge radically over time, with the


in a good proportion of one's games.

4 Main Roads or Side-Streets?

Another impOitant decision to make when formil1g one's opening reper-­ loire is whether to concentrate on pop­ ular main-line openings, or to attempt to avoid theory by employing little­ known sidelines. There an: advan­ tages in each approach, although I have clear views on which is better for certain types of players.

Following the Backstreets For many club players, there is a temp­ tation to avoid main-line opening the­ ory and instead play oHbeat openings. This has the great merit of avoiding one's opponenfs theoretical knowl­ edge and thus throwing him much more on his own re�ources. Tn addition, many of the lines Olle sees played at dub level lire gambits, which lead to the kind of exciting open play that many players fLild attractive. There is no doubt that such an ap­ proach can prove effective, even up to master level. Indeed, the anti-theoreti­ cal approach has been very popular in Englund over the past 20-30 years, with a number of the leading English players building their opening reper­ toires around offbeat lines. One of the tiHt to do so was the late Tony Miles, who, after using the Si<.::ilian Dragon extensively in his early career, gave up

the line in favour of less well-wol1l theoretical byways. Tony had great success with this approach. culminat­ ing in his famous win over Karpov at Skara 1980, when he played i e4 a6 2 d4 b5 as Black. Ont: of Tony's earlier favourites was J d4 e6 2 c4 h6 (or 1 c4 b6). 111is was also taken up by other English players, and became known as the English Defence. The following game is a good example of the effec tiveness of this defence in Tony'� hands against unsuspecting opponents.




Hustin,;s 1976f7 1 d4 b6 2 c4 i.b7 3 ttk3 e6 4 c4 ..Iib4 5 �c2 W'h4! At the time, such moves created something of a sensation, but they soon became a thematic idea in this opelllng.

6 <Jtd3 fS (D) 7 g3'? Already 11 seriolls error. White has to try the more adventurous 7 lLln �xc3+ 8 �xc3 \¥Ug4 9 0-0 fxe4 JO 4:Jes, although it is perhaps under standable that Farago was unhappy ahout heing forced to tish in SUl.:h murky waters so early in the game. 7 J!'hS 8 �e2 '/&f7 9 f3 fxe4 10 ..

fxe4 �f6 11 d5


Already White is having tremen­

d"us difficulty holding his centre to­

I' l'!her, and Miles's subtle play soon i>rings it crashing down.

11 0-0 12 lLit'3 '8g6! 13 �d3 't'Yh5! 14 0-0 lba6 15 a3 .!iLxc3 16 hxd ttJcs 17 i.e3 4lxd3 18 �xd3 ndS 19 exdS lOxd5 20 cxdS )hf3 21 I1d3 �xf3 (D) ...

White's once-proud pawn-centre 11.,,, lwenreduced to rubble, and he is a 1'''\\111 down with an exposed king. He ,,"Iy managed to struggle on for a few IIII,n' lIloves.


Miles won many such games in the 1970s, hefore opponents had worked out how to play against the English Defence. Gradually. however, White developed more �uhtle, less coopera­ tive ways to develop, and the line be­ gan to claim fewer drastic victims such as Farago ahove. Nonetheless, it remains playable to this day, with Jon Speelman being one who still uses it quite. regularly. Ifyou are interested in it, the most critical line is probahly I d4 e6 2 c4 h6 3 e4 .lib7 4- i.d3. Now the sharpest and IIJO�t thematic reac­ tion for Black is 4...fS, which leads to fanta�lic complications after the criti­ cal response S exfS .'ibg2 6 'iM'h5+ g6 7 fxg6 ,tg7 8 gxh7+ �f8 9 lbe2! lZJf6 1O '§'h4 .il.xh l (D).

This wa� tir�1 played in Browne­ Miles, Tilburg 1978. White has sacri­ ficed a whole rook, but will follow up with �g5. ltJf4, etc., with a ferocious attack. Many attempts have been made to justify Black's position, but it is hard to recommend 4...fS to any but. the very bravest players. Instead, Black



should probably prefer 4... �6 5 ttJe2 (or 5 fiJf3) 5 .. .ltJb4, when Black simply lakes the bishop-pair and settles for a cramped but solid position after a subsequent iJe7, ...0-0, . .d6, ele. A recent example was Lamprecht­ Speelman, Bundesliga 200112, which continued 5 fbf3 lllh4 6 d5!'! lLlxd3+ 7 �xd3 .th4+ 8 AdZ i.xdZ+ 9 ttJbxd2 ctJe7 l 0 0-00-0 I I lIfel !{)g6 l 2 "S"c3 '¥lie? 13 ctJn c6 14 dxc6 dxc6 1 5 ttJg3 l:.fd8 16 11adl c5 (D) with an equal . ..t



position. dictates the choice of opening, and there is also a good deal of flexibility in the way one can play Ihe white posi­ tion. Against 2 .liJe4, for example, White can choose between 3 .i..f4, 3 ..th4 and even the outlandish 3 h4, with which Hodgson scored many vic­ tories. AgainRt quieter replies, such as 2 d5, White also has a choice of strat­ egies. 'the miginal idea was to capture on f6 if allowed, doubling the black pawns. In more recent years, however, Hodgson and olher 'Trumpers' have tcnded to prefer simple development by 3 e3. Note that White keeps open the option of playing c4 at some poinl, which means that Black must be care­ ful not to allow a transposition into an unfavourable Queen's Gambit varia­ tion. Another potential advantage of the Trompowsky is that it can be played against l...dS as well, although it is generally accepted that the line has less bite in thai case. A typical line is 1 d4 d5 Z �g5 h6 3 Jih4 c6 4 e3 l/ikb6 5 'i'c I i.fS, when he will follow up with .. .e6, . ..'bd7, . li:lgf6, etc. Black has ..



Another English player who has made a very successful career out of offbeat openings is Julian Hodgson. His greatest weapon over tbe yean; has been the Trompowsky, I d4 !iJf6 2 Ji.g5 (D). Almost single-handedly, Julian has turned this opening into 11 much-feared weapon in British chess, so much so that it has become a main-line open­ ing. There are many obvious merits White avoid.>: his opponent's normal queen's pawn defence, hc also himself






fairly comfortable play here, although chances are no more than equaL At c1uh level, one also sees many examples of gambit play, Openings such as the Blackmar-DiemerGambit, Albin Counter-Gambit, Latvian Gam­ bit. etc., arc all played regularly at club and league level, whereas they hardly ever get an outing in masLer tournaments. There is a very simple reason for this, of course - although interesting <Iud difficult to meet over the board, such openings are not really sound, and very few grandmasters are willing to risk them. Nonetheless, there is much to be said for playing such systems at lower levels of play, particularly if you enjoy playing sharp atrucking lines. Many players find it uncomfortable to defend against an opponent who is prepared to sacrifice material in return for speculative at­ tacking chances. In such positions, the cost of a single error is much higher than in quieter openings. If you make an error in a typical Reti Opening middlegame, you may end up getting a small positional disadvantage, but if you make a similar mistake in the sort of wild tactical positions which often arise from gambits, you are quite likely to tind yourself being mated, or losing a substantial amount of material. One of the things to bear in mind when playing gambits is that it is not always necessary'to burn one's hoats in search of an insrunt knockout. Quite often, it is possible to play in a more restrained fashion, aiming for long­ tenn compensation [TOm piece activ­ ity. One example is the so-called Von

Hennig-Schara Gambit, which arises as follows: 1 d4 d5 Z c4 e6 3 lllc3 c5 4 cxd5 cxd4 This last move distingnishes the Von Hennig-Schara from the nonnal Tarrasch Defence (4...exd5). Themrun line now runs: 5 �a4+ i.d7 6 �xd4 exd5 7 'fNxd5 I1lc6 8 1ilf3 Iilf6 9 Yil'dl .ltcs 10 e3 W/e7 11 .teZ At this point, the normal move for Black is 1 1 . ,0-0-0, followed by .g5g4. However, the open c-jile means that Black's king does not feel terribly safe on the queenside, and White is able to open up further lines by return­ ing his extra pawn with the move b4, opening the b-file. Instead of all this, Black has a less explored and some­ what less risky way to play: 11...0-0 12 0-0 l:tfd8 (D) .


The white queen is slightly embar­ mssed for a comfortable post, and his queenside development is also lagging. Black has plans such as ...l:tac8, fol­ lowed perhaps by ...tZJe5, looking to



exchange off the knight which defend., 'White's king. If White exchanges on e5, the recapture ..."i!txe5 will probably provoke a weakening of the white king position by g3, after which Black can aim for moves such as ... i.g4 a.nd ...'t!¥h.''i, etc. Another idea fur Black, in­ stead of ...the.5, is ... a6, followed by, h-ying to create threats along the b8-h2 diagonal. While Black does not have full compensation for the pawn, he certainly has some initiative, and the white position is probably less easy to play over the board, especially at a fast time-limit. It is interesting to note that Fritz, usual1y an extremely materialistic judge of positions, only assesses White's advanrnge here as less Lhan half a pawn. I have already alluded to the fact that the biggest prohlem with such off­ beat lines, especially the gambits, is that objectively they are not fully sound. This is something which one must simply recognize and accept. If it were possible to get a lively attacking position without any risk, everybody would do it. Instead, one has to under­ stand that in order to obtain sharp ut­ t<u:king chances, especially as Black, one has to take risks; if you don't like doing this, you should he playing something much more solid. Another problem which one en­ counters in playing offbeat openings is that there is frequcntly a dearth of rdiabk books ab(Jut sut:h systems. Because such lines rarely get tested at master level, an awful lot of the analy­ sis which finds itself into opening books is of very poor quality, often

based on a mixture of 19th century mismatches and 20th century Internet blitz games. Few strong grandmaster� are going to write a book about a dis·· credited and probably unsound gam­ bit, so books on such lines ate usually written by weaker players, who are oftcn themselves enthusiastic practi · tioners of the opening and lack the necessary objectivity. I n his \xlOk Sc crets of Prm·tir:al Chess, John Nunn gave two splendid examples of thl' dangers of such books, even when writtcn by grandmasters. 1 recall one small instance from my youth, con eerning the Ponziani Opening. After the characteristic moves 1 e4 e5 2 lfJf3 lbc6 3 d Blad: has a nUlR ber of satisfactory defences, but one o( his sharpest and most ambitious tries is the Leonhardt Gambit, :L.d.5 4 �ac( liJto. With this move, Black sacrifice., the e5-pawn for rapid development During the mid-1 970s, a book wa� published advocating the Ponziani Opening for White, and r and a clull friend decided to take a closer look al it. My friend had played the openin;' in the past, but had been put off by ttl(" Leonhardt Gambit. Naturally, there fore, we were keen to see what the au thor recommended for White, He g�lV<" the line 5 lOxeS .id6 6 lOxc6 bxc6 "I d3 0-0 8 .lte2 ne8 9 ,ltg5 h6 \0 StX(h 'ID'xf6 I I 'lWxc6 i.e6. This had all bee'JI playcd in a top-level correspondclicT game, which had nmtinucd 12 4)d ' 'iiYgS, when Black had compensatj" 1 1 for his material. However, the autll(l1 instead recommended for White t(w 'obvious' move 1 2 0-0 (D), claimill!'


Illat with the white king now out of the 'Tntre, it was clear that Black had in­ sllfficient compenrmtion for his two Ilawns.


1 ':ven

in those bygone pre-Fritz it did not take me and my friend \'("IY many seconds to spot the reply 1 " _ .. �g4! after which White can re­ '.1)',11 with a clear conscience (the at­ h'IIlJll to harvest some wood for the '111" ,'11 by 1 3 'tlYxe8+ 1'4xe8 1 4 �.xg4 !'lIls to 14. . :114). Th� final problem with playing off1 ...:11 openings relates 10 one's deveiop1I1t'1i1 as a player. If you have serious ,1I11i1itions to become a master-strenglh I'L,V('I', the likelihood is that you will I O ,,,-h a point where your opening '''Iwl'loire becomes a limiting factor in V'"11 development. Playing offbeat and "'llh,·,)rctical openings can work vcry , I kd ivdy against Weaker players, but ,,110 ,- Y')l! start facing IMs and OMs on .I l<'l,ulm' basis, you will no longer be " hlo- 10 count 011 success with those '. .l111t' systems. Of course, you can at 11,.,1 11\11111 start to play more respectable ci:I,Y:-;.


openings, but if you have already spent most of yonr fonnative years avoiding such lines. you will be start­ ing the process some way behind other players who have been using main­ line openings most of their careers. I mentioned at the start ofthi� chap­ ter that the use of relatively offbeat openings has been a characteristic of the English school over Ihe past 25 years or so. One example of how such an approach can potentially rebound as a player becomes stronger is the English GM Joe Gallagher. Always an aggressive and courageous player, Gallagher has played the King's Gam­ bit ever since his junior days, and amassed innumerable scalps with it, even amongst GM opposition. He iH also the author of an outstanding book on the opening. However, over the pa�t few years, the Kiug's Gambit has been played at the very higheHt level hy such players as Morozevich, Short and others. As a result, it has been thoroughly tested against the very best opposition in the world, and occasion­ ally te�tcd to destruction. The follow­ ing is one brutal example of what a modem world-class OM can do to the King's Gambit:

Fedorov - Shirov Polanica Zdroj 2000 1 e4 eS 2 f4 exf4 3 4\f3 g5 4 h4 g4 5 .'Des d6 6 lZJxg4 1tJr6 7 liJf2 �g8 8 d4 -Ith6 9 lbc3 ttJc6 10 ctJd5 tbxd5 11 cxd5 (D) Fedorov had reached this position at least twice before, each game going



I I .JiJe7, when White has good play. However, a forewarned Shirov pro­ duced a brilliant piece sacrifice:

1l tWc7+! 12 i.e2 ibb4! 13 c4 kf5! 14 'ii'a4+�f8 1S �xb4lk8(D) ...


Black already has a ClUshing attack. The game finished:

16 �d2 J;txg2 17 'Ot>n. Itg3 18 �d1 .te4 1911h2 fS 20 .!t."ixt4 fxc4 21 .tg4 c3 22 i.f3 'ii'g7 23 �h1 .llg2 0-1 This and other games have shown that the King'� Gambit, while better than its fonner reputation, is ex­ tremely risky at top level, especially

against opponents who are prepared for it. In these circumstances, it is very difficult for a professional GM to con­ tinue using the opening as his main weapon. The reality is that a GM who opens I e4 needs to play the main lines of the Ruy Lopez as his chief weapon if he is to count on getting an advan­ tage with White against strong opposi­ tion. However, for a player who has virtually never played the Lopez all his chess career, it is very difficult to learn the opening from scratch. This dilemma is noticeable in the play of several GM-strength King's Gambiteers. PriOT to the above disas­ ter, Alexei Fedorov had scored a huge llUmber of victories with thc King's Gambit, in the course of which his rat­ ing had risen to over 2660. Once he started playing elite GM events, how­ ever, games such as the above soon dented both his confidence and his rating, the latter falling almost 100 points. At the tillle or writing, he ap­ pears to have abandoned the King's Gambit in favour of the Bishop's Opening, but although the lalter is un­ doubtedly a safer choice, it is hard to sec it striking fear into the hearts of potentiaJ 1 .,.c5 players. In recent years, Joe Gallagher has also suffered some problems against 1 ...eS. In a crucial last-round game al the Hastings Challengers a few years ago, as White against (gors Rausis. he was already reluctant to risk the King's Gambit against a strong opponent who had obviously prepared somethin)!, against it. Instead, with tournament victory and a place in the following


year's Premier tournament at stake,

avoided the King's Gambit in

he favour

of the Four Knights, only to fall into a poor position fairly quickly and even­ lually lose. On other occasions, he has


I e4 altogether when faced


defence 1 e4 c5 2 f4 d5 3 exd5 itJf6! drew much of the sting from the f4 Si­

cilian, while Hebden too began to lose faith in the King's Gambit after some nasty reverses against well-prepared master-strength

opposi tion. Mark's

with a strong opponent who was likely

response was to revamp his repertoire


1 980s.

It) reply 1 ... e5, the most recent exam­

I have seen being against Mark I lebden at Hastings 200112. That time Ilc opened 1 d4 and found himselffac­ ing the King's Indian Defence, Gal-

1;lgher's own favourite defence. Again,

he achieved


no advanlage from the

drawing a quiet game. It is

entirely, switching to

I d4 in the lateAlthough the switch was very successful, it is clear that having to make such a fundamental change in one's opening repertoire is not partic­ ularly desirable. Indeed, in a recent

\:onversation with me, Mark revealed

that he has begun doing some junior

rairly clear that Gallagher has never


( :ambit, and the resulting hole in his

study main-line openings, mther than relying on offbeat trickery to bamboo­

'luite succeeded in replacing the King's

H'pertoire continues to give him prob­ iL'llls against strong opponents.

Hebden himself is another example

and that his approach to

openings is to encourage his pupils to

zle weaker opponents.

"r Ihis phenomenon. His current rep­

Slicking 10 Ihe Molorway

III Chapter

All this leads us on to the alternative

'"flnire will be examined in more detail

9, but he stllfted out his

(";I!"Cer playing 1 Iwo


with White. His

main weapons were the King's ( 'ambit and 2 f4 against the Sicilian. Indeed, it was Hebden who taught ( tallagher the King's Gambit - in heu ,,[ paying off a large gambling debt,

II<' undertook to 'sell' Gallagher hifi l lIowledge of the King's Gambitl Play­ Ilig most of his chess in weekend

Swiss events in England, where beat­

III!'. weaker players was the key to

'.1 n,ess, Hebden',s, little-known but ag­

!'Il"Ssive lines were perfect. However,

,IS he hecame stronger and his lines W,'ll'

exposed to tougher and more

I'lUl"cssional opposition, they


Ii> yield less fruit. The discovery of the



biggest advantage in

sticking 10 respectable

main lines is that youcan rest assured that the open­

ing you are playing is fundamentally

sound, and

is not likely ever to be re­ futed. Of course, different lines go in and out of fashion at GM level, and if you discuss opening theory with a strong professional player you will of­ ten hear comments like "Such-and­

such a line is under pressure at pres­ ent" or "Nobody is playin g this line at

the moment". Typically, what happens

is that a game is played in which (say) White produces a new wrinkle in the

opening and gets an edge. If the next

few tournaments do not immediately



disclose an improvement for Black, the line will fall out of fashion and the top OMs will start avoiding il. How­ ever, these periods usually only last a relatively short time before the elusive improvement is found, whereupon the line returns to popularity again. There are alnKlst no cases of a respectable main-line opening being refuted, nor is this eveTlikely to happen. By defini­ tion, the lines we are talking about are based on sound opening principles, and for them to be refuted would entail a fundamental fe-think of the basic tenet� of opening play. Another benefit of the main-line approach i s that one always has a ready �ource ofGM games that can be used to study the opening. Further­ more, the fact that an opening is played extensively at top level means that what is written about it is likely to be that much more reliable than the pub­ lished games and analysis of other 1ine�. A third merit of playing main lines is that many of them offer great flexi­ bility. Thi� can be very useful if one particular variation comes under pres­ sore. Take as an example the Closed Lopa, Chigorin Defence, the basic position of which is reached after I e4 e5 2 4Jf3 €.Ic6 3 J�.h5 a6 4 ji.a4 tbf6 5 0-0 i.e7 6 :tiel b5 7 .ltb3 d6 (; c3 0-0 9 h3 4')a5 10 J tc2 c5 I I d4 �c7 1 2 Chbd2 (0). At this point, Black has at least five playable moves: 1 2.. .cxd4, 12...lbc6, 12 ....id7, 12....Jtb7, 12...l:.d8, etc. Of these. the most active lines are con­ nected with the opening of the c-fiJe by


J 2.. .cxd4 13 cxd4. Now once again, Black has several playable opti ons 1 3.. .�,d7, 13...l0::6, J 3. . .�b7, B. ..:tdS, etc. These lines are clearly all similar, yet slightly different. They lcad to po­ sitions which have many of the same fundamental strategic and tactical fac­ tors. If you play one of these lines for Black. and then a particular theoreti­ cal problem appears, you can always switch to one of the other variations, thus avoiding the specific sequence ynu are concerned about, hut still reaching the same general type of po­ sition where your overall experience and understanding of the structure is valuahle. Indeed, if you decide to take up a line like this, you should try to study the characteristics of the struc­ ture itself, without at first worrying too much about specific sequences. Another comment Mark Hebden made in the conversation refelTed to above was that he generally tries to get slu­ dents [0 study structures, rather than just specific lines. This same flexibility exists in many openings. Taking the King's Indian as


another example, suppose that after J ;J4 tLlf6 2 c4 g6 3 tLlc3 i.g7 4 e4 d6 5 tLlfl 0-0 6 iLe2 e5 7 0-0 you have al­ ways been in the habit of playing the line with 7 ... t;)bd7. However, a partic­ ular response for White proves diffi­ ..:ult to meet. and you decide you need to avoid this. Instead of having to abandon the whole King's Indian and switch to something entirely different, you can just explore one of the alter­ natives, such as 7 ...4.\16 or 7_ .. exd4, both perfectly playahle and respect­ ilble lines. In all l"aSeS, you arc still go­ ing to reach pesitions similar to those you are used to getting from 7 .'tJ1x17, so you will not be wasting your knowl­ edge and understanding of the King's Indian in general. The principal drawback of playing main lines is that there tends to be a lot of theory around them. TIlis CaJl be a problem if the line is very sharp, such as the Najdorf Skilian, because on..: simply ..:annol pl<ly the line safely without studying the complications and remembering a lot of (;onuetc variatiolL�. If you are averse to doing this, or you have a poor memory, you are better off avoiding such lines. However, that does not mean that you cannot play main-Iinc openings, only that you need to choose ones that re­ quire less concrete knowledge and are based more on understanding. The Closed Lopez lin�� considered ahove are one example, and can be played with few or no specific lines C-<lmmitted to memory. The other respect in which the vol­ ume of main-line theory can be a ..


problem is that one's opponents arc more likely to know what they are do­ ing in sllch lines. and can prepare them more easily. However, {his is again something one must simply accept. As with life in general, there is no such thing as a free lunch in chess, and everything (;()mes with a price tag at­ tachcd. In return for having a respcLi­ able, reliable opening. which has been tested extensively at GM level, and on which there is much material one can study, one must ac..:ept the (;onse­ quence that one's opponents arc also able to access that same material, and develop their own knowledge of the opening. Ifyou wanl an opening which only you know anything about, you will need to invent semething which has virtually never been played be­ fore. By all means do so, but remem­ ber the price which is attached to that approach - whatever you play is likely to he unsound!

Conclusions 1) Offbeat openings are generally otlbeat for a reason - they are not usu­ ally as good as the respectable main­ line openings. If you are going to play unusual lines, you must understand this and accept the risks involved. 2) Playing offbeat openings will probably prove more effective against weaker players, since one can thereby get tbem out of their book knowledge and on their own resources at an ear­ lier stage of the game. However, when playing stronger players, such an ap­ proach can backfire.



3) Be careful with books on un­ usual openings. All too often they are by relatively weak players, lack objec­ tivity, and contain significant amounts of unreliable and untested anaJy�is. 4) Main-line openings are going to be objectively sounder and generally offer more choice of variations within the opening. Also, books on sueh openings are more likely to be by strong and trustworthy analysts.

5) In general, despite the risks at­ tached to spending too much time on openings, I believe that young and am­ bitious players should be prepared to get stuck into main-line openings rela­ tively early in their careers. There is no point in wasting one's formative yeaf� on an unsound Of duhious open­ ing, which one will be forced to jetti­ �on as soon as one starts meeting strong opposition on a regular ha�is.

5 M ove-Orders and Transpositions

Paying ull-eful attention to move-order.> and transpositional tricks is an essen­ tial part of huilding a successful open­ ing repertoire. We saw in Chapter 1 how a player who fails to think out all the relevant move-order possibilities can come unstuck. However, move­ orders are not just a potential pitfall of which to be wary; they can also be used to one's advantage. By intelligent use of move-orders, one can often avoid particularly dangerous variations and restrict the opponent's choice of lines_

A World Championship Story The example quoted in Chapter I in­ volved the sequence 1 d4 �f6 2 c4 e6 3 lbf3 d5. As an example of how world champions approach issues of move-order, and how repertoires de­ velop over time, we will look at the Kasparov-Karpov world championship matcnes of the 1980s, and, more spe­ cifically. at Kasparov:� repertoire as White in these matches. At thc start of their first match in 1984, Kasparov's principal opening move as White was I d4. He had played this as his main weapon ever

since breaking through to GM level in the late 1970s, Against 1 ...4:lf6 2 c4 e6 his favourite continuation was 3 lZ\f3, avoiding the Nimzo-Indian. If Black then played the Queen's Indian, 3 ...b6 Kasparov would usually reply 4 a3. This last move, originally populruized by Petrosian, had been moulded by Kasparov into a formidable weapon. Tn the 1984 match, Kasparov followed his usual policy of not allowing the Nimzo-Indian, and the players had a long series of games with the Queen's Indian. In addition to 4 a3, Kasparov also played 4 g3 in many of these games, something he had only played on a few previous occasions and had clearly prepared specially for the match. As is well-known, this was the marathon match which was supposed to have no limit on the number of games, and it was game 32 before Kasparov finally scored his first win. That game featured his favourite 4 a3 variation against the Queen's Indian. Karpov's reaction to this was im­ mediate. In his next five games with Black, he answered 3 lLlf3 with 3...d5, transposing back into the QGD. Each of these games featured the Tarta­ kower Variation, 4 ibc3 iJ.e7 5 .tg5 h6 6 .i.h4 0-0 7 e3 b6 (Dr



avoid the line. Karpov duly followed the trend: 3 i.h4 Now Kasparov revealed the other point of his preparation: 4 11lf3 (D)




This !.ine, an old favourite of Kar­ pov's, proved very solid and he drew all five games. In the end, Kasparov actu­ ally gave up playing I d4 altogether, and in his last three games with White he switched to 1 e4, winning the 48th and final game, before FIDE President Campomanes arrived in Moscow and made his infamous decision to termi­ natc the match. Less than a year later, the two play­ ers faced each other again. Clearly, Kasparov had hecn working in the in­ taim on a solution to his problem. The result of his work was seen in the very first game of the match:

Kasparov - Karpov MOSCOlt' Well

(I) 1985

1 d4 ctJf6 2 c4 e6 3 lbc3(!)

This move was the first clue as 10 Kasparov's new approat:h. If Black now tries the QGD, White can play the Exchange Variation with €lge2, as outlined in Chapler I . This is gener­ ally considered a little dangerous for Black, and most players prefer to

This move had long been (;{)llsid­ ered harmless, and had neVer been played by Kaspamv before. Black now has several options. 4.. ,b6 transposes into a Queen's Indian line, but with Black having already committed his king's bishop to b4_ Although play­ able, this line is quite sharp after .5 .\tg5, and in the normal Quecn's In­ dian move-order 3 lbf3 b6 4 fbc3, many players prefer 4 .. th7 so as to answer 5 j.gS with 5_ ..1i..e7. Kaspa­ rov's move-order has denied Black this option. Similarly, 4...dS, the Rago­ zin Defence, is less solid than the stan­ dard QGD, since oncc again, White's .\tg5 movc pins the black knight. The Ragozin is a sharp and rather risky line, definitely not in Kal'pov's style. He instead opted to keep a NinlZo-In­ dian character to the position: .




4 c5 5 g3 ...

Once again, Kasparov makes intel­ ligent use of move-orders to restrict Black's options. He has now trans­ posed into the so-called Romanishin System against the Nimzo, which usu­ ally arises via the move-order 1 d4 li.:lf6 2 c4 e6 :; 4Jc3 .tb4 4 g3. Now 4...cS 5 tZ\f3 would transpose into the K-K game under discussion. However, in the normal Romanishin move-order, Black is not forced to answer with 4...cS. Instead, he has several other op­ tions, most notably 4...0-0 5 i.g2 d5 6 lUf3 dxc4 7 0-0 CLk6. This is cOllsi(l­ ered the most solid response for Black, since in this Catalan-like structure, the white knight is not ideally placed on c3 for winning baek the c4-pawn. By his clever move-order, Kasparov has avoided this line, by waiting until Black is wmmitted to ...c5 before playing g3. With the additional advantage of surprise on his side, Kasparov soon reaped the !:>enefLts of his suhtlety:

S lZJe4 6 �d3 �a5 7 �xe4 J..xc3+ 8 �dZ $.xdZ+ 9 ltJxdZ \'¥!b6?! 10 dxc5! �xbZ 11 llbl '/We3 12 '$"d3! 'i'xdJ 13 exd3 •••

White has a clear advantage in the ending, and Kasparov went on to win. Kasparov played 4 4Jf3 in five more games of the match, and Karpov never succt.-eded in linding a reliable equalizer. In the end, he decided to fall back on the QGD. but using a standard move-order trick to avoid the danger­ ous ibge2 Exchange lines: I d4 d5 2 c4 e6 3 lbc3 .il.e7(!) (D).


With his 3rd move, originally an idea of Geller's, Black aims to get a normal QGD, without allowing White to play the Exchange Variation set-up with 'Llge2. After 4 cxd5 exd5 White is unable [0 continue with 5 .il.g5, and must commit his pieces in some way. Then 5 .tf4 leads to a somewhat dif­ ferent position from the JLg5 and lUge2 lines. This theoretical duel continued in the third K-K match. played in wn­ don and Leningrad in 1986. Karpov started off by defending the 4 4)f3 Nimzo again, but after surviving a lost position in game 2. and being thor­ oughly drubbed in game 4, he went back to his QGD with 3 ...SLe7. This proved fairly solid, and gradually Kas­ parov started playing 1 e4 more often. The process continued throughout their next match in Seville 1987, but gradu­ ally Kasparov was coming to the con­ clusion that I e4 was the way to inconvenience Karpov most. When they played their final match in New York and Lyons in 1990, Kasparov played 1 e4 in 1 1 out of his 12 games



as White, only switching to I lbf3 in the final game, when he had retained his title and a draw was enough to seal victory in the match. Thus, we can see from this story how top playel":'l use move-orders and transpositions to avoid certain lines which they do not like, and to cut down their opponent's options in the open­ ing. This is something every player can do, with some intelligent thinking about his opening repertoire.

reasons, the Benko has always re­ mained popular, especially at club and league level, and it seems to score very well for Black in practice, As a result, many players prefer to avoid the Benko, which can be done without actually abandoning I d4, In­ stead, White can play:

1 d4 tDi'6 2 tbf3 c5 3 d5 b5 4 :tg5


Options Trading In planning one's opening repertoire, it is vital to understand how each move affects the player's range of op­ tions. We have already looked at some Nimzo-Indian and Queen's Indian ex­ ample:;; now let us consider some other openings. After 1 d4 tt.)f6 the most common follow-up is 2 c4. One of Black's most active defences against this is the Benko Gambit, 2.,.c5 3 d5 b5. With this long-term positional pawn sacri­ fice, Black aims at pressure down the open a- and b-files, Although the past 30 years have seen the Benko sub­ jected to intensive testing at all levels of play, no clear route to a signiticant advantage has been demonstrated for White. In many positions, White can gain a theoretical edge, but even then, it is very hard to make anything of it against accurate play. In addition, the active black pieces and clear plan makes Black's position rather easier to handle in over-the-board play, espe­ cially at fast time-limits. For these

This line first came to prominence in 1 980, when six-time US Champion Walter Browne used it to defeat Kor­ chnoi in II dramatic last-round game at Wijk aan Zee. White is aiming for an altogether different structure from the normal Benko. Since White has not played c4, Black is not threatening to open the b-file. In addition, White can play c3, blunting the effect of a black bishop on the long diagonal. Another key element in White's strategy is the undermining advance a4, after which the reply ...b4 will allow White to set­ tle a knightoll the powerful square c4. Black has a number of playable moves after 4 i.g5, including 4 ...d6,


4,..tbe4, 4...�6 and 4...i.b7. The fol­ lowing examples illustrate the kind of llOsitions Whit� is aiming for.

v. Mikhalevski - Knoppert Antwerp 1999

4 d65 Jtxf6 exf6 6 e4 a6 7 a4 b4 8 ..Itd3 SLe7 9 4jbd2 0-0 10 ttJc4 as 11 ttJfd2l:.e8 12 0-0 liJd7 13 4je3 (D) .•.


Sakaev - Andreev Sf Petersburg 1996

4..:5'b6 5 �xf6 �xf6 6 c3 �b6 7 e4 g6 8 ttJbd2 i.g7 9 a4 h4 10 lbc4 18b7 1 1 VWc2 d6 12 i.d3 O..() 13 0-0

bxc3 14 bxc3 'Ufic7 15 llJe3 i.a6 16 .txa6 11Jxa6 17 l21c4 (D)


White is better.

13 tbb6 14 .SLbS �d7 15 .1l.xd7 'tWxd7 16 fbdc4 lUxc4 17 tl'Jxc4 (D)

Kasparov - Miles


Ba.l'ie (3) 1986

White is dearly better due to his good knight vs Black's bad bishop.


4 0.e4 5 .th4 \li"a5+ 6 lbbd2 (D) ...




6 .tb7 7 a4 .i.xd5 8 axb5 YHfc7 9 l1a4 'i'b7 10 c4 ctJxd2 11 cxd5 ctJxfl 12 '*'d3 d6 13 e4 ctJd7 14 �xf1 h6 15 �e2 g5 16 J.g3 .i.g7 17 e5 0-0 18 h4 (D) .•.

More speciJ-ically, after 1 (/4 tiJf6 2 4.Jf3 d5 3 c4 e6 4 ttJc3 it.e7 5 cxd5 exd5 6 .tg5 he can play 6... c6! (D).


White has a powerful attack. While I would not claim that this system guarantees White an advan­ tage, it certainly scores very well in practice, and many Benko players pre­ fer to avoid it. If you dislike playing White in the positions which typicaUy arise from the Benko proper, this could be a useful way to play. However, if you decide to play Ihis system, you must understand that the 2 'bf3 move-order gives up a lotofop­ tions in other openings. We have al­ ready seen that you lose the chance to play QGD Exchange systems with your knight on e2. Even if YOll arc willing to play the Exchange QGD with 4.Jt], you cannot really get it via 2 4.Jf3, because the early commitment of the knight to 0 allows Black extra options to develop his queen's bishop.

Now 7 c3 .'tf5 is fine for Black, while after 7 �c2 g61 he can also force ".,t.f5. After the further moves 8 e3 Jif5 9 kd3 iLxd3 10 'S'xd3 4Jhd7 Black has comfortable equality. Thus, by playing 2 liJf3, you practically give up the chance to play the Exchange QGD with any effect. In addition, you also restrict your options against the King's indian. The normal King's Indian move-order is I d4 .:tJf6 2 c4 g6 3 f2Jc3 .\:Lg7 4 e4 d6. In this foml, White has a full range of op­ tions. He can play the main lines with 5 4)0, the Samisch (5 f3), the Four Pawns Atl<lck (5 f4), the Averbakh System (with 5 cte2 and 6 �g5), or less well-known line!; with 5 ltJgc2 or 5 h3. However, after 1 d4 tiJf6 2 t!)f3 While is deprived of all of these op­ tions, except those starting with 5 Itln The Queen's Gambit Accepted is yet another opening in which 2 lbf3


surrenders some options for White. After the normal move-order 1 d4 dS 2 c4 dxc4 White has. in addition to 3

ltJf3. the important alternatives 3 e3 and 3 e4. The latter, especial ly, remains one of White's most dangerous re­ plies. However, if you play the move­ order 1 d4 �f6 2 tt\f3, you need to have another system ready against the QGA. in case your opponent plays 2. . .d5 3 c4 d)tc4.

Another opening that has seen some interesting move-order developments is the Sveshnikov Sicilian. The char" acteristic position of the Svesilnikov appears after I e4 cS 2 lbf3 lbc6 3 d4 cxd4 4 lhxd4 lbf6 5 lLlc3 e5 (this move is called the Pelikan Variation) 6

lbdb5 d6 7 .tg5 a6 8 QJa3 b5 (D).


with 9 Xxf6 gxf6), deeply analysed and amongst the most interesting of contemporary openi ng practice. Because of the high theoreticaJ re­ pute of the Sveshnikov Variation, many players prefer to deviate earlier. One way is to pl ay 7 lbcis, but tbe most popular way to avoid it (together with

a variety of other Sicilian lines at Black's disposal ) is I e4cS 24Jf3 4Je6 3 J:.b5. By contrast with the related system I e4 cS 2 tnf3 d6 3 .ibS+, which bas a reputation of being rather tedious and usually a sign that White wishes only to draw, again:;t 2...4Jc6

the move 3 .tb5 (nicknamed 'The Awry Lopez' by a leading English cor­ respondence player of the I 960s!) has

a considerably superior pedigree. White has a variety of fol1ow-ups, bnt an early exchange on c6 is common, foilowed by d3. The pawn-structure is such that Black's bishop-pair will not easily be activated. while White can often open lines on the lcingsidc with f4. Tn this move-order, 3 ..ib5 is a fa­ vourite of many top players, including Kasparov and Adams. Indeed, EVgCllY Sveshnikov himself is on record as claiming that 3 �bS is White's only try for advantage ! While few players would go so far as that in their praise of 3 ib5. it is

At the cost of ceding the d5-square, Black makes use of his central pawn­ majority to scire control of key central squares. as well as driving White's king's knight back to the a3-square with gain of tempo. The resulting po­ sitions are highly dynamic (particu­ larly if White now exchanges on f6


true that the line is something of a nui­ sance for the would-be Sveshnikov player, since it leads to altogether quieter and less dynamic positions than the Sveshnikov. In order to avoid 3 iLb5 (and a few other possibilities for White in the standard Pelikan move-order), therefore. many players



have begun adopting an alternative move-order: I e4 c5 2 tbD e6 3 d4 cxd4 4 lbxd4 Qlf6 5 ttx:3 liJc6 (D).



years, with Kasparov in particular ad· 'locating the white side. After many trials with 8.. f5 and 8. .'l!Vc7, the cur­ rent trend is for 8, ...th7. which ha<; been the scene of two Kasparov-Gri, shchuk games, as well as severa] games involving Leko as Black. Thus, Sveshnikov Sicilian players have two options. They can play the normal move-order with 2...tDc6, in which case they have to be prepared for 3 i.b5, or they can use the Four Knights move-order. In the latter case, however, they must also be prepared for the 6 iDxc6 linc. Ultimately, it is a matter of taste, although I personally would tend to be less afraid Qf3 i.bS, .

By playing 2...e6 instead ofl.. /i::Jc6, Black has managed to avoid the 3 St.b5 variation. The position reached in the diagram is that of the so-called Four Knights Sicilian. However, after White's main follow-up 6 tbdb5, the Four Knights proper involves 6... .tb4. Instead of this, however, the Sveshni­ kov players continue 6...d6, when af­ ter 7 ..i.f4 e5 8 Sl.g5 a6 9 lLla3 b5 they have reached their intended line. with each side having spent an extra tempo (Sl.f4-g5 for White, ...e6-e5 for Black). As mmal, however, such a move­ order change has its drawbacks as well. Although White has been cheated out of his 7 lbds and 3 .tb5 oplions, the Four Knights move-order does give White other options which he would not have in the nonnal Pelikan move­ order. Chief among these is the contin­ uation 6 1Dxc6 bxc6 7 eS �5 8 liJe4 (D). This has become a major battle­ ground in elite GM events in recent


Making a Virtue of Necessity Although move-order consideratiOn>: may sometimes restrict one's options. they can also be used to one's advan tage to avoid r.pecific variations abolll which one is concerned. The Modcm Benoni is a good example of this: I d·J


tOf6 2 c4 c5 3 dS e6 4 li.:x:3 exdS S (;xdS d6 6 e4 g6 (D).


After its successful adoption by Tal in the 1960s, the Benoni became one of Black's most popular replies to 1 d4, l�ven achieving the ultimale accolade of successful adoption in a world cham­ Ilionship match (Spassky-Fischer, Rey­ kjavik Wch (3) 1972). However, by Ihe early 1980s, it was becoming an l,ndangered species al OM level, al­ mnst entirely because of one line, the dreaded Taiman()v Variation: 7 f4 ,�g7 8 :tbS+ (D).



This disruptive check had been known for years, bul. not considered dangerous after'Ifd7. This was because White usually retreated the bishop to d3 immediately. However, in the early 1980�, White began adopting a different strategy, beginning 9 a4!. In a sense, this is a high-dass waiting move . White knows that he will want to play a4 anyway, so he doe� it at once and preserve� options with the bishop. Although it may retreat to d3 anyway, it sometimes goes to e2 (or c4 or fI), aimiog for a tran�position to a King's Indian Four Pawns Attack with an ex­ tra tempo, if Black should have to re­ turn . his knight to f6. Alternatively, White may answer ...a6 by taking on d7, saving a temJXl for an immediate kingside attack. A particularly savage example of the latter strategy was �een in Kasparov-Nunn, Lucerne OL 1982: 9 ... l2Ja6 10 tDf3 tDb4 1 t 0-0 a6 1 2 ..\txd7+! i.xd7 1 3 f5! 0-0 14 JtgS f6 I S .tf4 gxf5 16 �xd6 .i.xa4 17 llxa4 '\W(xd6 1 S tOh4 fxe4 19 tOfS 'i'd7 20 fue4 Wh8 21 tOxcs 1-0. This is not the place for a detailed theoretical survey of this line, but suf­ fice it to say that despite 20 years of dedicated effort<;, nobody has been able to come up with a line that is trusted by leading OMs. However, rather than ahandon the Modern Ben­ oni entirely, many of its practitioners preferred to employ a little move-order trickery in order to dodge the above line. After 1 d4 ll.lf6 2 c4, they began playing not 2...cS but 2...e6. If White persisted with 3 lbc3, they would ac­ cept defeat and play the Nimzo-Indian



with 3 ....tb4, but if White played ei­

h3 leaves Black's quren's bishop short

ther 3 lbf3 or 3 g3, they would con­

of a good square, which in turn em­

tinue 3 ... cS and head back into the

phasizes the general cramp in his posi­

Benoni, having in the meantime de­

tion. Although White's results have

prived White of the chance to play the

not been so good here as in the f4 +

f4 + .1i.b5+ system. This is the way al­

iLbS+ system, the line is sufficiently

most all of the Benoni's leading prac­

unpleasant for Black to have sent quite

titioners now handle the opening, in­

a few Benoni players in search of a

cluding players such as Psakhis, de

way of avoiding it.

Firmian and Emms. Interestingly, if we continue the

The solution they arrived at was an­ other move-order subtlety. Instead of

story of the Benoni's theoretical de­

the routine 7

velopment into the 1990s, we find an­

play 7 ... a6 (lJ).


kg7, Black can instead

other example of such move-order subtlety coming to Black's rescue. Unable to use the f4 system, players were forced to look for ways to prove an advantage even with their knight o n n. After a few years, they came u p

with the following system: I d4 4.Jf6 2

c4 e6 3 4Jf3 c5 4 d5 exd5 5 cxd5 d6 6

lik3 g67 e4iLg7 8 h3 0-09 .1i.d3 (D).

B Now the natural reaction is 8 a4, af

ter which Black plays 8 ..i..g4, By this .

move-order. Black has pre-empt" 1 1 White's intended h 3 restraining mow (note that the preliminary 7...a6 is n,',' essary, since if immediately 7.. .,ig·1

White has 8 '/lWa4+ tnbd7 9 �2 wilh advantage). If White now plays


iie2, Black has avoided the .ltd3 pl:II' TIlis new plan, with the bishop on

after 9 .....i.xf3, practice suggests BI,I<'k

d3 rather than e2, also proved very

is fine. The sharper alternative 9 'f'I'h

successful fOT White. The extra de­

is also possible, but the eompliealioll"

fence of the e4-pawn limits some of Black's counterplay, while the move


after 9...i.xf3 J O 'Il!V:xb7 are gener"lh

considered OK for Black.


White's other response to 7...a6 is 8 h3, but then Black can expand on the queenside with 8. .b5. In the normal h3 + ..td3 system, Black would not be able to achieve this advance so readily, because White would always answer ...a6 with a4. However, Black's subtle move-order here has prevented White from doing so, and, once again, Black �eems to be fine here. Finally, I should point out that White himself has a way to try In pre-empt Ihis plan. Instead of7 e4, he can play 7 li3, trying to force his way back to the line he wants. However, Black can ex­ ploit the holding back of the white e­ pawn by 7.. .a6 8 a4 \\We7! when, yet again, he has prevented White from lcaching his intended set-up. All of Ihe above lines can he seen in the )'.;unes of John Emms, in particular. The Modem Benoni is not the only "pening whose dubious reputation (":111 be improved by some move-order -.llhtlety. Anolher example can he 'O lund in the Stonewall Dutch. After 11()l vinnik's successes with it in the I') I(ls, the opening entered a long pe­ "od in the doldrums, with Larsen be­ lli)', one of the very few lop OMs who ,'\,<'1" played it. However, during the I 'IXOS, the line achieved a significant 1II<Tl'ase in popularity amongst a new 1" 'Ill'l"iltion of players, including Short, '\ 1I';IIjXlV, Agdestein and others. The 1" lIlIg Kramnik aLso played it with ' ' 'II';idcrabie success. rill' basis ofthe n�w popularity was ,\ ,1I1Tt.'rent placing of Black's king's hl·,lIop. Compare the two positions


1 d4 f5 2 c4 e6 3 ttJf3 ttJf6 4 g3 d5 5 �g2 c6 6 0-0 �,7 (D)



1 d4 f5 2 c4 e6 3 lZJf3 lZJf6 4 g3 d5 5 �g2 <6 6 0-0 Jtd6 (D)


The first variation is the on� us�d by Botvinnik, with the black bishop on e7. However, this eventually fell out of favour, principally because White has the plan (invented by Botvinnik him­ self) of 7 b3 and 8 i.a3, exchanging the dark-squared bishops. In the mod­ ern interpretation, however, the bishop



stands on d6. Not only is it more active on this square, but it also enables Black to meet 7 b3 with 7...�e7!, pre­ venting the immediate exchallge of bishops. White can enforce the ex­ change, but only at some cost. Either he plays X a4, in which case he coo­ cede� the b4-square to Black, or he plays 8 il.b2, 9 'me l and then 10 ��>'L1, losing a tempo. His third option to ex­ change hishops is 8 Si.f4, but in this case the exchange is only achieved al the cost of a weakening of White's king position after 8 .i;:,xt4 9 gxf4. Practice suggests that, with accurale play, Black can achieve equal chances in all three cases. However, if yuu wish to play the Modem Stonewall as Black. it is im­ portant to understand that you can only reach the position with some co­ operation from White. In particular, the Stonewall is generally not to be recommended if White has not al­ ready committed his king's knight to 0 . If he can still play tZJh3, he has 11 number ofextra options. The knight is wcll placed to swing round to d3, while White's other knight can come via d2-fJ, leaving White with the opli� mal position for hoth of his knights. In addition, with the knight on h3. White can offer the exchange of bishops on f4, recapturing with the knight, thereby avoiding the weakening of his king­ side that would result from gxf4. One might think from the foregoing that White should simply stick his knight on h3 the mome,nt Black plays ...fS and ... e6_ However, this would ig­ nore one more subtlety. Although the ..

knight is good on h3 against the Stone­ wall, it is much less good on that square in lines where Black puts his d-pawn on d6, hecause it is then much easier for Black to organize the ad� vance ... e5, when the knight would just look silly on h3. Even ifthe knighl moves to f4 before that happens. it will be hit agaio when Black plays ...e5, and is unlikely to find a decent sljuare 1,0 move to (especially jf Black prefaces ".e5 wilh ... c6, guarding lhe d5-�quarc). So we can clllldude from the fore­ going Illat the Classical Dutch involves some fancy mow-order fCXltWork from both players. In fact, there is a kind of waiting game going on - White does not want to move his king's knight un­ til Black commits his d-pawn, whereas Black doesn't want to show his hand with his d-pawn until White has com­ mitted his king's knight The result of this mutual snspicion can he seen he­ low. After I d4 fS, White docs not know which line Black will play, hul he knows that he will want his bishop on g2 in any case (we are assuming that White plans to play the g3 lines against the Leningrad as well). So, maximum flexibility is preserved by 2 g3 €lf6 3 ,ig2 e6. Now White knows at least that Black is not going to play the Leningrad, hut he still does nol know whetller he is going to play thi' Stonewall or the ...d6 lines. As White wants to keep options of either ttJn or lLlh3, his natural response here is 4 c4. although this gives Black the addi tionnl option of 4....tb4+. Assumin,g


Black is angling for a Modern Stone­ wall, the obvious move is 4.. .d5, but thi� allows 5 41h3. On the other hand, 4 ... J..e7 avoids committing the d-pawn, but now Black can no longer play the Modern Stone­ wall with ....td6. The only way to keep al! options is 4. . .c6, when it is White's tum to tind a non-committal move. He can do so with both 5 '2ld2 and 5 �c2, although neither is ideal if Black continues 5 ...d6. The result of the guessing game is that Black has been Ul1<Ibk to achi eve the Modern Stonewall proper, but he has managed to force White into a sl ightl y sub­ optimal set-lip for the ...d6 lines, in which White would generally prefer to have his yueen's knight on c3 and his qUe<:':n on d l . At move 4, there is one other way for Black to head for a Stonewall . With the move 4.. .SLe7 (D) he agrees to play a Botvinnik-style Stonewall, with his bishop on e7 rather than d6.


plan which is one ofWhite's strongest lines against the Botvinnik set-up. Thu s, after 5 &If:3 0-0 6 0-0 Black de­ lays the ...dS advance in favour of 6...c6. Now if White plays 7 !be3, Black answers 7.. .d5, and the plan of exchanging bishops on a3 is no longer available to White. If instead White persist.'! with the latter plan by 7 b3 there follows 7 ...a5 8 .ta3 �xa3 9 4jxa3 WlJe7 1 O !be2 d6! (D).


Black is now able to force a quick ...e5, when he has good chances. This move-order has heen used many times

by English 1M Robert Bellin, a noted expert on the Dutch Defence.


These examples show how move­

order subtleties can be used in a posi­

tive way, to cnabk the player to reach a desired opening position, without al­ lowing a eel1ain counter-variation.

Compatibilities However, by clever use of move­

order, he can still avoid the b3 + .ia3

Another aspect of move-orders and transpositions is picking openings which naturally fit together. A typical



example is the Caro-Kaon against I

Within the King's Fianchetto com­

e4, and the Slav against 1 d4. By play­

plex, there is a very obvious overlap

ing both of these openings, one is able

between the PirclModem defences and

to answer the English Opening ( I c4)

the King's Indian. In fact, these lines

with 1 ...c6. Now after either 2 e4 or 2

also provide another example nf how

d4, Black can reply 2...J5 and trans­

move-order may be used to avoid a

po�e back into a line which is part of

certain dangerous variatioll. At the

his normal repertoire. It is true that

time of writing, one of White's most

White can insist on a pure English­

dangerous systems against

the Pife

type position hy 2 iLlf3 d5 3 g3, for ex­

Defence is 1 e4 d6 2 d4 ttJf6 3 t;x·3 g6

ample, but such lines are relatively

4 �.g5 (D).

tame and by using the l . . .c6 move­ order, Black is able to cut down on the numherof different lines he has to prepare. Another pair of openings that



wen together for Black are the French and the Classical Dutch. This enables Black to answer 1 d4 with l . ..e6, thereby avoiding the various


ous second-move alternatives open to White after l ...f5,

such as 2 &.g5, 2

QJc3 and 2 e4 (it should be noted, how­ ever, that this option is not open to the player who wants to play the Lenin­ grad Variation of the Dutch, since the

The strength of the line derives

move ...e6 is not un integral part of that

from the facl thai it combines several

line). For the same reaS(lll, the French

plans. White can continue with ¥iHd2

al�o tits quite well with the Nimw

and i.h6, just as in the lines with Sl..e3,

and Queen's Indian complex, since by

but he can also play for an early e5 ad­

playing I d4 e6 one can avoid the

vance, when the position ofthe hishop

Trompowsky - very handy if you hap­

on g5 servcs

pen to play most of your chess in Eng­

h4-d8 diagonal. After 4...i.g7 5 'iikd2,

land, where the Tromp

is remarkably

popular! Of course, one may say that

to create threats on the

5 ...c6 6 f4 b5 7 ii.d3 0-0 8 lZlf3 .ig4 9 0-0 leaves Black with great trouhle

the Tromp is not so fearsome anyway,

equalizing, while after the alternative

so there is no special reason to avoid it,

plan of chasing the bishop by S ...ltJbd7

but on the other hand, why bother hav­

6 0-0-0 h6 7 iH4 g5 8 it..e3 tZlg4 9 h4

ing to learn something against it, when

Black's kingside pawn-structure is very

you can just avoid it without losing

vulnerable. Black has scored very

any options yourself?

badly in both of these lines.


It is clear that most of Black's prob­ lems in this line stem from the fact that his f6-knight is vulnerable to the e5 advance. This leads on to the thought that perhaps Black is better off playing the Modern move-order in this sys­ tem, and analysis seems to bear this out. After I e4 g6 2 d4 kg7 3 ttk:3 d6 4 �.g5 the bishop does not attack any­ thing, and the eS advance will not hit a knight on to. In their outstanding book Pin.· Alert, Alburt and Chemin give two illustrative lines beginning 4...lbd7 5 �d2. one of which shows the hann­ lessnes� of e5 when Black has not played .. .'t'lf6: 5...a6 6 f4 b5 7 e5 and now the thematic riposte 7... f6! gives Black a good game after 8 exf6 4:Jgxf6 9 �d3 c5 (D).

Possibly simpler still is the more di­ rect 5 ... c5 6d..'5 '2'lgf6. Now thatWhite's d-pawn has been lured forward, it is much harder for him to play e5, so Black is able to put his knight on f6 without concerti. After 7 f4 a6 8 a4 'l!kb6 9 11a2 h6 lO.th4 '&'h4 (D) Black was fine in Kmic-Jausa, Sombor 1976.


Thus, we can conclude that the Mod­ ern move-order definitely looks like a good idea for Black if he suspects that his opponent will play the ..tg5 sys­ tem. However, as we know from the foregoing, we must consider what, if any, options Black is giving up by playing this move-order rather than the PifC move-order. He is not himself giving much up, since in most other lines, a timely ...'2'lf6 will trdnspose back into rirc territory. However, the Modem move-order does give White some more options. Firstly, he can play lines with c3, which are difficult to achieve in the rife order, because of the early attack on White's e4-pawn. However, the c3 lines are not terribly dangerous, so this should not be a ma­ jor problem. Much more significant is the fact that in the Modern move-order, White has the option of 3 c4, giving the game a queen's pawn flavour. This is where the compatibility issue comes in. If Black is happy to play the King's Indian, then he need have no fear of White playing 3 c4 against the Modem.



This in tum means that he can pre­ serve maximum flexibility, because he can use both the Pirc and Modem move-orders against I e4, depending on his opponent's preferences. How­ ever, if Black does not have the King's Indian in his repertoire, and does not want to play one of the pure Modern Defence lines against 3 c4, he has a problem. Effectively, it means that he can only play the Modem move-order against I e4 if he is certain his oppo­ nent will not play 3 c4. One leading player who does precisely this is Mikhail Gurevich.

Conclusions I) Move-orders are vital to gOl�1 opening preparation and should new, be ignored. 2) By clever use of move-orders. one can freqnently narrow down one\ opponent's options. This can m�':l1I less for one to study, and can also ell able one to dodge a particularly dall gerous line. 3) Wherever possible, choose opell iogs which tit together, so as to pn· serve the maximum flexibility will, move-orders and transpositions.

6 Use and Abuse of Computers

I hl' Increasing role of computers in , Ia'ss is unquestionably the most sig­ luli,'ant development in the game over til(' past 15 years, and in many ways it lias rlwolutionized the game. The use "I ,Iatabase programs such liS Chess­ II,IS,' and Chess Assistant has now he­ , " llIC standard notnnly atGM level, but 1"\'('11 amongst many club and league 1' 1 ;IYl't's. In addition, playing engines lIil\'<' improved so fast that any aver<lge 1,1;IYl'r can now walk to the shops and I'IIV IIII' a few pounds a program which I', "" strong that he is probably never )',u1W 1o be able to win a game against II l'l lwlly, there is the Internet, which 11,1', sl'l.'l'ded up information transmis­ ',h'lI m greatly that most major ches� 1" III IIOI111ents can now be followed live. \VIIII" Illese developments have their I " " .IIIVC; sides, there are also a number "I III.-;idious ways in which computers 11,'1'(- hl'llught new problems to the I"HIIL' In this chapter. we will look in 1111'''' dl'lai1 at the impact of comput­ . ,', ,lIld 'It how they may be Llsed more . tI,·, IIVt'ly in opening preparation.

No Hiding Place "hll' " " ',h a database program, such as 1 11<·,',II..s�, has the facility to search

out middlegame and endgame pat­ terns, it is in the field of openings that it has its greatest impact. This is seen in a number of ways. In the first place, the existence of huge collection� of tournament games means that a player's opening reper­ toire is now an open secret, to a far greater extent than in the past Prior to the arrival of databases, only a very small number of the master games played in any one year ever saw the light of day. Only the very best games were published in magazines or news­ paper columns, with the remainder ei­ ther relegated to tournament bulletins or not puhlished at aiL Those that made it into bulletin� would nonnally be seen only by professional OMs, who made a serious etfort to follow all the latest theoretical developments. Few could afford to subscribe to all of the relevant publications, so OMs were reduced to paying occasional visits to the central chess library in their home country. In the UK during the 1960s and 1970s, this meant COll­ ing to Londoll for a day and turning up atBob Wade's house in South Lon­ don, to beg use of his chess library for a few hours! Those who lived in a country where no such central facility



existed did not even have that oppor-­ tunity. The situation today could scarcely be more different. The games of virtu­ ally every tournament of significance are now availahle, free of charge, via the Intemel. In most cases, the games can he obtained within a few hours of heing played, and in many prestigious events. they can be followed live via an Internet link. In addition, specific puhlications, such as ChessBase Mag­ azine and the New in Chess Year­ books, provide detailed analysis of many thousands of games every year, plus opening surveys, statistics on the most successful opening variations, etc. The result is that any player who takes part in international opens or strong national events is hound to find many of his games appearing i n data­ hases. As an example, the Bigbaqe 200t datahase produced by ClIessBase contains over 1 20 of my games, de­ spite the fad that I am a pure amateur, who plays only 2-3 international opens per year. The consequence of this is that a player can no longer afford to have significant gaps in his opening reper­ toire, because any prospective oppo­ nent, armed with a database, will be able to locate these prob lem areas within minutes. In previom years, a player who lost a game in it certain variation could count on the fact that the game would not be published and so would not been seen by other future opp()nent�. This meant that he could sometimes get away with playing the same opening, even without a fully

convincing improvement, especially if the variation concerned was not the main line and therefore unlikely to be repeated by anyone who wa" not in the know . Nowadays, only a fool would take such a risk, given the widespread availahility of information.

One Improvement, One Point Only! The flipside of the ahove is that while in bygone days, a strong new move might catch several victims before it became widely known, this is much less likely to happen now. The fol­ lowing tale from the early 197()s is a typical example of what. could hap.. pen to unsuspecting amateurs in the pre-computer era, when they came up against professional GM opposit ion.

The story starts with the game llhl­ mann-Smy"lov, Hastings 197213: 1 c4 e5 2 CDc3 tbc6 3 liJf3 llJf6 4 g3 Xl.b4 5 �,g2 0-0 6 0-0 e4 7 ciJg5 .�xc3 K bxc3 .l4e8 9 n ext] I () 'Dxt3 dS 1 1 cxd5 (D).



Smyslov had previously won from this position after l 1 ...4:JxdS, but here Uhlmann produced the powerful in­ novation 1 2 e4!, which immediately places Black in trouble. The pawn cannot be taken because of 1 3 .QJg5 and 1 4 't!t'h5, so Smyslov was forced to allow White to build a powerful pawn-centre. The result wa� a typical piece of t:hlmann power play, as he swept the ex-world champion off the board: 12 . Q:-\b6 1 J d4 _1it.g4 14 h3 j,hS 1 5 e5 4Jd5 16 �d2 �d7 17 g4 .�,g6 18 'hgS !flaS 1 9 h4 h6 20 h5 hxgS 21 hxg6 fxg6 22 �f3 (22 't&a4!) 22 ...c6 23 j(xg5 4x4 24 '#fh3 �e6 25 :tf2 lIffi 26 iH] lOxc3 27 ,l;tafl klxf3 28 �xf3 lDbS 29 Wg2 lDxd4 30 '/WfR+ 1 -0. Some 18 m()nth� later, at the Nice Olympiad 1974, an unsuspecting ama­ teur from Monaco repeated the white side of thi� variation against the ex­ perienced Romanian OM Ciocaltea. Unfortunately for the white player, in the intervening period Smyslov had demonstrated a large improvement for Black, but despite at least two success­ ful OM outings, the games had not been publishcd anywhere in main­ stream chess magazines. The well­ prepared CiocaJtea of course knew all about them, which made the resulting Olympiad game something of a mi�­ match. Ciocaltea unleashed the im­ provement 1 1 ..:�xd5! (D). In the stem game, While had con­ tinued 12 lbd4, but lo�t quickly after I Z . :�S l 3 tt.)xc6 bxc6 14e3 .ig4 15 'tlVa4 lIe6 1 6 .l'.lbl .teZ 17 1le1? (17 ttf4!) 17...<'bg4 1 8 h3 '5'f5 1 9 I\xe2 �xbl 20 '/iYxg4 'IlWxcl + 21 WhZ l1d8




in SigUljonsson-Smyslov, Reykjavik 1974. Ciocaltea's opponent preferred 12 d4, but to no avail: l2...'UNh5 13 ':bl 'tWaS ( 1 3...4he4 had also favoured Black in Ribli-Vasiukov, Camaguey CapabJanca Memorial 1974) 1 4 e4 liJxe4 1 5 tI.'leS -'txe5 1 6 dxe5 '/j'cS+ 17 <J.>hl CUf2+ 1 8 llxfl �xf2 1 9 .ixe6 .tg4 1 20 'iWgl 'l*'xgl+ 21 Wxgl bxc6 when Black had the better ending and went on to win, TasiC-CiocaJtea, Nice OL 1974. One has to feel sorry for Tasie, be­ cause nowadays an improvement such as 1 1 ...'ffxdS would be all round the world within hours of first being played. In fact, the strength of the movc is such that the whok variation remained under a cloud for almost 15 years, until Kasparov surprised Ivan­ chuk with the new idea I I d4! (in�tead of I I cxd5) and won crushingly at the 1988 USSR Championship: 1 1 ... lDe4 L2 Wlc2 dxc4 ( 1 2....i..fS is �afer, with only a small edge for White) 1 3 llbl f5 (D). 14 g4! fjJe7 15 gxf5 �6 16 c.tJgS! Wi"xe2 17 $LdS+ 'it>h8 18 'iYxe2 �xe2



1 9 .ltt4 lbd8 20 .ltxd6 cxd6 2 1 llbe I Uxe I 22�xel i.d7 23 .'a.e7 .\i,c6 24 f6 1-0 Kasparov-Ivanchuk, USSR Ch (Moscow) 1988.

Information Overload While computer databases make In­ fonnation management a lot quicker, which is to the benefit of the amateur player who has limited time to spend on opening preparation, they also hring with them the problem of infor­ malion overload. The sheer number of games that are now available on chess databa�s means that even a full-time professional has difticully keeping up with the flow of information, The Bigbase 2001 database which I re­ ferred to above contains some 1 1{z mil­ lion games. The free weekly Internet chess magazine The Week in Chess (TWIC) rarely has fewer than 500 games per issue, and in particularly husy weeks of the calendar can have 2000 or more. This means that making efficient use of one's time depends on obscrving some ba�ic ground rules.

Firstly, you should not attempt tn memorize the many new games played in youI' favourite openings. Not only i� it simply impossible to do so, it is no! even particularly beneficial. as J haw repeatedly emphasized in this book. Understanding is what really counts, not �pecifie knowle<lge. If you �pend houn; poring over the thousands 01 new games played every week, YOII are simply wa�ting most of the time in volved. Secondly, be aware of the quality o[ the games you are looking at. One 01 the worst a�pects of the modern chess dataha�e is that it usually contains a huge number of games from very weak events. Few, if any, of these games ar.. going to teach you anything, or lw worth studying, If you wish to fl' search an opening, limit your studil's to high-quality games between GM�, and filter out the rest. This leads 111<" naturally on to the next sedion.

lying with Statistics One of the feature� of chess databa.�..:. which can he very useful if employ,·, [ properly is the statistil.:s and openill,!' report function. At the push of a butt'lil this enahle� you to obtain a picture or" given opening line. its success rail: III practice, the spo.'X-'ific results scored hI' different sub-variations. etc. This C OI l provide an enormous amount of usdnl infonnation. I:'<)r example, we know that in !'Yil cral, the advantago.'! of the first moy, should guarantee White chances 01 small plus. Statistically, White telld' "


to score somewhere around 52-55%, ,lIld Black around 45-48%. This in it­ " elf is valuable information when it mmes to judging an opening. If a hlack defence scores much below [.'i % this would tend to suggest that it IS inferior. Likewise, a white opening that makes less than 50% should be �Icwed with suspicion. Of course, such 'italistics are only a rough first guide, ;lI1d certainly should not be regarded 0I'i 'proof of an opening's merits, but Ilwy do provide a reasonable first indi­ ' 'Ilion. Even if a variation is objec­ tively satisfactory, a below-average percentage score may indicate that the l'(lSition is relatively difficult to play III practice, and this is certainly some­ lillllg the would-be exponenl of the hill' would want to know. However, in approaching the re­ "liltS produced by statistical analysis 110m one's database, there are a few I,nints which must be borne in mind. hrstly, as any statistician will tell you, ,II ways take account of sample size. l 'lll' fact that a certain opening line has ,I 1 1111% score may sound very excit­ III�'" hut if it turns out that your data­ h;ow only contains three games with III<' line, one of which, on do�er in­ " I",d ion, turns out to be a duplicate of " 11<' ()]' the others, this score is unlikely I" i", statistically significant. ''''','condly, quality of games is again " lIl·ial. This is a case where the old , " 111 ]>ulerprinciple of GIGO applies1'.Ilha)2c in, garbage out. If your data­ I\;I�" nmtains a large number of games hv wry weak players, these will dis1",1 Ihc statistics you get. J recently .


encountered a typical example of this, while looking at the following varia­ tion: I e4 e6 2 d4 d5 3 It:k3 dxe4 4

liJxe4 i.e7 (D).



The Ruhinstein French, character­ ized hy 3...dxe4, had been largely out of favour ever since the end of the 19th century, when Tarraschbranded it "the surrender of the centre". Only in the last five years or so has it become pop­ ular again at GM level, as various top players have realized that its solidity has been rathe,r underrated. Almost all of the recent attention has been di­ rected to either 4. . .tfJd7 or 4.. .�d7, but the text-move seems to me to be well worth a closer look. By contrast with 4.. .liJd7, Black intends to recap­ lure on f6 with the bishop. For exam­ ple, after 5 ttJf3 tZ:lf6 6 !{Jxf6+ Jtxf6 7 i..d3 c5 8 dxcS, all of 8 ..tOd7, 8 ...illVc7 and 8...'fea5+ appear perfectly satis­ factory for Black. If Black intends to recapture with the bishop on f6, it makcs sense to play ...ii.e7 immediately, without commit­ ting the queen's knight to d7, since in .



some lines, the knight can develop more actively to c6. The 4 j�e7 line was recommended almost half a cen­ tury ago by Cecil Purdy, the great Aus­ tralian pedagogue and writer, and his writings on the subject have recently been collected and published in book form (albeit under the unfortunate title Action Chess). Looking through the various lines, it seemed to me lhat the most danger­ ous tries for White were those where he tried to preserve the option of queenside castling. This is something which the 4...oie7 line makes rela­ tively difficult, since if White ex­ changes on f6, the resulting pressure on the d4-pawn may prevent him from ' 2. Probably the best move­ playing 1!&e order for White is 5 tDf3 tt:lf6 6 i.d3 lLlxe4 7 iLxe4 tiJd7 8 �e2 (D). ...


With this order, White prepares �e3, reserving (though not actually

committing himself to) the possibil­ ity of castling queenside. Meanwhile, there is also a tactical point to the move, of which more in a moment.

When I looked this position up on ;l database, I was rather shocked to SCt· the statistics. Out of a total of 3X games, Black won 6, drew 1 4 and 10s1 no fewer than 18. The relatively small number of black wins is to be ex·· peeted, since the line is relatively quil'1 and solid, rather than being a dynamit· attempt to seize the initiative. How ever, the large number of losses is cef tainly not what one would expect of;, line whose chief merit is supposed 1(1 be its solidity. It would be easy to at: cept the statistics at face value, an,1 dismiss the line as being bad hu Black, but this would be a serious mis take. Looking more closely at th,' games themselves, I discovered th'll no fewer t han 1 1 of Black's 1 8 losse� resulted from Black falling into theel· ementary trap R...ttJf6??9 i-xb7! .txh7 10 'lii'b5+, when White wins a pawil and destroys the black queenside. This little trap is the tactical point of White 's 8th move. Once one adjusts for these I I games. the statistics show exactly what on,. would expect, i.c. a black score frat' tionally under 50%, with a signifjcanl number of draws and few l osses. A typical example of solid black pla� from the diagram is McKay-Clark,. British Ch (Morecambe) 1 975, whil"i, continued 8 ...c5 () �e3 liJf6 10 j,d ( flic7 1 1 0-0-0 a6 12 dxc5 iLxe5 J \ �.g5 Ji.e7 14 lhe5 !Dds 1 5 bci .: �xe7 'h_'h. The veteran British 1�'1 Peter Clarke was a regular practition,·, ofthe ...!fl.e7 Iine during the 1960s and 1 970s (although he usually prefan'd the movc with 4 <!bd7 and only thell ...


� ...ii.e7; as noted above, I consider Ihis move-order to � less accurate than the immediate 4...ii.e7). As a tough auu wily defender, whose ambitions when Black against a strong opponent r:lrcly exceeded half a point, he was Ideally suited to the Rubinstein French, ;llId in his capahle hands it proved an ....;tremely tough nut to crack. This phenomenon of database sla­ tislics being distorted by large num­ i!t,rs of games by weak players can be wen in almost any opening, especially those that contain a trap. This wa� ,omething 1 realized when writing my previous book, 10J Chess Opening hops. It was really quite amazing at tilllCS to see just bow many victims the: \:Ime trap had claimed. Here is just Llile example, in the Pelikan Sicilian: 1 ,.,] c5 2 tbf3 tbc6 1 d4 cxd4 4 ttJxd4 I.; 11"6 5 Qx3 e5 6 lbdb5 d6 7 4Jd5 1,,; \�d5 8 exdS lOe7 9 c4 (D) (or 9 c3).

I':vcn my relatively small database L oiliains no fewer than 27 examples "I BllIck losing material immediately

\\"1111 9 a617 10 'i'a4!. ..•


Quantity and Quality The lesson from the previous section is that when using a database to re­ search an opening line, one must pay careful attention to the quality of game� on the databaseitself. Although the Internet has made master games much more freely available than in the past, it has also resulted in huge num­ bers of games from very weak events finding their way onto databases. If one thinks about the problem log­ ically, a database is generally used for two, quite distinct purposes. The first is f01' looking up the games of pro­ spective opponents during a tourna­ ment, in order to see what type of player they are, and which openings they play. For this purpose, one ideally wants a database which is as compre­ hensive as possible, so as to maximize the chances of finding one's oppo­ nent's games. In this case, it does not matter whether the database contains vast numbers of games hy weak play­ ers; quantity is everything. The second main use for a data­ base is for researching opening lines. For this purpose, however, one does not really want a huge database, full of games from insignificant amateur events, and played by relatively weak players. The majority of such games will fl()t he of theoretical importance, the results will be unreliable, and the size of the datahase will just serve to make .it more difficult to see the wuod for the trees. Instead, the ideal data­ base for opening research purposes will be small, and comprised only of



games from quality GM events, pref­ erably annotated by the players them­ selves. The solution to this dilemma is to maintain two separate dalabase�, one for each purpose. For looking up one's opponents during a tournament, keep a database which i� as broad and com­ prehensive as possible, regardless of the quality of tbe games. However, for opening research purposes, maintain a second database. consisting of only quality, annotated games. A good starting point would be the CD-ROM of the complete Informator series, per­ haps supplemented by the annotated games from ChessBa�e Magazine. By keeping this second database exclu­ sive, you should have a reliable source of information when researching open­ ings, while your first, catch-all data­ base will give you a wide-ranging collection of games, in which you are likely to be able to find games played by prospective opponents.

Silicon Innovations Thus far in this chapter, we have con­ centrated on datahascs, bul in the last decade, chess-playing engines have also begun to have a major impact on opening preparation. As one would expect. this is particularly the case in sharp tactical variations, wbere the computer excels. In Secrets (!/Pracli­ cal Chess, John Nunn {juoks an exam­ ple where a Sicilian variation wa<; refuted outright by a piece sacrifice found by a computer program. A more recent example of this saw no less a

player than Kramnik fall victim to sili con improvement. Anand-Karpov, Frankfurt rapidplay 1999 opened as follows: I e4 e5 2lbn lbfo 3 ltJxe5 d6 4 liJf3l2Jxe4 5 d4 dS (, .i.d3 4Jc6 7 0-0 iLe7 8 c4 l2Jb4 9 cxd5 liJxd3 10 �xd3 �xd5 II gel �f5 1 2 g4 ,tg6 13 tbc3 l2Jxc3 14 '&xc3 rx 15 .tf4 c6 (DJ.


Anand now played 1 6 11e3 and SOllil obtained a winning p()sition. only I" lose it after an horrendous onc-mo''",­ blunder. Interviewed in Che.nBII.\"'· Maga";ine, he told how he rcturned I" his hotel room with his trainer UIli lava, and began looking at the galli, Reaching the diagram, 'Ubi' ask,"" about the sacrifice on e7, which Analld had felt during the game was no IllOI!' than unclear. Purely out of curio�ll\ they put the position on Fritz. all.1 within a few seconds. the program evuluution of the p()�ition leapl I . . 'winning for White'. Examining Ill' board, Anand soon realized what I " had overlooked during the game v. llit Karpov, and duly concluded thai 1 ( "



sacrifice won out of hand. However,

forced to open the seventh rank for the

firmly convinced that nobody would

white queen. After 1 9 ...f6 20 �xg7

,"ver repeat th� line against him any­

fxg5 21 'ijixh8+ Wc7 22 'IlteS+ �xe5

way, he thought no more about it.

23 dxeS Black was hopelessly lost and

Huwever, twu days later, Kramnik

soon resigned.

olTered to repeat the whole line, having

Examples such as this demonstrate

('vidently expected only 16 l%e3, after

that the growing strength of playing

which he clearly had an improv�mcnt

programs is an increasingly important

prepared. Instead, Anand uncorked 16

factor in preparing sharp tactical vari­

ttxe7!!, whereupon Kramnik plunged

ations. At present, it is mostly just pro­

mto thought, a horrified look on his

fessional players who are likely to

face. On a neighhouring hoard, Kas..

analyse lines in detail with the help of

['arov and Karpov could harely main­

a computer, but one suspects that this

lain concentration on their own game,

will become more common at lower

as they looked up and saw what had

levels in the future. As a result, a lot of

happened. Kasparov, never a great

sharp gambit lines, which rely on sur­

1<ISS to the game ofpoker, looked at the

prise value and perhaps an element of

,kmo monitor, realized what was hap­

bluff. arc likely to come under greater

pening, then grinned broadly at Kar­

pressure. This has already happened

[lov and shook his head in amazement.

in postal chess, where it is dear that

They all realized that Black was to­

computers are being widely used,

lally lost. Kramnik eventually played

whatever the rules of postal compe­

1(}...Wxe7 17 �+ �d8 1 8 �xb7 IlcK (D).

titions may or may not say on the point.

Training Partners For the ordinary player, computer pro­ grams can be extremely llseful as a training partner. One of the main tech­ niques in Soviet training methods was the playing of special training games, either to try out an opening system or to develop certain middlegame and endgame techniques. Botvinnik was especially famous for using training games as an integral part of his prepa­ ration, playing secret games against Now Anand played 19 .tg5+!. This

such opponent� as Ragozin, Aver­


bakh. Flohr. Putlllan and others. The

,luring the Karpov game. Black is

hoary old anecdote about the cigarette


the crucial point that he had



smoke relates to his training games

significance. Having finished with one

with Ragozin. Many modern players also use train­

next game in the book, and repeated

ing games as part of their preparation.

the process

For example, I was once

given a de­

variation, they then moved onto the .

The results of this work were clear

scription of pmtof Grandmaster Yusu­

from the match. Out of 4 games with

pOY's preparation for his Candidates

the 4 e3 Nimzo, Yusupov won

2, and

1991. Yusupov

drew 2. Both of his wins were excel­

decided before this match that with White, he wanted to play 4 e3 against

hefore. Working with his trainer, Mark

lently played, one of them a brilliam;y, and he also missed it forced win in one of the drawn games. Confounding the critics, who had made lvanchnk a clear favourite for the match, Yusupov

Dvoretsky, the two set OUI to develop


match with Ivanchuk in

his opponent's anticipated Nimzo­ Jndian, something he had rarely done

Yusupov's knowledge and understand­


The average player does not gener­

ing of these lines. As a basis for their

ally have access to :1 master-strength

work, they used not an openings en­

trainer, but this technique of playing

cyclopaedia but a games collection by a grandmaster who specializ.ed in 4 d.

trdining games to team an opening can


The player concerned was Gligoric,

be employed with a playing engine such a� Fritz. By starting the game

and the lxxlk they llSed was the afore­

from a predetermined opening posi­

mentioned /grayu profiv Figur, which

tion, one can gain valuable practical

contains a significant number ofgmnes where Gligoric is White against the

experience, which will illcrea�c one's

Nimm. With each game, they first

played through the game and the notes, and analysed the key points . They then played a series of 15-minute games

with the variation concerned, alternat­ ing colours. By switching colours all the time, Yusupov was able to get a more rounded and objective view of

understanding of the opening and l ead to improved results. The computer can also assist in analysing the resulting training games, although its usefulnes� here is generally greater in tactical po­ sitions.


the positions, by seeing them from the

I) Computer databases have revo­

black side as well as the white. After a

lutionized opening preparation at al­

batch of four games, they would SlOp

most an levels of the game.

and analyse the main points of inter­

2) lnfonnation is now much more

e.�t in the games played, before play­

widely available than in the past, and

ing some more games. Finally, they

computers make opening preparation

would check their games against the

much quicker and easier than before.

l atest theory of the line, so as to ensure that they had not missed anything of

overload can be a significant problem.

3) On the other hand, infonnation


Although computer.'; can hold a great deal of information, much of it may be of poor quality, which brings its own problems. 4) Maintain separate databases for game preparation �md opening research.


5) As computer use becomes more

widespread, risky tactical lines be­ come more vulnerable to refutation. 6) Usc your computer as a training partner, if you do not have access to 11 trainer or regular sparring partner.

7 Universalities

One particular approach to opening

natural temptation to try to employ

repertoire managemen t is the use of

your favourite line as While. By doing

universal systems, i.e, openings which

so, you are likely to reach a position of

can be employed both with White and

a type you like and understand, and

Black, or against any particular open­

you can avoid having to learn lo[s of

ing move of the opponent. The usc of

theory on other openings. These prac­

such systems can enable a player to re­

tical points are perfectly valid, and

duce the amount of opening theory he

lllay in themselves provide sufficient

needs to study, and to reach positions

reason to adopt the reverse opening

ofa type he is familiar with and enjoys

approach. However, there i s a more

playing. It is to the pros and cons of

fundamental reason why one may want

this approach that we now tum.

to employ a reverse opening, uamely

Reverse Gears

objective advantage. If the King's In­

One pmticular aspect of the universal

should give some advantage when

opening approach is the use of reveL�C

used by White with an extra tempo?

the helief that it should give you an dian equalizes for Black, then surely it

openings. Typically, this involves tak­

Similarly, if the Sicilian is the best de­

ing a black defence, and trying to play

fence to

it as White, with an extra tempo. An

then shouldn't White stand better

example is the King's Indian Attack,

when he plays it witb an extra move?

I e4 (m; many believe it is),

in which White opens with the stan­

Surprising as it may seem, practice

dard King's Indian Defence moves ]

does not bear out this theory_ Open­

larly, the Dutch Defence player can

and the pure 'reversed Sicilian' ap­

I f4, hoping after 1...d5 or

proach to the English are not regarded

ttlf3, 2 g3, 3 Jtg2, 4 0-0, etc. Simi­ open

ings such as the King's Indian Attack

1 ...lLlf6 toachieve a position similar to

at top level as a serious way to fight for

his favourite Dutch line,

but with an extra tempo. In the English Opening, I

an opening advantage. Alex Yenno�

c4, the reply l. e5 produces a reversed

of another GM's reaction to this issue.

Sicilian, with White again having an

The 'Yerminator' wa.� present when

extra tempo.

somebody asked the Rmsiilll GM


linsky quoted an interesting example

If you have a �trong liking for one

Malaniuk about the Leningrad Dutch.

of these defences as Black, there is a

Malaniuk is the world's leading expen


on the line, and has played it with great success against opponents of all leveh. He was a�ked why he did not play I f4 a� White, with the idea of an­ swering 1 ...dS or J ...liJf6 with a fol­ low-up such as 2 liJf3, 3 g3, etc.. aiming for his favourite variation with an extra move. Malaniuk replied "That extra move's gonna hurt me." l11ere are a numher of rea�ons for this. Firstly, on a purely philosophical level, the simplistic argument that "the Sicilian equalizes for Black, so it must give an advantage with an eltlra tempo", does not hold water. I re­ member many years ago reading an intriguing book entitled My Chess Ad­ ventures, by C.W.Warburton. The au­ thor was a veteran player and organizer from the North of England, who be­ came a particularly strong correspon­ dence player. Brought up on the play of Lasker and others, Warburton re­ mained all his life a fervent believer in the 19th century classical virtues. He always opened I e4 with White, and always answered 1 e4 with 1 . e5 as Hlack. He swore that the Lopez was the best opening for While, and that defending the Lopez was Black's best reply, providing of course that he avoided the "positional error" 3...a6, and instead played 3 ...liJf6 and 4...d6, Ii fa Lasker. From the lofty heights of .� I st century theory, it is easy to scorn sllch an attitude, but Warburton upheld II is principles in many brilliant COITe­ \llImdence games during the 1950s dud 1960s, against strong opponents. However one may view Warburton's ,·,"Centricity, he gave lli1 interesting ..


retort to those who argued that the English should be good for White, be­ cause it is a Sicilian with a move in hand. His response was to tum the ar­ gument on its head. Many years of master practice have demonstrated that l ..e5 is a perfectly good answer to 1 c4, and en�ures Black equal chances. How much worse must Black's posi­ tion be, he argued, in the Sicilian De­ fence, whcn after I e4 c5 White has a reversed English with a move in handI The t11lth, of course, is that neither of these arguments is correct. Black does not stand markedly worse after ] e4 c5, but neither does White stand significantly better after I c4 e5. The main practical reason for thi� is that while and black openings generally have different aims, and when they are taken out of their natural hahitat, they often fail to adapt well to their new sur­ roundings. Thus White, having the first move, is generally trying to develop some initiative and obtain what we commonly refer to as an opening ad­ vantage. The opening lines he chooses are adapted to that purpose. Black, meanwhile, is aware that he slarts the game at a slight disadvantage, and that his main aim is to neutralize White's initiative. His play therefore tends to be geared to doing that, rather than to striving to seize the himself. The effect of this on reverse open­ ings can be seen by comparing the fol­ lowing main-line sequences: 1 e4 c5 2 lbf3 d6 3 d4 cxd4 4 QJxd4 .

lbf6 5 i.'Dc3 g6 6 J..e3 .1i.g7 7 (3 0-0 8 1Wd2 0)c6 9 Jl.c4 J..d7 10 0-0-0 ),lc8 11 �b3 1ile5 (D)



the second). In the first sequence, White, seeking an advantage t.o justify his first move, has adopted a highly aggressive set-up, castling queenside and preparing a kingside pawn-storm with g4 and 114. In the second exam­ ple, however, Black has reacted to the same Dragon set-up by adopting an altogether more restrained develop­ ment, castling kingside and intending to play more in the centre. In this line, Black's subsequent ... f6 move is usu­ ally played merely to defend the e5pawn, and not with any intention of

1 c4 e5 2 g3 qJf6 3 ig2 dS 4 cxdS 4:lxdS 5 lLlc3 tLlb6 6 .!iJf3 4Jc6 7 0-0 1J.e7 8 d3 0-0 9 a3 iLe6 (J))

supporting a . .g5 pawn thrust. In sh0l1, .

the character of the two positions is completely different, and theory con­ siders both positions to give White at most his normal orening initiative. This example encapsulates the main practical reason why adopting a re­


verse opening does not usually bring any advantage. It is because players tend to handle the position much Ie..'>;; ambitiously as Blad:. than they do when faced with the same position as White. If Black were to try to defend the reversed Dragon line by adopting the same set-up with

.too and ...�d7


that White uscs against the Dragon In the first. example, we have


proper. White's extra tempo would in­ deed prove significant, but against


�tandard Sicilian Dragon, and in the

less ambitious formalion, White is un­

second, a line of the English Opening

able to make much of it. Similarly,

known as the Reversed Dragon. Com­

there is absolutely nothing to stop

paring the two position�, whit:h reprc­

White meding the Dragon with SLe2.

�ent the most common main Jine of


each opening, one cannot but be struck

fectly respectable line of play. How­

by the difference in the set-up adopted

ever, it is rarely seen at GM level,

0-0, etc.; indeed, tbis is a per­

by the two 'anti-Dragon' players (i.e.

because it i s not considered aggres­

While in the first example, Black in

sive enough to fight for a significant


advantage, and most players are seek­ ing such an advantage when White. I learned this lesson in my youth, after playing the following variations: 1 d4 dS 2 c4 e6 3 ttJc3 c5 4 cxdS exd5 5 lZJf3 lZk6 6 g3 lZJt'6 ? �g2 i,e?

8 0-0 0-0 (/))


1 d4 liJf6 2 c4 c5 3 e3 g6 4 tbf3 i.g7 S lZJc3 0-0 6 ..tc2 cxd4 ? exd4 dS 8 O-OIild; (/))


The fint variation is the main line of the Tarrasch Defence to the QGD,


which was a favourite of several club­ mates and myself. Liking the Tarrasch so much (the follies of youth are sometimes hard to explain!), and re­ luctant to play the white side of the Benko Gambit, we hit upon the idea of playing the second �equence as White, thereby reaching a Tarrasch with an extra move. The line had the added merit that it could also be reached against the Griinfeld, via the sequence I d4 lhf6 2 c4 g6 3ltx:3 d5 4 4Jf3 iLg7 5 e3 0-0 6 i.e2 c5 7 0-0 cxd4 g exd4 4Jc6. Several of our club members played the line enthusiastically in the 1970s, and once we discovered that the leading player at 1:1 neighbouring club was also an addict, we even jok­ ingly christened it ''The Mid-Kent Variation", after the part of England in which we all lived. I wil l return to the last diagram be­ low, but for the moment I would just like to note one thing. My friends and I found that when we played the Tar­ rasch proper as Black, reaching the penultimate diagram, our opponents almost all played either 9 ..tg5 or 9 dxc5. As White in the Mid-Kent Vari­ ation, however, if we continued 9l:Xe1 , we found that very few opponent" played either 9... .tg4 or 9. .. dxc4. In­ stead, most players prcfened 9....iLf5, a quieter line which is known in the Tar­ rasch proper, yet hardly ever played. It was clear that when faced with the Tarrasch position with colours re­ versed, most players adopted an alto­ gether less ambitious approach than they would have done when White against the Tarrasch proper.



Information is Power In addition to the practical points listed above, there is another, more ob­ jective, reason why black openings are rarely as effective as one might expect when played with colours reversed and an extra tempo, Tbis is because, in the profound words of Grandmaster Suba, "Black's information is always greater by one move". Although the tirst move confers a (�ertain initiative on White, it abo means that White is always committing himself that tiny bit earlier than his opponent. Every move is to some extenL a commitment, and each time Black comes to make his move he bas the benefit tbat White has already made one extra move, thus revealing his hand that tiny bit more. As an initial and extremely basic example, consider the opening move I e4. Without Black having made a sin­ ,

gle move, he already knows certain things about the likely shape of the game. He knows, for example, that White is more likely to attack on the kingside than the queenside. Of course, a queenside attack cannot be ruled out, but if this is what White intends, lIe is far more likely to open I d4 or I c4; once he has played 1 e4, it is already more difficult to develop an initiative on the queenside. Similarly, Black knows that a subsequent c4 by White is also less likely than bef()fc, since it would leave a hole on d4. Thirdly, he knows that he is unlikely to have prob­ lems on the long h l-a8 diagonal, since White bas already obstructed it with a pawn and so is unlikely to fianchetto

his king's bishop with any effect. Let us assume Black replies L..e5 and

White continues 2 Qjf3. Once again, Black has learnt something. He knows that a quick attack on the f-file is now unlikely, since the white f-pawn has been blocked. He may even seek to take immediate advantage of this hy playing 2. . .f.5 himself, something he would perllaps have been less likely to do after other second moves by White. After 2. ./bc6 3 Jib5 Black knows that a quick attack on f7 is less likely than before, since a later .i.c4 by White would involve the loss of a tempo. This in turn means that Blat.:k t.:an later play a move such as ...Qjf6 without having to worry about the conse­ quences of lOgS or .i.xt7+, etc. On the other hand, Black also knows that Ilis e5-pawn is likely to be the main target for the next few moves, and this must influence how he develops. The above was a very basic example of how the information game works, but the same phenomenon exists in all openings, and goes a long way tn­ wards explaining why a black defence does not always work so well when converted into a white opening with an extra move. Larsen once gave 11 practi­

cal example of this. At a certain point in his youth, he became hooked on the NajdorfSicilian, I e4 c5 2lOf3 d6 3 d4 cxd4 4 4Jxd4 l/)f6 5 41c3 a6. So keen was he on the opening that he started playing as White I c4 e5 2 d3 tUf6 3 a3 "hoping for 3...d5 4 cxd5li1xd5 5 tiJn �6 6 e4, the dear variation with all extra tempo" Unfortunately, after a few successes at junior level, he SO()II



realized that "amongst masters, you

a) 9...i.g4 is probably the most

cannot persuade anyone to play 3 ... dS".

common plan. It aims to force White

Behind the rueful humour, however,

to clarify the central position, due to

lies a serious point, which illustrates

the threat of ...i.xf3. White's usual re­

very well the Suba dictum. In the

ply would be IO cxdS, followed by h3,

Najdorfproper, Black plays 5...a6 only

and then l1el, �f1, etc.

after White has already opened the

h) An aJtemative plan for Black is

centre with :3 d4. In order to get his Najdorf with colours rcversed, how­

9 ...dxc4, with the idea to �xc4 .!baS.

ever, White has to t:ommit himself to

e2, allowing 1 1 ....te6 (controlling the

Now the bishop must either retreat to

a3 before Black has played ...dS.

important dS-square), or to h3, when

Armed with the knowledge that White

Black can capture and secure the

is now well-prepared for 3...d5, Black

bishop-pair. White can react to this ei­

can simply avoid playing it, thus leav­

ther by

ing White with a pawn move which is

answering with the aggressive gambit

ceding the bishop-pair, or by

not of much real use in other struc­

10 d5 ttJa5 11 iLf4, with some initia­


tive for the pawn.

My clubmates and I encountered a similar problem in

our beloved Mid­


other moves are possi­

ble, these are the two main plans.

Kent Variation, discussed above. In

Moving on to consider the Mid-Kent

the diagram position below, we had to

position, we can draw certain conclu­

decide how best to utilize White's

sions. Firstly, if we know that Black

cxtra tempo.

wants to adopt the .. i.g4 plan, we can .

prevent it completely by 9 h3. Alterna­ tively, we can play 9 I:lel, with Ihe idea of meeting 9 ...ibg4 with

1 0 cxd5

QJxd5 1 1 h3 .Jie6 1 2 ..tfl , when White has a main-line Tarrasch position with an extra tempo - his lle I is a useful move in this position. On the other hand, if we know that Black wants to play

9...dxc4 followed by l O.. lLla5, .

there is something to be said for sim­ ply 9 a3. allowing the bishop to retreat safely to a2. The problem for White, however, is that he has to move, and so must com­ If we assume that itis Black's move

mit himself first. If he plays 9 a3. he is

III Ihe diagram position (as it would be III Ihc Tarrasch proper), the two main 1 'I;II1S are as follows:

advertising the fact that he is ready for the reply 9. .. dxc4, and it is unlikely that Black will be so nai've as to play



this move. Instead, he is likely to switch to 9... i.g4, in which case the move a3 is of little value to White. Al­ ternatively, if White plays 9 h3 to stop 9....tg4, Black can play 9...dxc4, and the h3 move is relatively ineffective in this line. After various experiments, my ciubmates and I settled on 9 lZel as the best move. As we have seen, this works well against 9....1g4, while against the 9 ...Jxc4 plan, we intended the play the gambit line with 10 dS. This is another line where llel fre­ quently comes in u�ful, pressurizing the black e7-pawn, which has been rendered backward after 1 0 dS. How­ ever, as narrated above, we found very few opponents willing to enter either variation. Most answered 9 l!e J with either 9 ....i.e6 or 9 ...iifS, when it is much harder to demonstrate the value of the rook on e L Indeed, against 9...J..fS, it can even tum out to be a negative factor, since in some posi­ tions Black's .. .ibb4 threatens a fork on C21

Universal Remedies Moving on from the subject of reverse openings, we come now to that of uni­ versal systems, i.e. opening lines which can be employed regardless of the op­ ponent's defence. For the present, I will look at systems that only apply from one side of the board, rather than openings that can be used with either colour. The great advantage of playing a line which can be used against almost any reply is that one reduces very

substantially the amount of ground one has to cover in opening prepara­ tion. There are several white openings that meet this criterion, but interest­ ingly, none of them starts I e4. The main examples are the English Open­ ing with 2 g3, the Collcn'orre/London Systems and the Verewv Opening. The first of the.<>c, the English Open­ ing, has become very popular in recent years, chiefly because of its advocacy by Tony Kosten in his b()ok The Dy­ namic EnMlish. At first sight, one may be forgiven for thinking that to de­ scribe I c4 and 2 g3 as a 'dynamic' opening is a case for the Trades De­ scriptions Act, but in fact Kosten makes quite a good case for his varia­ tions. The thing which enables the opening to be described as a universal system is the use of2 g3 as White's al­ most invariable follow-up move, rather than alternatives such as 2 liJc3 or 2 lbf3. There are not many normal black responses to 1 c4 that cannot be met satisfactorily by 2 g3, and the avoid­ ance of an early �3 also enables White to avoid various defences, such as the following (see diagram,� on jf)l" lowing page):

1 c4 e5 2 4:k3 Q\f6 3 4:Jf3 4Jc6 4 g3

>t.b4 ... and ...

I c4 t2Jf6 2 liJc3 c5 3 liJf3 e6 4 g3 b6 5 .tg2 .th7 The first line is a popular system af­ ter 1 ...c5, and was tried out more than once in the Kasparov-Karpov malches in the 1980s. The plan of ...I:i.b4 and a later ... �xc3 has proved very solid for Black over many years. In the Kosten



meet a laterd4 advance by exchanging


on d4 and setting up a row of pawns on his third rank, by ...d6, .. .a6, ....'ike7, .. .0-0, ., etc. This flexible set­ up contains the seeds of a venomous counterattack, should White overreach himself in advancing his kingside pawns (those who are unconvinced of the Hedgehog's inestimable merits arc referred to the game Vaganian-Gid­ dins, Antwerp 1996!). Once again, however, the Kosten move-order pre­ vents Black from achieving the Hedge­ hog set-up, because White's early fianchetto interferes with Black's . .. b6 move; e.g., 1 c4 illf6 2 g3 c5 3 �g2 e6 4 'bf3 and now if Black persists in try­ ing to play the Hedgehog, he runs into trouble ,Liong the hI-aS and a4-c8 di­ agonals: 4.. .b6? 5 'DeS d5 6 fbc3 .i.b7 7 �a4+!.

Of course, there are many satisfac­ tory ways for Black to answer the 2 g3

move-order, however, Black cannot achieve this, because the white knights are not committed so early. Thus, after I c4 e5 2 g3 t{Jf6 3 i.g2 lbc6 4 t1Jc3 i,b4 White can answer with the im­ mediate 5 ltJd5, avoiding an exchange on d. By contrast with the first dia­ gram above, his hng's knight ha� not heen committed to t3. which enables him to follow up with c3 and tbe2. This seems good enough for a small advantage. The �econd diagram is the popular Hedgehog Variation. Black intends to

move-order, despite the fact that cer­ tain of his options have been elimi­ nated. My personal favourite is the continuation I c4 eJf6 2 g3 c6 3 ..tg2 d5 (D).




My use ofthis was inspired by wit­ nessing the game S.Webb-Petrosiall, Hastings 1977/8, in which the ex­ world champion destroyed his solid English 1M opponent in little over two hours' play: 4cxd5?! cxd5 5 lOf31Oc6 6 O-O?! e5 7 d4 e4 8 lbe5 i.d6 9 i..f4 0-0 1 0 lbc3 neg 1 1 lbxc6 bxc6 12 i.xd61Wxd6 1 3 Ikl h51 14 tfd2 h4 1 5 ,*,g5'!h3 16i.hl 't'tHb4 17�d211b8 1 8 b3 lOg4! 1 9 W1f4 Uh7 (D).

chances. After trying out 4...g6 a few times with satisfactory result'>, I would now prefer 4... ..ig4. Kosten describes this as a very dangerous line for Black, and it is indeed true that Black must tread carefully. After his recommended 5 l£Je5, the natural 5 ...�h5?! is an error: 6 cxd5 cxd5'! (6... ltJxd5 is safer, but still somewhat better for White) 7 \'tffa4+ ctJbd7 8 ltJc3 e6 9 g4 i.g6 10 h4 (D) and Black is in deep trouble.


20 lbxe4 (already desperation - 20 e3 is met hy 20 ....+We7!, planning ...g5) 20...dxe4 2 1 it.xe4 iLd7 22 ':'c5 .nb5! 23 S:i.xc6 1:!.xc5 24 Sl.xd7 �b7! 25 f3 't!:Vxd7 26 dxc5 CDe3 0- 1 . I still recall the shell-shocked expression on Webh's face as he walked around the tourna­ ment room after the game! Of course, White can play the posi­ tion much better than Wehh did. Kos­ ten recommends the interesting pawn sacrifice 4 ltJf3. Although I am scep­ tical as 10 the true objective merits of White's JXlsition after 4...dxc4, there is no doubt that in au over-the-board situation, White has good practical

Kosten now considers ] O... .t.d6 and 1 O... i.c2 (the latter refuted hy the tac­ tical JXlint I I lbxf7!), hut he does not mention Black's other try, W...bS!? which was played as far back a� Kon­ stantinopolsky-Goglidze, Leningrad 1936. This spectacular game continued 1 1 QJxh5! ( I I 'ID'xhS allow,q Black to play 1 1 ....tab8 1 2 'lli·c6 Iib6 J 3 '@'a4 nh4, ele.) 1 l ...41xe5 1 2 lDe7++ We7 I3 liJxa8 '{Wxa8 1 4 h5 4Jex.g4 15 hxg6 fxg6 1 6 d4, when White had won the ex.change. After further complications, he won as follows: 1 6. <.t>n 1 7 i.f4 .JI.e7 1 8 11c1 .!iJhS 19 i.d2 �8 20 ..th3 lbhf6 2 1 'iWb3 liJxf2! 22 Wxf2 ..


4le4+ 23 '*'g2 Q]xd2 24 �xb8 lhb8 25 Irc6 i.f6 26 i.xe6+ rl;e7 27 h3 i.xd4 28 lIdl �e3?! 29 i..xd5 J..g5 30 e3 1-0. However, Black can improve sig­ nificantly on all of this, by answering 5 .!be5 with 5 .i.e6! (D). ..


Although not mentioned by Kosten, this was the move recommended as best hy Konstantinopolsky at the time of the above game, and it is also rec­ ommended by NCO. Although the move Il'Ioks unnaturaJ, it avoids expos­ ing the bishop to further attack by a g4 advance and Black can always de­ velop his .kingside by ...g6 and ..iLg7. NCO, for example, now gives 6 cxd5 iLxd5 7 liJf3 c5 8 .!Dc3 i.c6, when Black is fine. Another set of openings which White may use with relative impunity against almost any defence is the QP complex, which can mean any of the Colle, London or Torre systems. All are characterized by 1 d4 and 2 4:Jf3, with the branch then depending on White's third move: 3 e3 is the Colle .


System, 3 .tf4 the London System and 3 J..g5 the Torrc Attack. These have always enj oyed a fair bit of popn­ larity at club level in the UK, since they are all solid, easy to learn and en­ able White to avoid nasty theoretical surprises. In general, these line..;; do not seek to achieve a significant advan­ tage from the opening, but aim instead to shift the weight of the struggle to the middlegame. The fin;t two have not been popular at master level since the early part of the 20th century, but the Torre has enjoyed periodic patron­ age at higher levels, by playen; such as Petrosian, Miles, Yusupov and others. As we will see in Chapter 9, the Eng­ lish GM Mark Hebden has turned the Torre into quite a formidable weapon in recent years. The problem many players have with the above lines is that they tend to be very slow and positional, which is not to everybody's taste. The club­ level player who would like to have a universal easy-to-Iearn opening but finds fiancbettoed bishops and c3-d4e3 pawn-centres a bit too tedious, has a couple of other options. Both �tart I d4 4:Jf6 (or L.d5) 2 tbc3. Now most players will continue 2...d5 (D) (or 2...llJf6 if they started with 1 ...d5), af­ ter which White has two options. The gamilileers amongst you can try 3 e4, the infamous Blackmar-Die­ mer Gambit. Unless Black is willing to transpose into a Classical French by 3 .. e6, he is virtually forced to capture the pawn. After 3 ... dxe4 (3 ...liJxe4 is also fine), whereupon White contin­ ues in gambit fashion with 4 f3. The .



look. White's main strategic threat is to capture on f6. doubling Black's pawns, and then to play d, i.d3, 'iWf3, 'lJge2, etc. Often White can castle queenside

and launch a kingside pawn-storm. Probably the soundest and most popu­

lar defence is 3 . 4:lbd7, preserving the ..

integrily of Black's pawn-strudure. White then has several plan�. 4 n, aiming to advance in the centre with e4, is the silarpesl linc, but positionally

discussion groups. Given the ahun­ dance of available material, I will not

rather clumsy and not really to be rec­ ommended. The line preferred by Ver­ esov himself wa� 4 lUt 3, foll(lwed by e3, �d3, 0-0 and an eventual e4. Tony Miles played a number of games with this in the 1980s, with good results. However, for those looking to get even further from the beaten track, there is the intriguing move 4 f4!?

cover it in any detail here except to say

This rather bizarre-looking move is

that the Blackmar-Diemer is similar to

an idea of A.M.Stewart, several-times

not 100% sound, but

White can get the key position against

B r itish cOlTe:;pondence champion. The idea is to obtain a Stonewall set-up, with hi:; traditionally bad bishop de­ veloped outside the pawn-chain. After 4...e6, Stewart's patent i:; the sneaky­

90% of opponent" is a significant ad­

looking 5 a3 (D).

Blackmar-Diemer is a line which has achieved virtual cult status, both in OTB and postal chesH, being the sub­ ject not only of books, but also of dedi­ caleu theme tournaments, websites and


many gambit';


extremely dangerous in the hands of a

well-prepared and imaginative attack­

ing player. In addition, the fact that

vantage over many other gamhit lines,

which depend on Black's cooperation. The alternative to the Biackmar­ Diemer after I d4 'lJf6 2 fbc3 d5 is 3 �g5, the Veresov Opening. This en­ joyed a brief spell of populari ty in the UK in the mid-1970s, following publi­ cation of a long article by 1M Robert Bellin in Chess magazine. Unfortu­ nately, it was soon overtaken by its near-relative, the Trompowsky, hut it seems to me that the Veresov has its own merits, and may be worth a closer



The point of this is to prewnt Black from playing ...i.b4 and taking on c3. Although such a manoeuvre would


lOf6 2 g3 g6 3 b4. My suspicions that I b4 might be more than just a Basman­ esque joke were first aroused when

relinquish the bishop-pair, Stonewall

looking through some old Soviet mag­

stmctures tend tu favour knights, and

azines from the 1 950s and 1 960s, in

the removal of the c3-knight, an im­

the library of the splendid Max Euwe

portant defender of White's weakened

Centre in Amsterdam. In one of them I

e4-square, would more than outweigh

came across a highly impressive win

the surrender of the hishop-pair. After

hy Sokol sky himself, from a USSR

5 a3, the follow-up would probably he

correspondence championship:

e3, .i.d3, and either �f3 or VWf3. Al­

Sokolsky Andreev USSR con: Ch /960-1

though White's 4th and 5th moves


look rather artificial, StewaJ1 has won numerous games with this, many of them against strong correspondence players, so the plan should not be un­ derestimated. All in all


I believe the

Veresov (with or without the Stewart

1 b4 e6 2 .1b2 �r6 3 bS d5 4 e3 i.d6 5 4J1'3 0-0 6 c4 c5 7 .teZ 4Jbd7 8 0-0 b6 9 a4 -Ifl.b . 7 10 a5 dxc4?! 11 fDa3 lIc8 12 4Jxc4 i.b8 (D)

plan) is a good choice for the club and league player who is looking

for an

easy-to-learn way of avoiding his op­ ponent's theoretical knowledge. It is sounder than the Blackmar-Dicmer,


but sharper than the CollefLondon complex, and also flexible enough to permit different interpretations. As a final example of a 'universal' opening which may merit sume con­ sideration, I would like briefly to con­ sider the Sokol sky I b4. Unlike the ,

lines considered above, the Sokolsky is usually relegated firmly to the junk­ box section of openings, but I believe this to be unjustified. If handled in the right way, it is really not so different

Now White made an excellent posi­ tional decision: 13 d3 �e7 14 e4!

from many English and Reti systems.

White has a clear edge and Sokol­

I;or example, Reti himself used to play

sky went on to win a positional mas­

J 1Of3 �f6 2 c4 g6 3 b4 (including his



f:lmous victory over Capablanca at New York 1924), while Smyslov has

I subsequently located many more

often played the related system 1 1Of3

Sokolsky games with this opening,



and soon realized that there is much more to it than I had appreciated be­ fore_ Rather than being merely a piece of tactical trickery, the Sokolsky has a sound positional hasis. As in many lines of the Engl ish and Reti, White's basic strategy is an early space gain on the queenside. However, unlike those systems, White's king's hishop usually ends up on c2, mther than g2, si nce from e2 it defends the b5-pawn and supports the c4 advance. In many lines, White rcaches a structure similar to a reversed French Defence. Here are a few examples of how play can develop:

Sokolsky \#)rld con:



The factors mentioned in the previ­

ous note give White grounds to hope for a small plus in the long term.


Ch semi-final 1958·60

Sokolsky - Strugach Byelorussian Ch 1958

1 b4 e5 2 i.b2 d6 3 c4 fS 4 e3 lllf6 5 tOn i.e7 6d4 e4 7 itJrd2 d5 8 b5 c6 9 a40-0 10 �b3 ..te6 11 fLlc3liJbd7 12 .1i.e2

1 b4 e5 2 .i..b2 f6 (D) The most ambitious, but relatively rare in practice.

White has an edge. Sokolsky



8yelorussian Ch 1959

1 b4 eS 2 Ab2 i.xb4 3 �xeS In this exchange variation, White's hopes are normal ly connected with his central pawn-majority and half-open b- and c-files. His development lag means that he needs to be careful in the initial stages, but if he can emerge from the opening without suffering any misbaps, he shoul d have the better long-teon chances.

3,.,itJf6 4 c4 0-0 5 e3 lUc6 6 .i.b2 d5 7 cxd5 tOxdS 8 itJf3 .i.g4 9 .ie2 l!c8 10 0-0 (D)

3 e4?! As the present game shows, thi s gambit is bags of fun for White, but it is probably not full y sound. White



should prefer the positional 3 b5, fol­ lowed by e3, c4, tDf3, d4, etc.

3...�xb4 4 i.c4 lJlc6 5 f4 exf4!? 6 ibh3 liJge7?! 7 tiJxf4 liJa5 (DJ


White has a clear advantage.

8 iLxr6!? lUs 9 .!bh5!? .!bxc4 10 lUxg7+ �f7 11 0-0 �g8 12 �h5! :xf6 1311xf6 tng6 14 gxg6! hxg6 15 1!t'xg6 WhS? The critical defence is 1 5 ...lbe5! 1 6

tt'g3 �h7!.

16 ttJeS! '6e7 17 llJf6 1-0 Sokolsky

Pelts Mimk 196! -

It is clear from even these few ex­ amples that the Sokolsky is a perfectly sound opening, in which there is much scope for creativity and many chances to obtain interesting, original positions. Given its surprise value as well, it is well wmth looking at ifyou are seeking an opening in which you can avoid op­ ponents' theory and to a considerable extent dictate the layout of the game.

Is Black OK? Thus far, we have considered univer­

1 b4 lUf6 2 i.b2 g6 3 c4 i.g7 4 e3 d6 5 d4 0-0 6liJf3 lLJbd7 7 iLe2 ncs S lbc3 e5 9 a4 The start of an interesting plan. The simple 9 0-0 is also playable.

9...c6 10 as 'Be7 11 b5 a6 12 b6! *"d8 13 d5 c5 14 'iWb3 Now that White has blocked the queenside, he intends to castle there and attack Black's king.

14..,h6 15 h3 <'lJh7 16 g4 f5?! 17 gxfS gxfS 18 0-0-0 (D)

sal openings only from White's point of view. Not sUllJrisingly, it is rather more difficult for Black to adopt a uni versal system, since White's first move allows him much morc scope to push Black around. Nevertheless, Black ­

does have a number of possibjlitie�.

The most popular one is the Pirc/M (k I em complex, which I will disnrs� ill the next section Another npl iOll I'l( Black, which I shall consider hell', ( 1 . 'bc6 against anything .







It is not unfair to say that for most of the 1980s and 1990s. L..lbc6 was one of the great undiscovered secrets of opening theory. It was hardly played or taken seriously by anyone, despite the fact that Tony Miles had been ter­ rorizing the OM circuit with it for years. lbe following massacre illus­ trates the damage Tony was able to do with it:

GeruseJ - Miles Porz 19X2 1 d4 "bc6 2 dS tL\es 3 f4 'Lig6 4 e4 e6 5 'Lif3?! exd5 6 exdS �cS (D)

Already White'� position is a sham­ bles. Unable to see any way of com­ pleting normal development, he now lashes out in desperation, but the retri­ bution is not long in coming.

12 h3lhr6 13 g4liJe4 14 !lh2 lLlh4 15 lDg5 lDxgS 16 t"xgS �e7 17 'ifYg3 (D) Miles now finally annexes material with a move sufficiently photogenic a� to deserve another diagram:


White's decision to treat the open­ ing like a left-handed A1ekhine is al­ ready looking suspect, as his king is prevented from castling kingside.

7 'ii'd3 d6 8 iLe2 'Lif6 9 lbc3 0-0 10 i.d2 White is now finally ready to evac­ uate his king from the centre, but at the last moment, Miles strikes out to pre­ vent this. to_. ttJg4! 11 tDdl ne8 (D) .! Personally, I would have been em· barrassed enough to resign here, but



Gerusel struggled on for a few more

Of course, Black can now play 2 ..e5,


transposing into nonnal

18 ttf2 iLxf2+ 19 ti'xt2 1lWe4 20 ttJe3 '&hl+ 21 .to 4Jf3+ 22 �dl It'lxgS 0-1 I recall myself once suffering a sim­


1 e4 c5 main

lines (with some of White's options cut out, e.g. the King's Gambit, Vi­

enna, dC.), but the player who chooses 1 ,".'�k6 to avoid main-line theory will want an independent alternativc. There

ilarly drastic defeat in a blitz tourna­

are three possibilities.

ment in Moscow, as White against 1M Nikolai Vlasov, another 1 .. .itJc6 devo­

2...e6, which after 3 d4 d5 transposes

Nimzowitsch's preference was for

tee. Thankfully, I can't remember the

into an unusual kind of French De­

exact course of the game, but I know it started 1 d4 ltJc6 2 c4 e5 3 d5 !bee7 4

fencc. Although the line is just about playable, I cannot recommend it, as

many moves later, when something

obstructs the thematic French freeing

landed on g4 and the f2-square col­

advance ...c5.

e4 li:lg6 5 the3 1.c5, and ended not

the position of the black knight on c6

Option two is the move 2 . d5, which


. .

The hasic idea of this system is to play [ .}:.6 and 2. . e5. Thus, the typi­ .

cal main lines go I e4 lbc6 2 d4 e5 and 1 d4 <'Dc6 2 c4 eS. It should be noted

will transpose into a variation of the Centre Counter (or Scandinavian De­ fence, as it is more generally known nowadays). This is a favourite of the

that in the fonner line, Nirnzowitseh

aforementioned 1M Vlasov, who has

himself used to prefer 2 . ,d5, but this

published extensive analysis of the lines following 3 exd5 �xd5 4 lbc3 'ID"a5 5 �b5 .Jtd7 6 0-0 0-0-0 (D),


leads to a different type of game, and does not enjoy a very good thcoretical reputation nowadays. By contrast, the positions after 2. e5 seem perfectly I e4 lhc6 2 ..

OK for Black. Indeed, after

d4 e5, it may well he that White's best try for an advantage is to transpose into a Scotch with 3 121f3. Over the past de­


cade or so, the Scotch has hecome a serious alternative to the Ruy Lopez, even at the highest leveL Nonetheless, there is no reason why Black should not he able to equalize, providing he is well prepared, and the possibility of lransposition into a Scotch should not worry the would-be L.lbc6 player. Ironically, the most critical chal­ 1 e4 �6 is probably 2 tUf3.

lenge to

These lines are extremely sharp, and while it is difficult to trust the



hlack position entirely, it is likely to

Again, the most accurate, pre-cmpt­

prove effective in the hands of a well­

ing Black's plan of playing ...dS him­

prepared player. For those looking for


7...liJe7 8 .tb5+!

a sharp tactical fight, with plenty of opportunities to seize the initiative as

Emphasizing the drawhack of the

Black, these lines can be recommended

queen's bishop's absence from the

for further study.

black queenside.

However, the most popular option is 2...d6. This was the line preferred by

8...c6 9 dxc6 bxc6 10 :ta4 'V¥ic7 1 1 '&e2 �d7 1 2 g4 ,ig6 13 0-0-0

Tony Miles, and one with which he

It is clear that White has a signifi­

won numerous games. However, he

cant advantage, with more space, hetter

mel his Waterloo in 1995, in the fol­

development and the �afer king. He

lowing game:

went on to win an attractive sacrificial game:

lIIescas - Miles Linares Z J 995

13...eS 14 'Wc4 lIe8 15 <i)h4 llJb6 16 .b:b6 axb6 17 lid3! dS (D) 17 ...b5 loses to 1 8 liJxb5 cxb5 1 9

1 �4 QJc6 2 liJf3 d6 3 d4 lbf6 4 lbc3

�xb5+ lbc6 20 ttJxg6 hxg6 ll


1i.g4 5 1i.e3! This is the start of White's most ac­ curate plan. Until this game, theory had generally recommended 5 i.b5 as strongest, but Miles demonstrated in a


number of games that Black's position is satisfactory after 5 ...a6 6 i.xc6+ bxc6, Ligterink-Miles, Wijk aan Zee 1984 being one dra<;tic example.

5,. e6 6 h3 �h5 7 d5! (D) .


18 exdS! .ixd3 19 �xd3 r:td8 20 IIdl g<i 21 d6 illi6+ 22 111bl 1iIb8 23 4:Je4 bS 24 llJf6+ <t>f8 25 dxe7+ <3;xe7 26 fhd7! bxa4 27 .!iJfS+ gxfS 28 �a3+ 'Wt>e6 29 gxfS+ <io>xfS 30 �f3+ 1-0 After this rather crushjng defeat, Miles's enthusiasm for 2...d6 waned rapidly. After a few, largely unsuc­ cessful, experiments with 5 ...a6 (so as


to prevent the check on b5) he switched to answering 2 �f3 with 2... eS. Play­ ers who wish to continue using 2...d6 clearly need to find some improve­ ment here. Illescas himself suggestoo 9...QJxc6, with only a small advantage for White. Another try is 7 ... exdS 8 exdS fbes, when White's most critical line is probably 9 g4 �g6 10 lZxi2, planning to drive Black back with f4. However, after I O..:*Ve7 1 1 Si.g2 hS! 12 g5 QJh7 1 3 0-0 J..fS (D) the posi­ tion was fairly unclear in de la Riva­ Spraggett, Santiago 1995.

White eventually won that game, but not because of the opening. This is another avenue that 2...d6 exponents may wish to explore further.

Colour-Blindness The final universal system I wish to discuss is the most universal of all, in that it can be played with both col­ ours, as well as against viltually any opposing set-up. I refer to the King's InmanlPirclModem complex. These


lines are characterized by a kingside fianchetto, with such moves as ...g6, ... i.g7, ...d6, ..."bf6, etc., or their equivalents as White. The lack of early central contacL is what makes these systems so universal, because it means that there is little chance of vio­ lent early contact knocking the player out of his preferred scheme. With Black, the principal move­ order decision against 1 e4 is whether to play the Modem (1 .. g6) or the Pirc (1...d6 2 d4 "bf6 3 lbc3 g6) move­ order. As mentioned earlier, the deci­ sion depends mainly on how Black in­ tends to play against a set-up with c4. If Black is happy to defend a King's Indian proper, he can play 1 . ..g6. Thus, after 1 e4 g6 2 d4 �g7 3 c4 d6 4 lik3, the move 4 ...liJf6 transposes into the King's Indian proper. Alternatively, Black has other options involving delaying ...liJf6, such as 4...lbc6, 4...0.d7 and 4...e.5. HoweVer, at the time of writing, none of these non-KID options looks terribly good from a theoretical perspective, although certain Modern Defence die-hards, such as Azmai­ parashvili, continue to play them. If Black is not happy with these lines, he needs to play the Pirc move­ order, only committing himself to a kingside fianchetto after White has blocked his c-pawn with lbc3. On the face of it, this might seem to mean that he no longer has a universal defence as Black, since after 1 d4 d6 White is not forced to play 2 e4, but can prefer 2 c4 or 2 Q:\f3. Now 2...g6 would allow While to force Black back into lines he wishes to avoid (e.g. 2 ttlf3 g6 3 (4).



However, in recent years, an alterna­ tive repertoire has evolved based on 2 c4 eS and 2 t'iJf3 Jt..g4:

I d4 d6 2 c4e5 (D)

king will find a secure spot on c7 (af­ ter the pawn advance ...c6), and his de­ velopment is comfortable. In practice, Black has a healthy plus score from the position after 3 dxe5 dxeS 41/fexdlH ,.wxd8, and even after allowing for the fact that in the majority of games Black is the stronger player, it is clear that Black's chances are fully equal at least. Since the advance 3 dS abo hrings nothing (Black can develop in severnl ways, 3.3S followed by 4 ..QJf6 being one of the simplest), White's main Iry from the penultimate diagram is 3 lDf3. Play then continues 3 ...e4 4 QJgS fS, which, once again, has proved very rohust in practice. Both sides have a lot offkxibility in how they de­ velop, but Black's basic plan is gener­ ally to play ...QJf6, ...0-0, ..c(:i, ....QJa6-c7, etc. Alternatively, he can advance ...c5, trying to induce the re­ ply d5, which would give him the c5square for his pieces, The. resulting positions offergoodchances for Black. In the second diagram, Black will answer 3 e4 with 3...<'Df6 4 CDc3 e6, followed by ... Ji.e7, ...0-0 and ...dS. His intention is to aim for a French­ style pawn-structure, but with his light-squared bishop developed out­ side the pawn-chain. Black's results here have been very respectable. A more critical plan for White is 3 c4, hut here he must reckon on the ex­ change, damaging his pawn­ structure. Mickey Adams is one top player who has defended these lines successfully as Black. All in all, the 1 ...d6 repertoire seems to offer many .

1 d4 d6 2 1ilf3 �g4 (D)



One important step in the develop­ ment of this repertoire was the realiza­ tion that, in the first diagram, Black has nothing to fear in the ending after the exchanges on eS and dB. By con­ trast to the Berlin Defence ending we discussed in Chapter 3, here the black




interesting possibilities for Black,

of the same policy, and the same

while remaining less explored than

drawbacks apply. If you exclusively

most main lines

Universal Problems

1 g3 as White and 1 ...g6 as Black, you are going to restrict your practi­ cal experience to a relatively nan-ow range of positions, and in the long run this will not be good for the overall de­ velopment of your chess unde rstand ing A� a result, young and ambitious players should think seriously before adopting such an approach Although it can bring undoubted practical bene­ fits in the short term, it can prove a limiting factor later in a player's de­ velopment.

In the foregoing, we have discu��ed




As White, there is also some move­ order choice. The standard King's In­ dian Atta<:k usua!ly starts with 1 tbfJ, followed by 2 g3, 3 kg2, etc How­ .

ever, While can aLso .�tart with

1 g3,

which offers slightly more flexibility and scope for originality. The English

GM David Norwood is an example of uses I g3 and 1 ... g6 ex­

a player who clusively.

the advantages of a universal opening

approa<:h - less to study, the ability to reach positions and structures one is familiar with, etc. For the amateur player especially, these are significant benefits. However, there is one great disadvantage to such an approach, and that is the limiting of one's under­ standing to just a small group of posi tions. We discussed the same problem in Chapter 2, in the context of concen­ trating on a very nan-ow opening rep­ ertoire. Using a universal system is really jmt a more extreme application ­




J ) Universal systems can save a player a great deal of time and prepa­ ration by reducing the number of dif­

ferent openings he needs LO study. 2) Playing 'black' openings as White with a move in hand rarely promises a significant advantage, de­ spite the extm tempo.

3) The main drawback of universal systems is that a player restricts the range of positions and structures that

he plays, which in tum can limit his development as a player.

8 I nfidelity and Divorce

Hitherto, we have con�idcred issues surrounding the formation of an open­ ing repertoire, I now want to discuss situations where a player deviates from his normal repertoire, either for a onc-off game, or as a more permanent change in his repertoire.

"I Didn't Play My Usual Line Because... " The first situation to consider is the temporary deviation from one'� nor­ mal repertoire, in a �pecific, one-off game. This may be for a number of reasons, which can be either positive or negative, One of the most common reasons to depart from one';; normal repertoire is a desire to avoid the opponent's prepa­ ration. This may be a fear that the op­ ponent has prepared a specitk line for that actual game, perhaps because he knows your play and has seen you play a certain line before. Maybe you have even played the line against the same opponent previously, and there­ fore believe that he has prepared an improvement. Altern[ltlvely, you may just have a general fear of a certain op­ ponent's theoretical knowledge - hav­ ing. seen the opponent at other chess evcn!S, you know that he is a player

who knows his choRen lines very well, or you know that he always plays a particularly unpleasant line [Ig[linst your favourite opening. For one or more of the� reasons, you decide to depmt from your normal line and play something he will not expect. To [I certain extent, the degree of success you can expect from such switches depends on how broad your own repertoire is. If you have a very narrow repertoire, and do not know very much about openings other than those you play regularly, it is going to be difficult to adjust to playing some­ thing new. On the other hand, if you are accustomed to playing a wide vari­ ety of systems, it is easier to avoid a certain line, without ending up in something totally unfamiliar. In general, just as J believe that most amateur players should stick to a relatively narrow repertoire, so I also believe that it is generally a mistake to departfrom one's normal lines just be­ calise of fear of an opponent's prepa­ ration. In this belief, I am supported by Yermolinsky, who had the following to say on this subject in his book Jhe Road to Chess Improve­ ment: ..... no player should ever aban­ don his openings out of fear of the opponent's superior preparation. Often I see my students going into inferior


lines because 'Oh, I didn't want to play a normal line, I've seen him play this stuff so many times ... '. To reveaJ a little secret, I study theoretical lines in the hope that my opponents will avoid them! Nothing makes a GM happier than when his less experienced oppo­ nent gets 'creative' from the very first moves. Don't make that mistake. If you think your openings are good, play them against anyone, e�pecially grandmasters!" I have myself fallen into the trap he describes, with disastrous results. In the 1990 British Championship, I lost a bad game against Dave Coleman in the line 1 e4 e6 2 d4 d5 3 ttJc3 i.b4 4 c5 c5 5 a3 �xc3+ 6 bxc3 CiJe7 7 �g4 VfHc7 8 il.d3. This sharp move, little known at the time, was a favourite of my opponent, whereas I had never met it before. The least sharp reply is 8 ...c4, but biUer experience has taught me (and continues to teach me) that this move is almost always wrong in the Winawer, so I headed for compli­ cations: 8 . cxd4 9 fbe2 .!bbc6 10 'i!!!xg7 11gB I t �h6!? (D). .




This last move, instead of the more obvious 1 1 '1i'xh7, came as a complete surprise. J subsequently discovered that it had becn awarded an exclama­ tion mark by no less an authority than Uhlmann. White's idea is to take on h7 with the bishop, forcing the black rook to make an important decision. Taken by surprise, I played too passively, fell into a very difficult position, and eventually lost Although unpleasant, sllch a loss is to some extent forgivable. Falling into a little-known, highly sharp variation, of which one's opponent happens to have made a speciality, is something which can happen to anybody (it cer­ tainly could in those pre--computer days). What was not forgivable, how­ ever, was the sequel. Despite having analysed the line and prepared some improvements, when I found myself facing the same opponent a few months later in the Hastings Challengers, I lacked the contidence to repeat the opening. Instead, 1 made a spur-of­ thc-moment decision to play some­ thing else, and chose L.eS. All I suc­ ceeded in doing was jumping from a warm frying pan into an extremely hot fire, as my opponent rattled out a sharp line of the Goring Gambit, 1 e4 e5 2 t.iJf3 �6 3 d4 exd4 4 d, another of his favourites. Knowing even less about this than I had previously known about the 8 .id3 Winawer, I was soon in the toils and lost in 25 moves. It is nice to be able, for once at least, to say that r learnt from my mistake. In the Gausdal Troll Masters 1994, I had the following opening as Black



against Swedish GM and theoretical expert, Thomas 'The Hit-man' Ernst: I e4 e/l 2 d4 d5 3 lbc3 jib4 4 e5 cS 5 a3 �xd+ 6 bxc3 !iJe7 7 '/i'g4 0-0 8 i.d3 fS 9 exf6 ':xf6 10 .tg5 un 1 1 �hS g6 12 �dl llJbc6 1 3 .1bf3 '/WaS 14 i.d2 c4 IS .te2 COfS (D),

In the post-mortem, we started ana­ lysing the alternative reply 16'hliJd6. After 1 7 tbg5 �g7 1 8 h5 h6 Ernst de­ clared that the piece sacritice 19 hxg6 (D) yielded White a winning attack.


Thus far, we were following my pre-game preparation, but in this po�j­ tion, instead of the usual 1 6 0-0, Ernst produced the new move 16 h4, the �ort of direct approach that his pugili�tic nickname would lead one to expect. After due thought, I decided to follow the classic recipe of answering a wing attack with a counter in the centre, and continued l 6...Jtd7 17 h5 eS but after the further moves 1 8 hxg6 hxg6 t 9 dxeS :s.e8 20 g4 liJg7 21 COgS! I was unable to justify the loss of material. Despite some desperate attempts to land a haymaker punch, I duly went down after 2 1 ....:Lfe7 22 'Z'lh7 nxeS 23 l[)f6+ wn 24 'Z'l"e8 hg4 j 25 f3 COfS 26 lLld6+ �g8 27 'Z'lxf5 ,i"f3 28 llJg3 i.xhl 29 "Llxhl �d8 30 �f2 .ttfS+ 3 J �g2 1-0.

I was scepticaJ, but after being mated a few times in the analysis, I soon gave up and went off to the bar, to drown my sorrows with a few bottles of shockingly over-priced Norwegian beer. However, by this time I was in my 'Russian period', and rather than aban­ don the whole opening as a forced loss (which would probably have been my reaction a few years earHer!), I duly �pent a few of those cold Moscow winter evenings having a closer look at the whole 16 h4 line. I soon came to the conclusion that Ernst's proposed piece sacrifice was unsound, and that 16...lbd6 was a g(X)(/ answer to his new move. As luck would have it, some four months later, at the next Gausdal tout­ nament, the first round draw threw up the pairing Ernst-Giddins. Rather than



doing what I would have done in the

triumph ot" faith in one's opening, de­

'bad old days , and hastily improvis­

spite having suffered a bad defeat


ing a new defence to 1 e4, I instead

As a tinal footnote to this episode, I

chose to call Em�t's bluft lt was clear

should point out that the wonderful

to me that he would not play the piece

magnum opus Franziisisch Winawer. Bund 1 by Kindermann and Dirr,

sacrifice, but even so, I wanted to see what he would do instead. Attt'T just a

published in 2001, mentions 16 h4 as

couple of minutes' play, the position

<In interesting but untried move; evi­

after 16 h4 was OIl the board, and 1 un­

dently, my 'Rumbles in the Jungle'

corked 16 ...tbd6. To my astonishment,

with the Hit-man have not reached

after only ,I few minutes' thought


most databases! To some extent, how­

Ernst confidently rattled out 17 4lg5

ever, the theoretical significance of the

the further moves 1 9...hxg5 20 j�d3

line is not so great, since 14..:�c7! is nowadays regarded as superior to my

cxd3 21 fBh5 <i.>f8 22 W'xg5 We8 (D)

1 4...c4.

llg7 1 8 h5 h6 1 9 hxg6. However, after

the 'Hit-man' began to consume vast quantities of time, as he realized that the attack was not succeeding.

Playing for Results Another reason (read 'excuse') players give for departing from their normal repertoire is that they were deliher­ ately playing for a certai!) result, and wanted to choose an appropriate open­ ing. For example, a player needs to win with Black, in order to win the tournament or fulfil a title norm. Con­ cerned Ihat his opponent might play for a draw, he ahandons his normal French Defence, fearing the Exchange Variation, and instead plays the Sicil­ ian, which hc has never played before. Alternatively, a normal!y sharp attack­

After 23 1:.h8+ �d7 24 %th7 tbe8

ing player only nceds a draw from a

25 'ifh6 lIe7 26 g7 ttJxg7 27 lIxg7

certain game. Instead of playing his

dxc2 28 §,C I 'iit"xa3 29l'hc2 b6 it was

normal game, he adopts some kind of

clear that Black had succeeded in

quiet �et-up, hoping to avoid risk. In

shielding himself on the ropes and

both cases, the result can easily be di­

warding off all White's best punches,


and even I was able to wrap up the

Even at

GM level,

there have been

game with the extra piece. A triumph

many classic examples of players fall­

of preparation, certainly, but also a

ing into thi� trap. One of the most



dramatic in recent years occurred in the final round of the 1990 Interzonal in Manila. Going into the final round, Mikhail Gurevich was on +3, a point

13 �g3 'Zld7 14 'Zld2 i1lf6 15 \!In c6 16 'tiVb3 '§'b6 17 '6'xb6 axb6 (D)

off the lead. Having led for most uflhe tournament, his tie-break was such that a draw with White against Short in the final round would put the Russian GM into the Candidales cycle. Short, on the other hand, had a poor tie-break, and needed to win if he were to reach the Candidates. The game opened thus:

M. Gurevich - Short Manila IZ JI)9{)

I d4 e6 2 e4'? The question mark is not so mueh due to my reverence for the French, but instead reflects a very bad psycho­ logical decision by Gurevich. A COtl­ finned 1 d4 player, whose theoretical knowledge was (and still is) one ofilis greatest strengths, he would in any other circumstances have played 2 c4, and taken on a main-line queen's pawn opening. Knowing that he necds only a draw, he allows temptation to over­ take him, and instead heads for an Ex­ change French. His position never recovers.

2...d5 3 exd5 exd5 4 .!iJfJ Ji.g4 5 b3 i.b5 6 .ac2 .i.d6 7 QJe5!? Another move he would probably not have played in nonnal circum­ stances. Although objectively, Black has no mure than a rather dead equal­ ity, the psychological initiative is aJ­ ready with him.

7...i.xe2 8 �xe2 liJe7 9 0-0 0·0 10 .�f4 1:.e8 1 1 '*Yg4.i.xe5 12 i..xe5 .!iJg6

The endgame offers Black an edge. Although White's hishop cannot be described as 'bad' in the conventional sense, it is actually less effective than Rlack's knight, despite the relatively open position. In addition, the half­ open a-file means that White must constantly won)' about a black pawn advance ...bS-b4, breaking in down the a-file. Although While's position is objectively tenable, Black's cndur­ ing pressure means that it is highly un­ pleasant to play in practice, and a dispirited Gmevich is slowly pushed off the board. The remaining moves must have been agony for him.

18 a3 lbe4 19 ctJxe4 nxe4 20 .ti.fdl b5 21 0,11 f6 22 '3 Iie6 23 Iie1 IiIf7 24 11xe6 �xe6 25 .t:te1+ ct>d7 26 'life2 h5 27 cJr>d3 h4 28 J.h2 !.be? 29 �f4 .!iJf5 30 Ji,d2 b631 ':e2 c5 32 J..e3 b4 33 axb4 c4+ 34 '.tc3 lLld6 35 l:r.el 11a4 36 �d2 1hb4 37 nal ':'xb2 38 Ita?... >i'e6 39 .uxg7b5 40 ..tf2 b441 <t>cl d 42 i.xb4 .!iJf5 0-1


Gurevich's misfortune had remark­ able consequences for the chess world. As a result of this game, Short reached the Candidates, went on to win it, and then participated in the breakaway from FIDE, which has resulted in to· day's shambolic world championship situation. Just think, if Gurevich had played 2 c4, we might still have a worthwhile world championship cycle ! The late Soviet grandma.�ter Lev Polugaev,�ky provided some highly in­ structive advice on how to approach the opening phase in a decisive game, Once again, the scene was the last round of an Interzonal tournament, this time at Petropolis, Brazil, in 1973. In order to have any chance of making the Candidates, Polugaevsky needed to beat Portisch with White. The man himself now takes up the story of the discussion with his second (Bagirov), and his eventual decision: First I Iuui to decide the question: should I pia.)! that whkh I normally play, or should I try In surprise my op­ ponent with my ('hoice of openinx? ... we begun con.�idering opening with J e4. In its favour, apart from its sur­ prise value, was the fact that after J e4, Portisch Jeels much less confi­ dent... "But if it should be a Lopez, what then?" I asked dubiously. "Play the Italian Game!" "But I never played it, even a.� a child! " "So much the better! Portisch only plays the variation with .. .J.d." Arui J was shown a multitude ofvari­ ations of primordial antiquity, which


had been worked out taking Portisch 's games into account. J was ready to agree, when sud­ denly J sensed: this i.� no Wily to play! Thi.\' i.r rwt the way to pian a decisive battle. Ajter all, ifr were tofail to gain an advantage fmm the opening, I would neverforgive myselffo r having betrayed 'my sort' of chess, and this would inevitably tell on my condition during the game. (Grandmaster Preparation, p. 194) Instead. Polugaevsky opened I lLlf3, secured an advantage from the open­ ing, and went on to win a fine techni­ cal game. in his best style. On this same theme, I recall an in­ teresting exchange with my trainer, when L was in Moscow. Facing a game where a win with Black would put me in the prize list of a tournament, I made the classic mistake of playing a dubious gambit that I had not played hefore. I emerged from the opening with a bad position, and was lucky to draw. Back in Moscow, Igor quizzed me about my choice of opening, and I explained my reasoning. "Ah", he replied, "So in other words, you think it will be easier to beat someone from a rotten position you know nothing about, than from a decent one that you've played umpteen times before?" (he then went on to add that "every Russian schoolboy knows better...", but you'd probabJy guessed that bit anyway...). There is one other factor which makes it dangerous to switch open­ ings in a crucial game. I have repeat­ edly emphasized in this book that



understanding the typical middlegame structures that an opening leads to is the single most important factor in playing an opening successfully. If you adopt an opening you have rarely played before, you are almost inevita­ bly going to finish up in a position you arc not familiar with. Even if your position is objectively good, your la<.:k of familiarity with the structure is prohably going to detract from your ahility to play it properly. This phe­ nomenon was illustrated very nicely in the lOth and tinal game of the Shot1-Karpov Candidates Semi-Final match in 1992. Going into this crucial game, Kar­ pov trailed 5-4, and thus had to win with Black in order to save the match and force a tie-break. It is difficult for any player to win with Black to order against a world-class opponent, but for Karpov it is even more so, since his normal opening repertoire is not very well suited to the task. Karpov has al­ ways believed in playing primarily for solid equality with Black, rather than doing anything very ambitious. His normal responses to 1 e4 were 1 ...c6 and 1 ...e5. However, in the match with Short, these lines had brought him nothing hut trouble - from two games with cach, he had amasscd a grand to.­ tal of just half a point. Faced with a must-win situation in the final game, it was not so surprising that he chose to switch to the Sicilian, even though he has played the opening on only velY few occasions in his career, and then usually against significantly weaker players.




Linares Ct (10) 1992

1 e4 cS 2 liJf3! The exclamation mark is not for the objective merits of White's second move. but for an excellent psycholog­ ical decision. Needing only a draw, there must have been a great tempta­ tion on Short's part to adopt a quiet line, such as 2 c3. However, at that time in his career, Short had rarely played such lines since his youth, and instead always took on the main lines of the Sicilian (with great success, it should be added). Rather than back down now, Short avoids the mistake that Gurcvich had made against him (see above) and sticks to his usual ap­ proach. 2 l1k6 3 d4 cxd4 4 4lxd4 lbf6 S lbe3 d6 6 .tgS e6 7 '@Id2 iLe7 8 0-0-0 .••

0-0 9 ctJb3 a6 10 it.xf6 gxf6 11 h4 'iith8 12 g4 bS 13 gS b4 The tirst new move. 1 3 ...l1g8 hav­ ing been played previously_

14 ctJa4 :g8 15 f4 �b8 1 6 <iJ'bl .ars 17 i.c2 (D)


Looking at the position objectively, it is dear that Karpov has achieved all that he could reasonably have wished for. In a must-win situation as Black, he has reached an extremely sharp and unbalanced position, in which Black's chances are reasonable. However, the problem he faces is that while tbe position may objectively be OK for Black, it is of a type that Karpov had hardly ever played in his lite before, and in which he must have felt like a tish out of water. The effect of this showed in his next move.


Black's extra pawn is of little rele­ vance.

21 Q\b6lbe7 (n)


17 eS?! ...

This act of larceny betmys Kar­ pov's lack of experience on the hlack side of the Sicilian, and the move caused uproar amongst the watching OMs in the Press Room. In particular, the excitable Ljubojevic went quite berserk, ex:claiming to all and sundry "Karpov may understand chess, but he does not understand the Sicilian De­ fence!" The extent of Ljubo's criti­ cism may be somewhat exaggerated, but there is no denying that 1 7...e5 is a pour move, after which White stands clearly better. Short instead recom­ mended 1 7 ...'i!fc7. 1 8 f5 fxgS 19 hxgS l:l.xgS 19 ...'Wxg5 is no better in view of 20 �d5.

20 �e3 �f6 Karpov subsequently suggested the alternative 20 h6, but Short counters with 2 1 :lc4 6tf6 22 lLlb6 !De7 23 ltJaS. It is clear from this and similar variations that Black's 17th move ha" allowed the white minor pieces to be­ come very active, while in this position ...

In this position, Short now regained his pawn with 22 <tJxc8?, but instead the immediate 22 J..xa6! would have been virtually winning for White. Af­ ter a �ubsequent further error by Shorl, Black obtained serious counterpJay, and only went down in a dramatic time-scramble. This tense and dramatic game illus­ trates two things. Firstly, the problem faced by a player who switches to an opening with which he is unfamiliar. However good his specific prepara­ tion, and however objectively reason­ able the position he obtains, he i� likely to tind himself playing a posi­ tion and structure in which he has little or no experience. In such a case, eVl'n a� great a player as Karpov can find il difficult to handle the position \1'<" 1 1 The second thing the game iliusll.>!' is the rewards which Clln ,",.111,· I , , ,, , , being courageous when 11<"<"0 11111" " "I a draw, and not hein�� ;]1<;11< 1 I" 1" I I . · .




one's normal lilles and accept a sharp position. Short's play in this game is an excellent example of this.

Heading for the Divorce Courts Thus far in this chapter, we have con­ sidered mcrely temporary departures from one's repertoire, in a one-off game. I now want to consider the more fundamental decision to change onc's repertoire on a permanent basis, and to abandon a particular opening or varia­ tion for good. As you will have gathered from the foregoing chapters of this book, I am in general quite hostile to the idea of changing one's opening repertoire in a fundamental way. Just as in other walks of life, divorce can be a messy and expensive business, and in my opinion, it is rarely justified in the context of chess openings. The mllin cause of my hostility is the impact that changing openings has on one's un­ derstanding of an opening. I have repeatedly stressed through­ out this book that understanding typical positions is the single most important factor in successful use of an opening. Such understanding is first and fore­ most the product of experience. and must he built up over a period of years, by playing and analysing the opening. It follows from this that by changing openings completely, one is abandon­ ing the experience and understanding that one has built up in the line con­ cerned, and thereby putting oneself

back to square one in this respect. In the great majority of cascs, the change will backfire. I know a player in England who is a particularly drastic example of the ef·· fects of this constant switching of openings. The player concerned, rated around 2000. was a friend of mine in the 1990s. in the 8 years or so that we wcre playing tournaments together, his white opening repertoire changed five times. Hc started ofT playing 1 e4 (Bishop's Opening and 2 f4 Sicilian). After a couple of tournaments in which his results with these openings were disappointing, he switched to the Colle System. One year on, he decided that this was too quiet. and that the path to triumph lay in going back to I e4, only this time taking on main-line Sicilians. After six months of sitting up until 4 a.m. each morning, studying Nunn's Beating the Sicilian, he played one international open, scored 0/3 on the white side of the Sicilian, and de­ cided that he should instead play Torre systems (1 d4, 2 lbf3 aod " .tg5). A couple of years later, it was the Eng­ lish Opening with 2 g3 that became the secret of all succe�s. Not surpris­ ingly, over the period he scored about 30% with White, and even that owed much to his tenacity in defending lost opening causes. Yet, over the same pe­ riod, a� Black he never varied from the French against I e4 and the Slav against I d4, and his score as Black av­ eraged dose to 70%! The moral of the story could scarcely be clearer. At a much higher level, the top Rus­ sian GM Dreev underwent a similarly


painful transitional ordeal in the mid1990s. For most of his adult chess ca­ reer. Dreev stuck faithfully to the French Defence, playing lines with 3 lbc3 Lbf6 and 3 .!bd2 c5 4 exd5 '§'xd5. In general, his results were extremely successful, but somewhere around J 997, he became dissatisfied with the French (in pm1icular, the 4 .�xd5 Tarrasch line came under serious the­ oretical pressure) and began to look for another defence to I e4. At first, he chose the Classical Sicilian, but it was very clear that he was like a fish Ollt of water in such positions. A few brief examples will demonstrate this:


After this further mistake, his posi­ tion is critical. 13 0-0-0+ <3lc7 14 ..tbS! (D)






Linares 1997

1 e4 cS Z Lbf3 <'tk6 3 d4 cxd4 4 Qlxd4 Ci.ir6 s lbc3 d6 6 g3 g6 7 �g2 .liLg7?! (V)

Already a significant inaccuracy. Corrr--ct is 7.. CDxd4. .


S �xc6 bxc6 9 eS dxe5 10'iVxdS+! 'itxdS II i.xe6 1IbS 12 .ie3 Uxb2?

14 I1M 15 a3 .llxbS 16 lLlxbS+ �b7 17 tbxa7 •••

Black is an exchange down for no adequate compensation. Dreev duly lost the ending.




LifUlres 1997

1 e4 eS Z liJf3 ibc6 3 d4 cxd4 4 Q'\xd4 iLlf6 S �3 d6 6 JigS e6 7 �d2 a6 8 0-0-0 �d7 9 f3 h6 10 i,e3 "fie7 11 g4 liJe5 12 b4 bS 13 iLe2?! g6 14 gS hxg5 15 hxgS nxhl 16 J:.xhl iLlhS 17 I!1bl l1lg3 18 lIh7 (V) Thus far, Dreev has played OK and has a satisfactory position, but now his lack of experience in Sicilian posi­ tions comes back to haunt him. In­ stead of playing 18 ... QJxe2, which promises balanced chances, he makes two serious errors in quick succes­ sion, and loses in just eight more moves:



his style, this move smacks of desper­ ation.

2 exdS ltJf6 3 Chf3 ChxdS 4 d4 i.g4 5h3 ..ih5 6 c4 tUb6 7 tUc3e5 8g4 (D)



� :Jc4? 19 SI:.xc4 "I1i'xc4 20 b3 18.... 'lWt1+'! 21 �b2 a5 22 11h8 a4 (D)

Already Black is in trouble. Play­ ing with great elan, Svidler smashes through rapidly:


8, exd4 9 ltJxd4 .Jtg6 10 £g2 c6 11 0-0 ii.e7 12 £4 h6 13 fS jLh7 14 cS! Q"J6d7 15 gS!! 0-0 16 g6! fxg6 17 tiJe6 'tWc8 18 tiJxf8 hc5+ 19 Whl j,xfS 20 fxg6 �xg6 21 -txh6!! ••

Black's position is quite lost. Fine play by Svidler, but Black scarcely reo sembled a world-class GM.

23 lOdxb5! J.xb5 24 tUxbS axb3 25 tUxd6+ '<&d7 26 ltJf5+ 1-0 $vidler - Dreev Rus�'ian Ch (Elista) 1997

I e4d5 Afler his toDid time with the Sicil­ ian at Linares (in addition to the above games, he also lost heavily to Judit Polgar), Dreev decides to try some­ thing else, hut coming from a player of

After these and several other simi­ lar disasters, Dreev finally settled on the Caro-Kann with 4 ...i.f5. This proved to be the sort of solid and sound defence which suil.ed his tech­ nical style, and he soon regained his solidity against I e4, something he re­ talus to this day. Nonetheless, his ex­ periences during the period when he was trying to change openings are a salutary lesson; if such a thing can



happen to a world-clas� GM, it can

attitude was altogether different -

is not the fault of the

'This position

happen to any of us. So, are there any circumstances

opening - I'm the one who has mis­

when a o::hange of opening repertoire

played it, and it's up to me to fight my

can be justified? Despite my previous

way out of trouble". The result was

strictures, I believe that there are, al­ though these cases are

relatively rare.

Essentially, they hoil down to two


I started to defend significantly

more tenaciously than I had ever been able to do previously. This cxperience reminded me of a

main situations.

story from the world of professional snooker. Some years ago, a multiple

Bagirov's Law

world champion described the psycho­

The first situation is when one'8 cooti­

logical effect on his play of breaking

dence in an opt:ning has been so se­

his cue. Like most snooker players, he

is having a

had used the same cue throughout hi.�

verely undermined that it

negative effect on one's play. In the In­

professional career, only to have it

troduction, I described the way J used

broken beyond repair one day. The new

to hlame the opening for all of my de­

two-piece high-tech cue he started us­

feats. before I began studying chess

ing was objectively in better condition

properly and developed a cohesive

than the 30-year old, slightly warped,

repertoire. Having no proper opening repertoire, and constantly


lines which were objectively dubious

one-picce wooden one he had used all his life, yet he

was completely unable it, and his play col­

to gct uscd to

and aboutwhich I knew very little, had

lapsed. When asked what the differ­

an extremely negative effeet on my

ence was between the two cues, he

psychological state during a game.

replied "With my oId cue, if I missed a

Whenever I fell into a bad position, I had a

strong tendency to cave in psy­

straightforward pot, it was my fault; now if! miss one, it's the cue'sfault".

chologically, just saying to myself

Such psychological problems are

what's the

one reason why it may be right to

point of trying to defend this stupid

change your opening repertoire. If you

something Iikc "Oh heck, position; it's all

the opening's fault".



many defeats that you lose

Such a feeling makes it extremely dif­

contidence in the opening, the nega­

ficult to motivate oneself to mount a

tive associations are liable to haunt

in bad positions,

in doing so. Onc thing 1 no­

you whenever you play the line, and in this case, the only solution may h... to abandon the opening. In this conl,'xl. I

ticed after I began playing proper

had an interesting conversation ill _'I Ill]

successful defence

and it is notable that I hardly ever suc­ ceeded

openings was that I found it much eas­

with the Latvian OM Igors Rall�l\ J k

ier to cope with bad positions. If the

told me that the late GrallliJll:iSI'·1 Il.ll·

opening did go wrong, I found that my

irov (who became a I,atviall "III'" '"



his later years, and lived in Riga) used

never likely to be refuted. There may

to say that if you lose three game� in a

be a theoretical development which

row in the same opening, you should

�eems to put the line under pressure,

give it up, even if it's your favourite


opening, because of the negative emo­

book, such developments tend to be of


we discussed earlier in the

tions associated with repeated defeats!

temporary effect, and sooner or later

At the risk of incurring the wrath of

an improvement will

my editor, 1 have to say that 1 person­

rehabilitates the line. In the meantime,

ally find it hard to believe that a man

you may be able to avoid the problem

who played Alckhine's Defence virtu­

by varying your move-order, or adopt­

be found which

aUy all hi� life never lost three succes­

ing a slightly different line within the

sive game� with it, but there you are!

same opening. It is unlikely that you

Seriously, though, Bagjrov's basic point

will ever need to abandon your open

is well made - if your confidence in an

ing entirely.

opening goes irretrievably (whether after three consecutive losses or not), it can sometimes be best to abandon it, if only temporarily,

Dodging the Issue The last comment leads on to the sub· ject of precisely



10 change your

repertoire, if you feel that you need to. The main advice here is to try to avoid

The other main circumstance where a

changes which arc too radicaL As I dis­

change of opening is justified is if the

cussed earlier under the topic of play

line you play has been refuted. In such

ing main lines rather than sidelines, the

a case, you have little choice but to

great merit of most respectable main

abandon the line and take up some­

line openings is that they otfer flexi

thing else. especially given the extent

bility, and this can be of great use il

to which theoretical knowledge now

you need to amend your repertoire. 11

trave/s. If you stick to your old line in

your Sicilian Dragon is under pressun:

the hope that none of your opponents

and you feel the need to change. you

will know the refutation, you are tak­

do not have to abandon the Sicilian en

ing a considerable risk, and will be

tirely. Instead, you can switch perhaps

found out incrcasingly often as time

to the Accelerated Dragon, therehy

goes on, thanks to the ever-growing

obtaining positions similar to thos('

use of computer databases.

you are used to playing, but avoidinl' the specific variation you are worried

Having said that, the ri�k of a tine being refuted or rendered effectively

about_ Similarly, if you lose conli

unplayable is very small, unless you

dence in defending the 4.. tZJd7 Caw

are playing dubious and offbeat lines.

Kann, you can consider switching h '

If you play some kind of respectahle

4 �f5 lines. Such a change enabk:;

main-line opening, it is in practice

you still to exploit your knowlcdgl'




and experience of Carn-Kann posi­ tions in general, and White's various alternatives on moves 2 and 3, whereas abandoning the whole opening and taking up (say) the Pire Defence would involve learning a whole raft of brand new lines and structures.

Conclusions I) In general, it is highly risky to depart from your usual opening reper­ toire in a particular game. It may he tempting lo do so to avoid a certain variation, or to exploit a perceived gap in the opponent's repertoire, but the risks associated with playing an unfa­ miliar type of position are generally likely to outweigh the benefils. Unless you really believe there is a good chance to catch the opponent in a vari­ ation which leads to almost forced loss, you arc probably heUer off stick­ ing to your usual lines. 2) The above advice is especially important in cases where you need a certain result (Le, a draw is enough to win a tournament or fulfil a norm). Many games have been lost by players who needed only a draw, and were

11 7

tempted to play for that result straight from the opening. The best way to se­ cure half a point is to obtain a large or even decisive advantage. and then of­ fer your draw from a position of strength. You are far more likely to do that if you play something YOll know and understand, rather than something you are unfamiliar with. 3) Remember Belov's words - it is easier to win from an equal position that you have played before, than from a bad one you know nothing about! 4) Be very careful about making wholesale changeH to your repertoire. Be,ar in mind that all openings have their problems, and giving up your reg­ ular opening because of one variation you dislike is a very committal deci­ sion. The work involved in learning and mastering a whole new opening is much greater than that in improving your understanding of your existing lines. 5) If you really feel the need to al­ ter your repertoire, consider whether a smali change in variations and/or move-order will enahle you to avoid what you are unhappy about, without a wholesale change of opening.

9 Some Players' Repertoires Analysed

This final chapter consists of an exam­ ination of the opening repertoires of a set of grandmasters, from Pischer and Kasparov downwards. The idea is 10 see how leading OMs deal with tile various issues we have discussed ear­ lier in the book. In addition, you may be able to draw on some of these ex­ amples in modelling your own reper­ loire, Indeed, one effective way of forming a repel10ire is simply to lift most of it from a leading player, whose style you feel is close to your own. By following thaI player's games, you will be assured of a regular flow of fresh information on the lines COll­ cerne<l, and will be able to follow how his repertoire evolves over time.

Bobby Fischer White: I e4 - Ruy Lopez, main-line Open Sicilians, 3 .!bc3 vs the French, etc. Black 1 e4: Najdmf Sicilian Black 1 d4: King's Indian, with the occasional Gtiinfeld The outstanding feature of Fischer's repertoire was that it tended to be very narrow_ As discussed ahove, this was true of most top players of his

generation. out Fischer was perhaps even more focused that most of his contemporaries. Another point I made earlicr in the book is that players with narrow repertoires tend often to he the sort who believe very strongly in the correctness of their own ideas, and are willing to defend them through thick and thin. Fischer was certainly that sort of stubborn character, and his rep­ ertoire bears that out. Apart from a small handful of I 4:Jf3 games in his youth, and a similarly small number of games with 1 b3 and 1 c4 during 1970-2, he scarcely ever opened with anything other than 1 e4, and was on record inMy 60 Memorable Games as describing it as "best by lest". Like al­ most all top-level I c4 players, lh,' main-line I.opez was his chief weapon against 1 ... e5, and the Open Siciliau against L.cS. One particular I�ischer favourite, right from his earliest youth, was th,' Sozin Attack (6 i.c4) in the Siciliall, which he used almost exclusively against both the Classical (I e4 c5 ltJf3 d6 3 d4 cxJ4 4 4:Jxd4 ltJf6 5 li.\<' \ lbc6 6 .l1i..c4) and Najdorf ( I e4 c5 lLlf3 d6 :3 d4 cxd4 4 liJxd4 4:Jf6 5 4.10: \ at) 6 $t,c4). Th.e following is typical u! Fischer's use ofthe line: '



Fischer - Rubinetti Palma de Mallorca [Z 1970

1 e4 c5 2 ¥'lf3 d6 3 d4 cxd4 4 liJxd4 .!bf6 5 lLlc3 36 The game Fischer-Olafsson, Stock­ holm rz 1962 was an example of Fischer's successful use of the same set-up against the Classical Sicilian: 5. lLlc6 6 i.c4 e6 7 .i.b3 J..e7 8 f4 0-0 9 i.e3 liJxd4 I 0 J�.xd4 h5 I I c5 dxe5 1 2 fxeS 4Id7 13 0-0 h4 14 l1\e4 �b7 15 .'bd6 St.xd6 16 cxd6 �g5 17 "t!Ve2 iLd5 1 8 l1adl .i.xb3 1 9 axb3 e5 20 'li'h5 a5 2] �xd7 exd4 22 �f5 �xf5 23 l:!xf5 .t:.fd8 24 lIxd4 and White won the ending.

f4. That game continued S b4?! 9 ttJa4 .!i:lxe4 10 0-0 g6 1 1 f5 ! gxfS 1 2 liJxf5 :g8 1 3 i.d5 �a7 and now 14 �e3! would have given White a vi­ cious attack. 8 �b7 9 nel <L:Jbd7 10 .:tgS h6 1 l ...


�h4 liJc5? (DJ



6 .Ji.c4 e67 .1i..b3 (D) One of the earliest Fischer wins in this line went 7 0...0 .idn 8 .i.b3 iLe7 9 .te3 0-0 10 f4 '¥fic7 I I g4! <;Ph8 12 g5 4:Jg8 1 3 f5 e5 14 QxIs �d8 15 4Jf3 with a crushing advantage, FischcT­ Witte, US Open 1957.


7 b5 8 O�O The famous encounter Fischer-Tal, BledtZagreb/Belgrade Ct 1959 saw 8 •••

12�d5! exdS 13 exdS+\t>d7 14 b4 liJa4 15 lLlxa4 bxa4 16 c4 cJtc8 17 �xa4 'i!Vd7 18 \1!&'b3 gS 19 .\tg3 .!bhS 20 c5! dxc5 21 bxc5 1J!'xd5?! 2Zlle8+ 'Oitd7 23'i!Va4+ St.c6 24 liJxc6 1-0 Against the French, Fischer gener­ ally stuck to 3 !be3, preferring the positional a4-based lines against the Winawer ( I e4 e6 2 d4 dS 3 �3 St.b4 4 e5 c5 5 a3 .i.xc3+ 6 bxc3 liJ.e7 7 a4). Indeed, in My 60 Memomble Games, he declared that he doubted the Wina­ wer's soundness, but he also admitted that hi� results had not been encourag­ ing. In fact, it was not until the open­ ing game of his 1971 Candidates match against Larsen that Fischer pro­ duced a really convincing demonstra­ tion against the Winawer. In his youth



he had similar problems against the Caro-Kann, although these vanished once he finally abandoned the Two Knights Variation (1 e4 c6 2 .!i:Jc3 dS 3 tiJO). Another early favourite of Fischer's. for which he retained a penchant throughout his career, was the King's Indian Attack. In his early youth, be­ fore deciding that I e4 was best by test, he frequently opcned I lLlo, 2 g3. etc. Although he soon abandoned this, he continued to use the King's Indian Attack as an occasional weapon in po­ sitions where Black had committed himself to the move ...e6, i.e. 2 d3 against the French, and 3 d3 against I e4 c:'i 2 lLlf3 e6. A typical example was the following:




Buenos Aires 1970

1 e4 c5 A celebrated French Defence exam­ ple was Fischer-Miagmarsuren. Sousse IZ 1970: 1._.e6 2 d3 dS 3 It:xJ2 t,f)f6 4 g3 c5 5 i.g2 tDc6 6 �gO iLe7 7 0-0 0-0 8 eS li.Jd7 9 1ilel bS \0 ttJfl b4 1 1 h4aS 12 .i.f4 a4 1 3 a3! bxa3 14 bxa3 ttJaS l S lbe3 .ta6 16 .ih3 d4 17 l.iJfl ibb6 18 tlJgS lbd5 1 9 i..d2 i.xgS 20 i.xg5 �d7 21 �h5 nfcR 22 tlXi2 .'Dc3

4...d5 S fiJbd2 .!il.d6 6 .\kg2 lJJge7 7 0-0 0-0 8 4:Jh4! b6?' 9 t4 dxe4 JO dxe4 ..Ita6 1 1 lIe I c4 12c3 ct"Ja5 1 3 cS i.cS+ 14 <o&hl 4:ki5 1 5 tDe4 i.b7 1 6 'ilVh5 CDc7 17 g4! .ixe4 1 8 .i.xe4 g6 19 'ilhl6 CLld5 20 f5 lle8 21 fxg6 fxg6 22 fiJxg6! with a decisive advantage. S ..t.g2 kg? 6 0-0 tbge7 7 l:.e1 d6 8

c3 0-0 9 d4 cxd4 10 cxd4 dS 11 e5 i.d7 12W l'tcS 13 i.f4 <'baS 1411cl bS 15 b3 b4 16 lbe2 �b5 17 �d2 CLlac6 18 g4! (D)

(D). 23 ..tf61 'ti"e8 24 tbe4 g6 25 't&'g5 IDxe4 26 11xe4 c4 27 h5 cxd3 28 I1h4 tl.a7 29 £.:.g2 dxc2 30 'tWh6 �f8 3 1 'axh7+! 1-0. 2 11lf3 e6 3 .3 Iilc6 4 .3 g6 Black chose a different set-up in Fischer-Ivkov. Santa Monica 1966:

18...a5 19 CLlg3 'iYlb6 20 h4 ibb8 21 .th6 tLld7 22 '!Yg5 lhcl 23 ':xcl


.t.xh6 24 'i!fxh6 ncS 25 .u.xc8+ ttJxc8 26 hS \!!IdS? 27 IilgS IilfS (D)


malch against the same opponent, Fischer adopted the Nimzo-Indian, Semi-Tarrasch and Modem Benoni against Spassky's

I d4.

World champions naturally tend to have a major impact on setting trends in openings, and Fischer was certainly no exception, It was his patronage of the Najdorf Sicilian which really put


that variation on the map, and it was already the most popular of all Sicilian variations by the time Kasparov took it up. Another Fischer favourite, the Ex­ change Lopez ( 1 e4 e5 2 liJn �6 3 .xt.b5 a6 4 .i.xc6 dxc6), has retained a fair deal of popularity to this day, It

28 .i.e4! VJiJe7 29 Qjxh7! lDxh7 30 hxg6 f'xg6 31 �xg6 ttJgS 32 CDhS tUf3+ 33 ;i,>g2 QJh4+ 34 '>Yg3 QJxg6 35 tUf6+! cl;f7 36 '6'h7+ 1-0

had scarcely been played at all since Las.ker's time, until Fischer reintro­

duced it with such effect at the Havana Olympiad in 1966 (Fischer's particu­ lar contribution being the follow-up 5

As Black, Fischer displayed similar faith in his chosen lines, sticking with

0-0, instead of Lasker's preferences 5 d4 and 5 Cbc3),

almost religious fervour to his Najdorf and King's Indian, His Grlinfdd us­ age was pllrticularly directed against

Garry Kasparov

the latter usually adopted the Samisch

White; I e4, I d4, I c4 Black 1 e4: Najdorf Sicilian Black 1 d4: Grlint'eld, plus occasional

Variation against the King's Indian ( 1

Nimzo, QGA, QGD

Spassky in their early encounters, and probably resulted from the fact that

d4 c'tJf6 2 c4 g6 3 tOe 3 i.g7 4 e4 d6 5


In the course of his career, Fischer

lost a number of games against the

Kasparov is without doubt the hest­ prepared player the world has ever seen, and the depth of his preparation

Samisch, and generally seemed to feel

has hrought ahout

somewhat uncomfortable against it.

the overall standard of preparation at

However, he lliso lost both Grlinfe[J

all levels of the game, What I find


very clear rise in

games against Spassky (although nei­

most interesting about Kasparov's rep­

ther l()s� C<ln be attributed 10 the open­

ertoire is that he combines the two

ing) and it is significant that when it

main approaches we discussed earlier

came to the 1972 world championship

in Chapter 2. As White, he tends to



have a very wide repertoire, regularly using all three m:Jin opening moves, in search of an opening advantage. As Black, however, he tends to be very much more limited, for example, stick­ ing almost exclusively to the Najdorf against I e4. I believe this reflects the fael that Kaspurov plays almost all of his chess in super-tournaments, and rarely plays against players who :Jre more than one dass helow him. Con­ sequently, he prefers to rely on one sharp, ambitious opening rather than dabbling in several. With White, how­ ever, he i� striving the whole time to benefit from the element of surprise, and this generally means being pre­ parcd to vary one's opening� much more. Perhap� the mo�t interesting aspect or Kasparov's repertoire development in recent years has been as Black against I rl4. Up until 1997, the King's Indian had been his favourite defence ever since his youth (except for the years 1983-7), with the Gr[infeld and the occasional QGD as an alternative. However, after several King's Indian losses against Kramnik, particularly in the so-called Bayonet line ( I d4 <'iJf6 2 c4 g6 3 ctJd kg7 4 e4 d6 S ttJf3 0-06 ..i.e2 e5 7 0-0 lbc6 8 d5 (f}e7 9 b4), he has abandoned the opening altogether, as indeed have most other top-level KID practitioners (Polgar, Shirov, Gel­ fand, Van Wely, etc.). From 1997 until his world championship match against Kramnik in 2000, he stuck pretty con­ sistently to the Grilnfeld, but after los­ ing game 2 of that match, he �tarted scratching around with other openings,

mainly the QGA, Nimzo-Indian and 4...a6 Slav. At the time of writing, he continues to vary his approach against 1 d4, but he seems to have decided that the Griinfeld is not really solid enough against some the leading members of what he rather contemptuously refers to as "the Pepsi generation". Against ! e4, Kasparov\ advocacy of the NajdOli" has been almost total, since abandoning the Caro-Kann in his youth. HoweVt'T, he has on one or two occasions shown how psychologi­ cally effective an unexpected switch can be, most notably when he sur.. prised Anand with the Dragon during their 1995 world championship match. At the time of writing, he has begun using the Sveshnikov Sicilian more regularly, most notably surprising Shi­ rov with it in the last round of Linares 2002, to notch up yet another one­ sided victory against his favourite cus­ tomer. 'The most stunning quality of Kas­ paruv's opening repertoire is his abil ity to produce innovations that arc nol merely small improvement� on whal ha� gone before, but which com· pletely overturn existing opening th,' ory. There arc maoy examples of thi� in his games, one notahle l:ill-ie bein.!' the line 1 e4 e5 2 t/:Jf3 Cbc6 3 iLhS a6 1 : &.a4 €lf6 S 0-0 €lxe4 6 d4 b5 7 Jl.h\ dS 8 dxe5 Si.e6 9 4Jbd2lbcs 10 c3 ( 1· 1 ! 1 lbg5 dxc3 1 2 lbxe6 fxe6 13 bx<' ( '@'d3 (D). This had been con�idered satisfa, tory for Black ever since its first ;\1' pearance in Karpov-Korchnoi, Baglll<' City Weh (10) 1978. However, wlh"1I



Anand wa::; unwise enough to repeat the line in game 1 0 of his 1995 match with Kasparov, the latter uncorked 1 4 .ie2! 'lWxe3 1 5 liJb3!, offering a rook for the attack. The idea itself was not. totally new, with 1 4 i.c2 having been casually suggested by Tal in notes to the Karpov-Korehnoi game. However, nobody had played it or analysed it se­ riously until Kasparov came along and demomtrated to Anand that it wins by force. Nobody bas played J L..dxc3 at lop level ever sincc. In addition to refuting entire open­ ing lines, Kasparov is also responsi­ hie for turning several previously ne­ glected opening� into serious proposi­ lions at top GM level. One such is the Scotch, which he first used against Karpov in game 1 4 of the final match in New YorklLyons in 1990. Thanks to Kasparov's patronage, the Scotch has arhieved widespread popularity and now rivals the Lopez as "'·hite's main Iry for advantage after I e4 e5. Another line which Kasparov is al­ lIlost single-handedly responsible for ll1<1king into a major weapon is 4 �c2


against the Nimzo-Indian. Apart from a few games by Seirawan, no top GM had regarded this as a really serious try for advantage, until Kasparov started using it in the late 1980s. [t soon be­ came overwhelmingly the most popu­ lar reply to the Nimzo, and remains so to this today. One of the bedrocks of Kasparov's revival of the line was the variation 1 d4 t['lf6 2 c4e6 3 Q\c3 i.b4 4 fffc2 d5 5 cxd5 exd5 6 ;tg5 h6 7 .ih4c5 (D).


Until Kasparov got to work on it, this had been regarded as fine for Black for almost half a century, ever since the game Keres-Botvinnik, USSR Ab· solute Ch (Leningrad/Moscow) 1941, which had famously ended in a 22move debacle for White: 8 0-0-0'1 !.xc3 9 Wkxc3 g5 10 .Jig3 cxd4 1 1 W'xd4 tbc6 1 2 'ii'a4 i.f5 1 3 e3 Itc8 (it is already clear that the white king is caught in a lethal crossfire) 1 4 i.d3 �d7 I S Wbl �xd3+ 16 ltxd3 'i§'f5 (D) 1 7 e4ttJxe4 1 8 Wal 0-0 1 9 :d I bS 20 �xb5 iLld4 21 �d1 Cbc2+ 22 tirbl '1lb4 0-1.



fared little better: 9 e3 gS 10 .itg3 lDe4 t I lLJO fff6 1 2 lbxc3 1 3 iLxc6+ bxc6 (D).

However, Kasparov's extensive and never--ending ,malytical work produced the major improvement 8 dxcS!, which completely overtumed the theoretical verdict on 4...dS. The first time he used the move was in Kasparov-Korchnoi, Tilburg 1989. Korchnoi reacted badly with 8...0-0?! and stood clearly worse after 9 e3 ti'lbd7 10 Ji.d3 '/iaS 1 1 <'be2 .i.xc3+ 12 'tWxc3 iYxc3+ 1 3 itJxr3 "bxcs 14 iLc2 gS I S i/..g3 �e6 16 f) a6 17 h4 (D) (1-0, 27)


A year later at Linares, Spassky at­ tempted to improve with 8...ttJc6 but

14 a3 g4 IS i.eS tLJe4+ 16 axb4 �fS 17 .ixh8 gxB 1 8 llgl l/Wg4? 1 9 'iWd 1 ttJg5 20 'iVd4 ttJe4 21 '&e5+ .te6 22 'm'f4 �g6 23 �xB f6 24 �f4 �f7 25 f3 tiJg5 26 'it>d2! �f5 27 h4 'liYxf4 28 ext4 "bh7 29 g4 1-0. Finally, Nigel Short made his at­ tempt to strengthen Black's play in his world championship match against Kasparoy in [993: 8...g5 9 J..g3 0.e4 to e3 �a5 (D).


Thir> all-out attack on the pinned c3-knighl was the basis of Short's new idea. On its initial outing, in the fifth game, the idea secured a comfortable draw, but when Short wa� unwise enough to repeat the line four games later, he walked into another powerful piece of Kasparov preparation, in the shape of I l ltJc2! .tf5 12 .kcS! : 0-0 13 "iJM iLg6? (NCO giver> 13 .. .tLlx:c3 1 4 liJxf5 4:\e4+ 15 Wdl ltk 6 1 6 .td6 with a slight advantage to White) 1 4 .!Db3 l'Llxc3 15 i.xc3!. White was much better and went on to win.

Anatoly Karpov White: I d4

Black 1 e4: Cam-Kann with 4. tt:hl7 Black 1 d4: Nimzo and Queen's In­ ..

dian Kafpov is a leading example of a player who has always had a narrow repertoire. In addition, his approach is unlikl': that of both Pischer and Kas­ paro\', both of whom tend to believe in active countelplay with the black pieces. Karpov is much closer to the more classical 'win with White, draw with Black' approach, and throughout his career he has generally concen­ trated on solid equalization as his first priority with the black pieccs. Against I d4, he has played the Nimzo and ()ueen's Indian almost exclusively �ince his earliest youth. varying only with the occasional QGD, Tartako­ Wl:f Variation (I d4 d5 2 c4 e6 3 &3 1,;\f6 4 <'tJf3 iLe7 5 i.g5 h6 6 .i.h4 0-0 7 ( \ b6). During most of his world '


championship years, he generally met 1 e4 with 1 ... e5, initially playing the Breyer Defence (1 e4 e5 2 liJf3 .'tJc6 3 .�b5 a6 4 ii.a4 4:Jf6 5 0-0 SJ..e7 6 ld.el b5 7 .i.b3 d6 8 c3 0-0 9 h3 CDbS), and later switching to the Zaitsev Varia­ tion ( I c4 e5 2 liJf3 fbc6 3 .ib5 a6 4 j.,a4 <?lf6 5 0-0 i.e7 6 11e l bS 7 .ib3 d6 8 c3 0-0 9 h3 .Jtb7 10 d4 neS). However, around the late 1980s. he started using the 4...�)(17 Caro-Kann regularly, and this has since become his almost exclu�ive defence to 1 e4. In particular, he has played many games with the rather odd-looking variation I e4 e6 2 d4 dS 3 t,;'\e3 dxe4 4 lZ.lxe4 !bd7 5 iLd3 liJgf6 6 tbgS e6 7 ttJ I f3 i.d6 8 'iVe2 h6 9 lDe4 liJxe4 1 0 'l§'xe4 ffie7 1 1 '1lVg4 'it'fS (D), which does not look terribly Karpovian at all, yet has become a firm favourite of his.

A� White. Karpov has played ex­ clusively 1 d4 for the past 1 5 years or so. However, it is interesting to note that up 10 the time of his world cham­ -pionship matches with Kasparov in



the mid-1 980s, he had always been overwhelmingly a I e4 player. Wh�n examining some of the K-K opening duels in Chapter .5, 1 pointed out that Kasparov had started that series of matches as primarily a 1 d4 player, but had begun using 1 e4 as his main weapon by the end. Over the same pe­ riod, Karpov made the opposite tran­ sition, starting out with I e4, but switching to I d4 when he found that he could not get the advantage against Kasparov's Sicilian. Just as Karpov has always believed in the Queen's Indian as Black, w as White he has al­ ways preferred to take on that open­ ing, rather than allow the Nimzo. Indeed, jf there is any one opening which Karpov has really made his own over the yean;, it is the Queen's Indian, in which his deep and subtle understanding has brought innumera­ ble points on both sides of the board. Precisely because he has almost al­ ways answered 1 d4 t'tJf6 2 c4 e6 with 3 4:'10, you win find very few exam­ ples of the E)(changc QGD in Kar­ pov's games as White; as we noted earlier, the Exchange Variation does not work well with White's king's knight committed to L=) early on, and Karpov instead usually plays main­ line QGDs as White. Karpov ha� always displayed a quite exceptional natural feeling for the Queen's Indian's nuances. In his early years, he generally played the main lines with .,.�b7 as Black, hut in more recent years the popular ...,l1.a6 line� have hecome Karpov's home ter­ ritory from both sides of the board.

The following is one recent examp],­ of his prowess in this variation:

Karpov - Adianto Rali 20()() 1 c4liJf6 2 d4 e6 3 ti)f3 b6 4 g3 $.a(. :­ b3 j,b4+ 6 .ltd2 j,e.7 7 .i.g2 c6 8 .!it.d rl5 9 ttJe5 q)t'd7 10 tZlxd7 4Jxd7 I I £Dd2 O�O 12 O�() (D)

This position has arisen many tinJ('", in the last two decades. 12...Il:c8 Karpov-Ehlves\, Bali 2000 took III<" following interesting course: 12 ... 4\11) 1 3 e4 b5 1 4 11el dxe4 1 5 �c2 I I , l::t.adl .!bdS! 1 7 ,h � e4! ttJxc3 18 'ili'x< \ JH6 1 9 ..:.5!? j..xd4 20 '&c2 f5 2 1 ,O)' (D). White has compensation for Ill< pawn, in the shape of Black's bad d(, bishop and weak c6- and e6-pa\'liJ i After some errors by his OPPOlll"l,1 Karpov went on to win. 13 e4 dxc4 The old main line with 1 3 ...c5 h,,·. been the scene of countless olh, , '



This impatient central break allows White a powerful passed d-pawn. The superior 1 6...�e7 brought Black equal­


ity in Karpov-Ka�parov, LondonlLe­ ningrad Weh (21) 1986.

17 d5 exd5 18 exd5 i.f6 19 Itadl 4:\b6 20 .txf6 '!'fxf6 21 a4 ':fd8 22 QJe4 'iVg6 23 d611b8 24 �c3 i.b7 25 as ltJa4 26 iVa3 QJb2!? 27 'i!fxb2 "�,4 (D)

Karpov games. One recenl outing saw him win with astonishing ease: 14 exd5 exd5 1 5 dxc5 dxc4 1 6 c6 cxb3 17

ltd b2 18 .Jixb2 �c5 1 9

\'"I'g4? (a poor innovation; 1 9 4Jb3 and 19 Jta3 arc better tries) 1 9 .. .;if6 20 .

.li.xf6 �xf6 2 1 lDe4? (losing a pawn ror nothing, but the threats of . . .lZki3 ilIId ....tb5 were already difficult to meet) 2 1 ...tWxc6 22 �xc5 ¥&'xcS and Black was simply a pawn up for noth­ illg in Van der Sterren-Kiupov, Wijk aim

Zee 1 998.

14 bxc4 b5 IS %1el bxc4 16 t'!fc2

('5'l! (D)

28 'lli'xb8! Although materially White will have only rook and bishop for the queen, the passed d-pawn will be decisive.

28 lhb8 29 ttxe4 fS 30 llxc4 .ti.dS 31 d7'/>l¥a6 3211xc5 �e2 33 l:.ccl 'ot>1'7 34 .id5+ <t>f6 35 .tIel 'S'd3 36 nedl �e2 37 ±L'6 g6 38 lael Wild2 39 nedl �c2 40 ttd6+ ""f7 41 ne8 1-0 .••

Vladimir Kramnik White: 1 d4, 1 liJf3 Black 1 e4: Ruy Lopez

(Berlin), Pe­

troff, Sveshnikov Sicilian,



Black 1 d4: Nimzo, QGD, Semi-Slav



The immediately striking thing about Kramnik's repertoire, compared with most of the players ahove, is how much more varied it is, particularly with Black. In Ihis respect, Kramnik is

Nimzo-Indian, and it is much harder for Black to find the necessary coun­ terplay to offs et the surrender of the bishop-pair. A few of Kra mnik's games in this line illustrate the point:

typical of the current generation of super�GMs, who tend to use far more


�ors . In fact, Kramnik is le ss varied than some of the othe rs, which is partly why I have included him- if we tried to examine Anand's repertoire, for example, it would encompass at least half of all the main-line open­ mgs. As White, Kramnik has varied be­

tween I d4 and I ttJf3, At the lime of writing, he tends to be using 1 d4, as he was when he first broke through to the top level in 1991-2, but for a num­ herof years between 1993-?, he almost exclusively u sed I ltJf3. He tended to use this primarily as a transpositional tool, often tmnsposing back into regu­ lar queen's pawn openin gs such as the QGD, King's Indian, etc., but at the same time avoiding certain other lines, such as the Benoni, Benko, Griinfeld (although sometimes he would go back into regular Grtinfeld lines), etc, One of the lines with which Kramnik scored especially heavily was the anti-Nimzo-Indian system 1 ttJB (fjf6 2 c4 e6 3 lbc3 .tb4 4 t\IIc2. This line



Vierlflu 19%

different openings than their pred{.'Ccs�

1 'iJf3 ltJf6 2 c4 e6 3 ttJc3 iLb44 'j&c2 4 'i!Yb3 is simil ar Although the .

queen is not always so well placed on b3 as on <.'2, the move has the merit of forcing Black toplay 4. .c5, a move he m ay otherwise choose to delay or avoid altogether. Kramnik has used both queen moves, an example of 4 'ab3 being the following: 4...c.5 .5 g3 ttx6 6 .\tg2 0-0 7 0-0 d6 8 d3 h6 9 e3 eS to a3 .liaS 1 1 QJd2 �e7?! 12 ibdS .

liJxdS 13 cxdS .lixd2 14 ilxd 2 4Jb 8 I S d 4 0.d7'!! 1 6 dxe.5 QJxe.5 17 f4 ± iZld7 (0).



bears a super ficial similarity to the 4 'ii'c2 Nimzo, but here there is the im­ portant difference that White has not committed his d-p awn This means that he retains the option of d3, keep­ ing control of the e4-square, so often a basis for Black 's counterplay in the .

18 e4 b6 19 :tfd �a6 20 i.c3 c l 21. lWd I ibc s 2 2 'ii'g4 f6 2 3 eS .ltc8 ].j 'ijl'h4 �3 25 exf6 fie? 26 h31 gxf6 1.'! .lixf6! lbxcl 28 }hel 't&'t7 29 j.e \



'ffg6 30.!:!.e7 �n 3 1 l;tc8+1H8 32 ,te4

followed by 1 6 liJd2, would have

1-0 Kramnik-Hratek, Berlin 1996.

given White a clear advantage.

4...0-0 Againsl 4 ..c5 Kramnik introduced .


.Qjf3 is extremely flexible, but the


early commitment of the knight to f3

6 g3 lbc6 7 ..tg2 i.xc3 8 '1Wxc3 �a5 9

does cut out certain options. Whcn

a promising pawn sacrifice: S a3

b4!? After 9 ... cxb4 10 axb4 li'xb4 1 1

Kramnik first came to prominence, he

'is-'xb4 ctJxh4 Iz lhd4 dS 1 3 il.a3 ttJc6

tended to use the Samisch against the

1 4 lbb5 he had plenty of compensa­

King's Indian, winning many excel­

tion, and went on to win handily in

lent games with it. However, when he

Kramnik-Romanishin, Belgrade 1993.

decided to start opening 1 CilB, he

5 a3 i..xc3 6 �xc3 c5 7 b4 b6 8 g3 �"6

was forced to find another anti-KID

S".i.b7 9 Ji.g2 tr ansposes to Kram­

years with the Petrosian System (1 d4

nik-Illesl.:as. Dos Hennanas 1997, in

lLJf6 2 c4 g6 3 lbc3 Ji.g7 4 e4 d6 S 1LJf3

whirh Black preferred the more natu­

0--0 6 ..te2 e5 7 dS), including two vic­

weapon. After a couple of successful

ral development of the knight on d7,

tories over Kasparov, he took up the

but after 9_ ..d6 10 0-0 Qj1xl7 1 1 �bZ

newly-popular Bayonet Variation re­

'¥Ie7 12 d3 !tfc8 1 3 bS a6 1 4 a4 axb5

ferred to above, with equal success.

15 axb5 '5'fS 1 6 e4 White stood some­

what better.

9 i.g2 .�.b7 10 0-0 d6 11 j�b2 e5 12e3 1lc8 13 d311e8? 14 b5! !be7 (D)

As noted above, this line has been re­ sponsible for


shmp decline in the

popularity of the King's Indian at the highest level. The great mnge of black openings is in contmst with the approach of someone such as Kasparov. Whereas the lalter has tended to be fairly pre­


dictable as Black (certainly until very recently), but more varied as White, Kramnik is rather the opposite, show­ ing rather greater variety with the black pieces. I think this can be ex­ plained largely hy the difference in style between himself and Kasparov. The latter is far more of a maximalist, who tends to be playing for a large ad­ vantage as White. For this reason, he


KJamnik now initiated what proved

is constantly searching for new paths.

be favourable tactical complica­

and thus has to vm'y his white open­

tions with 15 lLlxe5!?, but even in the

ings a great deaL Kramnik, on the

absence of these, the simple 15 e4,

other hand, is rather less aggressive



and instead excels in exploiting rel­

style, he has hardly ever been tempted

atively small positional advantage�.

hy opening� such as the King's Indian

Consequently, he is able and willing


to play a smaller variety of openings

healthy living by beating up such sys­

as White, knowing that while he is un­

tems with White.

Grtinfeld. and has instead made a

likely to secure really big advantages much of the time, he is nonnally going to be able to count on


small edge at

least, and this normally satisfies him. As Black, on the other hand, he has tended to move over the years from

Michael Adams White: I e4 Black 1 e4: 1 .. e5, Marshall Attack., .


shatTJCf lines towards much more solid

Black 1 d4: Nimzo and Queen's In­

variation�. Against 1 e4, fOf cX<lmple,


he played the Classical Sicilian a great deal during the early 1990s, and then

Michael Adams is the last of the

switched to the highly-sharp Svesh­

super-GM� whose repertoire we will

nikov line

he examining here. His style has al­

( I e4 c5 2 ll:lf3 tDc6 3 d4

cxd4 4 ttJxd4 lbf6 5 �3 e5 6 lbdb5 d6

ways been strongly reminiscent of the

7 £g5 a6 8 c';)a3 b5). However, over

Capahlanca/SmysJov/Karpov school,

the last 5 years or so, along with other

in that il is based primarily on a superb

members of the super-tournament cir­

natural instinct for where to place his

cuit, he has begun playing much less

pieces. In geneml, Adams shies away

sharply as Black, instead aiming only

from ultra-sharp openings, and is nor­

for solid and drawish equality. To this

mally content to seek a small advan­

end, he has played the Petrotf exten­

tage with White, and to play primarily

sively, and more recently the Berlin

for equality with the black pieces.

Defence, discussed in Chapter 3.

This tendency has become even more

He has also shown similar flexibil­

clear in his play in recent years, since

ity as Black against 1 d4. In the early

his career has concentrated on super­

1990s, he generally stuck faithfully to

tournaments, where the emphasis is

the Semi-Slav (1 d4 d5 2. .:4 c6 3 4.Jf3

on winning with White and drawing

4.Jf6 41Lk3

e6), a favourite of a whole

generation of ex-Soviet�, including

with Black. With White, Adams has always

1 e4 player, apart from a hand­

Shirov, lvanchuk, Dreev, M.Gurevich,

been a

etc. However, as that opening ha� been

ful of experiments with the Trompow·

progressively analysed almost to death,

sky. At the time of writing, he usually

Kramnik gradually �topped playing

plays main-line openings, such as the

it, and has instead used the Nimzo�

Ruy Lopez against 1 .. ,e5 and open Si­

Indian and QGD, plus occasional al­

cilians against 1 ...c5, However, until

ternatives. It is notable, however, that

recently, he was notable amongst top

with his solid and ciassicalIy con-cct

GMs for being relatively reluctant to



enter into theoretical disputes, particu­ larly in the Sicilian. To that end, he ha$ played a number of offbeat lines again$t the Sicilian, including the Clost:d, 2 c3, and I e4 c5 2 1Z1f3 d6 3 i.c4. Although he won many games against evea world-c1a,<;s GMs with thi� approach, he ha� over the pa�t couple of years begun to move away from such non-critical lines, in favour of main-line open Sicilians. As Black, Adams's repertoire is marked principally hy its solidity. After playing the Caro-Kann quite ex­ tcnsively in his youth, hc gradually started playing 1 ...e5 more frequently. until it took over a.<; his almost sole re­ ply to I c4. One interesting point is that he now plays the Marshall Attack as his main defence, which may at first sight seem out of keeping with his generally solid approach as Black. However, contrary to its fearsome rep­ utation, the Marshall has been ana­ lysed to such an extent at the top level that it is regarded by the leading play­ ers as primarily an equalizing weapon. Thert: are many lines which lead to a forced draw. and White's prohlem i� not in drawing against it, but in finding a way to secure some advantage. At the sort of supcr-toumament level at which Adams plays most of his chess, the Marshall is an ideal choice, but at lower levels, the presence of numer­ ous drawing lines for White may make it unattractive to players who wish to play for it win as Black against weaker opponents. This is an issue to which we will retum below, when we look at the repertoire of English GM Mark

Hebden. Adams's usual solution is to switch to the MolierlNew Archangel lines ( I e4 eS 2 ibf3 'bc6 3 JibS a6 4 J..a4 Q\f6 S 0-0, and now either S...bS 6 i.b3 .tcS or S ....i.cS) when facing lower-rated opposition. The following is one very nice example: Quillan - Adams British League (4NCL)


1 e4 eS 2 0if3 41c6 3 i..bS a6 4 L4 lDf6 S 0-0 bS 6 �b3 .i.c5 7 c3 d6 (D)


Ii d4 i.b6 9 kg5

9 ..'ilLe3 SJ..g4 1 0 dxeS €lxeS 1 1 i.xb6 .txf3 1 2 gxf3 cxb6 1 3 f4 tbc4 1 4 e5 dxe5 15 Si.xc4 bx("4 1 6 fxeS tbe4 with an unclear position wa,<; Wahls-Ad­ mns, BundesJiga 1994/5. Black even­ tually won. Another disaster for White was 9 h3 .th7 10 lleI 0-0 I I il.e3 CZ'la5 12 i..e2 ibc4 1 3 i.el d5 14 b3? dxe4 1 5 lllxe5 lZlxe5 1 6 dxe5 ttJd7 17 .i.f4 "@h4 18 g3 W'xh3 19 i.xe4 i.xe4 20 tl.xe4 lbc5 21 rie2 ztad8 f)-J Morovic-Adams, Santiago (6) 1997.



9 h6 W .ltb4 gS 1 1 .ltg3 0-0 12 dxe5?! ...

1 2 �bd2 cxd4 13 cxd4 �xd4 14 4Jxd4 �xd4 15 1Of3 ii..b6 16 'tic2 liJh5 17 i..dS 'Llxg3 1 8 hxg3 llb8 1 9 e S <;i;>g7 2 0 lladl with compensation, IPolgar-Adams, Dos Hennanas 1 995,

12 'Llxe5 13 .i.xe5 dxe5 14 1:!1r'xd8 ':xd8 15 0:lxc5 <Ji>g7!! (D) •..

GMs, his Benko began to suffer quite badly and he soon abandoned it. I remember personally witnessing two particularly severe defeats: Barecv­ Adams, Hastings 199112, and Timo­ shchenko-Adams, London Lloyds Bank 1992. With the Nirnzo and Quecn's Indian complex, however, Adams has been highly successful, and jllst as with Karpov, it is clear that these clas­ sically solid systems suit his style per­ fectly.

Mikhail Gurevich White: I d4. 1 c4 Black 1 e4: French, Pirc Black 1 d4: Semi-Slav, occasional Leningrad Dutch

16 a4 .i:le8 17 Q:jd3 ii..b71 18 eS lbg4 19 axbS i..e4! 20 Jtc4 .l:tad8! 21 h3 �xd3 22 i.xd3 4lxf2 23 i.c4 axb5 24 1i.xb5 X!xe5 25 liJa3 lId2 26 liJc4 lhb5 27 QJxb6 'tJxh3+ 28 gxh3 J.'!.bxb2! 0-1 Against 1 d4, Adams has played the Nimzo for virtually the whole of his ca­ reer. The only significant exception was a brief period in the early 1990s, when he used the Benko Gambit. This was at a period of his career when he was playing in opens and rdatively weak all-play-all events, and winning with Bl ack was rather more of a prior­ ity. Once he slarted moving up to stron­ ger events. and encountering top-class

Thus far, with the exception of the inactive Fischcr, all of the players whose repertoires we have examined arc top-flight GMs, whose chess is al­ most all played in super tournaments against other elite GMs. Gurevich, the former Soviet star now resident in Bel­ gium, is in a somewhat different posi­ tion. Although extremely strong, he is not one of the charmed circle who tend \0 monopolize the invitations to the top all-play-all cvents. Instead, he plays the majority of his chess in open toumaments, which means that his priorities are somewhat different. Rather than being faced with players of his own strength all of the lime, Gurevich plays many games against weaker players, and the nature of open tournaments is such that it is fre­ quently necessary to be ahle to win -


against such players, regardless of colour. The classical 'win with White, draw with Black' approach rarely suf­ fices for success in such events. Faced with this problem, many play­ ers adopt very sharp systems, aiming to maximize winning prospects by stirring up early tactical imbalance. However, Gurevich is an outstanding example of how a sounder and more technical style can he equally effec­ tive, and his approach is reflected in his opening repertoire. He is basically a velY correct player, who relit:s on outplaying opponents ill strategically difficult positions and long endgames. rather than taking undue risks to cre­ ate sharp positions. The effectiveness of such an approach, in the right hands, is shown by his results; over the past five years or so, he has won innu­ merable international opens through­ out Europe. and in 200I scored a Fischeresque 9/9 in the Belgian Cham­ pionship. As White, Gurevich usually opens I d4 or I c4. From my observation, he tends to use the latter move more often against the weaker opponents. proba­ bly because the flexibility and numer­ ous move-order options in the English make it easier to lure such opponents into less familiar territory. Nonethe­ less, Gurevich is certainly not the sort of player who backs out of theoretical disputes. He has long enjoyed a repu­ tation as one of the most erudite of OMs, and is certainly a world expert on many of his favourite lines, such as 4 �c2 against the Nimzo and 5 �f4 in the QGD.


The latter in particular is a line in which Gurevich has won innumerable game� and considerably enriched the theory. For example, after J d4 d5 2 c4 e6 3 tbc3 Qjf6 4liJf3 i.e? 5 .i..r4 0-0 6 e3 c5 ? dxc5 i.xc5 8 a3 lbc6 9 �c2 'l!Va5 it was Gurevich who invented the line beginning 10 0-0-0 (D).

Gurevich introduced this move in two games in the 1988 USSR Cham­ pionship_ The effect of surprise can he seen i n the way Andrei Sokolov, runner-up to Karpov in the Candi­ dates cyde only one year earlier, wl­ lapsed when faced with Gurevich's new idea: 10 ... dxc4 I I i.xc4 iLe? 1 2 g4 b5? (current lheory recommends 1 2...e5 as equalizing) 13 fi.xb5 i.b7 1 4 ttJd2 iLlh4 1 5 axb4 iLxb4 1 6 <'bc4 �al+ 1 7 ""d2 .i.xc3+ 1 8 'i.'e2 't'J"a2 1 9 lIal 1-0 M.Gurevich-A.5okolov, USSR Ch (Moscow) 1988. Despite his initial success with 10 0--0-0 - . however, Gurevich has rarely played it since. generally preferring the quieter line 9 ..i.e2. After a subse­ quent pawn exchange on c4, this leads



to positions with a symmetrical pawn­

well as generally avoiding major early

structure, where White has a very

simplification. Nevertheless, Gure­

small edge, based on his more active

vich's main defence is the French, and

qucen's bishop. It may not look like

herc his choices are an excellent ex­

very much, but in practice it often

ample of how one can have variety in

proves to be the sort of nagging little

one's repertoire without involving

edge that a superb technician such as

many different openings. Like almost

Gurevich is able to exploit to the max­

all leading French players nowadays,

I e4 e6 2 d4 tl5 3

imum. Yet another way to play the

fie I , followed by an ex­

Gurevich answen;

position is 9

.:Dc3 with 3 ... COf6, rather than the

change on d5, leaving Black with an

Winawer. After 4 i.g5, he almost al­

isolated d-pawn. One typical example

ways continues 4 ... dxe4 5 <L\xe4, but

of this i� 9 llc 1 fi.e7 10 cxd5 lbxd5 I I

here there are several options for

.:t:lxd5 exdS 1 2 i.d..1 ..tf6 1 3 0-0 j,e6

Black, all of which Gurevich uses. His

( l 3 ...Jbb2 1 4 .txh7+ is hetter for

most common choice is 5...St.e7 6 .I11.xf6

White) 1 4 b4 a6 and now the thematic

i-xf6, but he also uses the sharper re­

manoeuvre 1 5 �c5! (D) left White with

capture 6 ...gxf6, as well as the 5th

a �mall bul steady edge in M.Gure­

move alternative 5 ...4"\bd7. Similarly,

vich-Peelen, Dutch Chi 1 998.

against the Tarrasch French (3 tDd2), he generally eschews both the sharp 3."l2)f6 and the trendy 3 ....te7, in fa­ vour of the IQP positions reached after 3.,.c5 4 exd5 exd5. However, here too he varies his systems after 5 .:t:lgf3, generally choosing between 5 ....:t:lf6 and 5...a6. Both lines usually lead to IQP structures for Black, but there are small differences between them, which makes specific preparation by White more difficult, while still allowing Gurevich to draw on his vast experi­ ence and understanding of the generic structure.

Gurevich's repertoire as Black ver­ sus I

e4 consists of the French and

Against 1 d4, Gurevich sticks pretty faithfully to the Semi-Slav. bllt he

Pirc, and once again, he tends to use

does venture the occasional Lenin··

the Pirc more often against the weaker

grad Dutch. The .lauer is an excellent

opponent'>. This makes good sense,

choice against weaker opponents, but

since the Pirc!Modem complex aJlows

it is notable that Gurevich uses it spar­

a lot of flex.ibility and is harder to

ingly against. grandmaster opposition.

prepare specific sequences against, as

The Semi-Slav is a system which has


been analysed in enormous depth, and in which there is a huge range of move-orders and alternative set-ups. However, most of the choiccs lie with Black, whi(;h means that White often has to know far more than Black. By picking his own preferences, Black can nalTow down the range of lines he needs to prepare, which makes the Semi-Slav a good practical choice. Gurevich is one of a whole genera­ tion of Soviet players brought up on thi� opening, and his vast knowledge and experience make i t a formidable weapon in his hands_


great belief in the value of the pawn­ centre. Thus, he always opens 1 e4, and is finnly convinced that any reply which allows him to follow up with 2 d4 mu�t already give White some ad­ vantage. On Ihis basis, the French, Pirc, Caro-Kann, etc., are dismissed as inadequate for equality in Sveshni­ kov ' s scheme of things. Effectively, this leaves L.e5 and 1...c5 as the only correct replies. His patronage of 2 c3 against the latter move is further evi­ dence ofbis faith in the two-pawn cen­

Sveshnikov is probably the single greatest contemporary example of a player with a tru ly fanatical belief in the correctness of his own ideas about openings . Throughout his career, he

Ire, whereas to play the Open Sicilian mean abandoning forever the chance of estahlishing an e4-d4 pawn­ centre. It is interesting to note that in recent years, he has also begun experi­ menting with 2 1'4 against the Sicilian (principally because of the extent to which 2 c3 has been analysed), but even here he never misses the chance to establish a big pawn-centre. After 2 f4 liJc6 3 <iJf3 g6, for example, he al­ ways continues 4 c3, rather than the 4 i.h5 which is preferred by most f4 Si­ cilian players. This attitude also applies as Black.

has remained faithful to virtually the

l11roughout his career, Svcshnikov has

Evgeny Sveshnikov White: 1 e4 - c3 Sicilian, Evans Gam­ bit, Advance French, etc. Black 1 e4: Sicil ian 4.. .e5 Black 1 d4: Semi-Slav


same systems. and continues to up­

believed in the value ofBlack's central

hold his beliefs, both in practical play

pawn-majority in the Open Sicilian,

and polemical discussions. Anybody

and he has always practised lines which involve the eady advance ...eS. In these systems, Black makes imme­ diate use ofthis central pawn-majOlily to secure control over key squares such as d4 and f4, and to squash any chance of a white central pawn ad­ vance with f4 and e5. Initially, Svesh­ nikov's attention was focussed on the highly sharp Pelikan Variation ( 1 e4

who plays Sveshnikov can be 99?(I

certain of the position which will he on the board after 10 moves or so, but he can also be equally sure that the same position has been on Sveshni­ kov's board in home analysis for count­ lcs� hours during the past 30 years.

The theme which

derpins all of


Sveshnikov's opening systems is a



c5 2 tiJf3 lbc6 3 d4 cxd4 4 lDxd4 lDf6 S lDc3 eS), and his contribution to this line was so great that the whole com­ plex is nowadays often referred to as the Sveshnikov Variation. However, during the 1980s, the man himself switched to the alternative line 1 e4 c5 2 lLJf3 liJc6 3 d4 cxd4 4 iiJxd4 e5. Although this had been known for many years under the appellation of the Lowenthal Variation, Sve�hnikov's interpretation was radically new, in that he answered 5 t.1Jb5 with 5...d6 (rather than 5 ... a6, which character­ izes the Lowenthal proper). The simi­ larity of this new line (which has acquired the rather silly name of the Kalashnikov Variation) to the Svesh­ nikov Variation is clear, in that Black once again uses the e-pawn to estab­ lish central influence, at the cost of ceding the d5-�uare. However, in the Kalashnikov, the fact that Black's King's knight is not yet developed pre­ vents White from pinning it with i.g5. Instead, White has the extra option of clamping down on the dS-square with 6 c4, which leads to slower and alto­ gether less sharp positions than those in the Sveshnikov Variation. Sveshnikov has an almost fanatical belief in the correctness of his ideas in these lines, and is on record as claim­ ing that after 1 e4 c5 2 LtJf3 ibc6 White's only try for an advantage is 3 .tb5. Although he has attracted few regular supp0l1ers (ironically, his fonner favourite the Sveshnikov Vari­ ation remains popular even at the highest level), there is no arguing with his own results - on my database, he

has scored almost 70% as Black in the Kalashnikov. The following thematic example is typical:


Sveshnikov Moscow /992 -

1 e4 cS 2 'Llf3 'Llc6 3 d4 cxd4 4 lbxd4 eS 5 iiJbS d6 6 c4 Against 6 lblc3 a67 fi)a3 h5 s lbds,

Sveshnikov generally prefers 8. ..4Jce7; C.g 9 c4 tt.lxd5 1O �xd5 JiLc6 1 1 �b7 liJf6 1 2 cxbS nbS 1 3 �xa6 d5 with compensation, Bologan-Sveshnikov, Riga 1995. .•


te7 7 <illc3


7 .i.d3 J.e6 8 0-0 lIc8 9 b3 liJf6 10 .1b2 0-0 1 1 VJlc3 a6 1 2 -LJa3 4::le8 13 lbd5 ..tg5 1 4 fbc2 the? 15 liJceJ liJxd5 16 cxd5 j.d7 Tiviakov-Sveshnikov, Podolsk 1 993. 7...a6 8 liJa3 .Jte6 (D) 0=

9 tiJc2 9 .1l.e2 tiJd4 10 ttJc2 liJxe2 1 1 'iWxe2 lDf6 12 0-0 llc8 1 3 4Je3 0-0 14 ndl?! 'fIle7 15 b3 b5 1 6 lbed5 ctJxd5 1 7 liJxd5 & : xd5 1 8 cxdS '/!¥c2 1 9 "iVg4??


f5 20 exfS h5 2 1 '§'f3 e4 22 �xh5 llxf5 0-1 Ulybin-Sveshnikov, USSR Chi (Naberezhnye Chelny) 1988. Another alternative is 9 .'Dd5 �c8 10 i.e3 �g5 1 1 i.b6 �d7 12 iLe2 .Jtd8 13 0-0 ttJge7 14 �e3 0-0 15 c5 0.d4 16 .:lxd4 exd4 17 cxd6 '5'xd6 1 8 ttJc4 �c5 1 9 b4 �a7 20 1Dcb6l:tc6 21 �)(e7+ !li.xe7 22 liJdS .i'.xd5 23 exd5 �c3 and Black is better, OJ!-Svesh­ nikov, Helsinki 1992.

9 iI.gS ••.

A standard strategic idea in these positions. Black exchanges his had hishop and weakens White's dark squares.

10 �e3 CLlge7 11 iLe2lbd4 12 0-0 0-0 13 il.d3 gc8 14 i.d2 g6 15 b3 f5 16 ttJxfS fbexfS 17 exfS gxfS 18 Jl.xgS 1!UxgS 19 f4 'IIi", 20 1JId2 (D)

2O b5! 21 �ael bxc4 22 .i.xc4 i.xc4 23 bxc4 ttxc4


tends to prefer slightly less fashion­ able lines than those played by most other Semi-Slav practitioners. For ex­ ample, in the Meran Variation after 1 d4 dS 2 c4 c6 3 lbf3 liJf6 4 tbc3 e6 5 d fbbd7 6 i.d3 dxc4 7 ..Itxc4 b5 8 �d3, the mosl popular moves are 8...a6 and 8....th7. Sveshnikov, however, has generally preferred 8 .. b4, a line which does not enjoy so good a reputation. Once again, however, he has his own ideas, and has been �uccessfuJ in up­ holding them in practice, a� in the fol­ lowing examples: .

Dreev - Sveshnikov Russian Cht (Podolsk) 1992

1 d4 d5 2 c4 e6 3 fbf3 Cbf6 4 liJc3 c6 5 e3 lUbd7 6 �d3 dxc4 7 .i.xc4 b5 8 �d3b4 (D)


...and Black won. Against 1 d4, Sveshnikov is another ex-Soviet player who has used the Semi-Slav all of his life. However, he

9 lLle4 c5 In more recent years, Sveshnikov has preferred 9...ltJxe4; e.g" 10 �xe4 'f!fb6 II 'iYa4 i.b7 12 4Jd2 1ic8 1 3 a3 (13 4.:Jc4 �a6 14 'itxa6 i.xa6 1 5 iLd3 c5 16 lZld6+ i.xd6 1 7 it.xa6 Itc7 is



equal, Baikov-Sveshnikov, Russian eh (Elista) 2001) 1 3 . ..c5!? 14 dxc5 .ltxc5 15 axb4 iLxM 1 6 0-0 lIc7! 17 i.xb7?! �xb7 1 8 thb3 0-0 19 litdl llb8 with an edge for Rlack, Yako­ vich-Sveshnikov, Erevan 1996. 10 thxf6+ gxf6!? 11 0-0 NCO suggests either 1 1 .te4 or 1 1

e4. lbe fonner proved unsuccessful in Lugovoi-Sveshnikov, Novgorod 1995: 1 1 i.e4:Ib8 1 2 O-Of5 1 3 i.c6 '!Wc7 14 d5 i£.g7 15 e4 fxe4 16 thg5 0-0 17 i.xd7 .ixd7 18 thxe4 C41 and Black's bishop-pair and queenside majority proved more important than White's kingside chances. 1 l...cxd4 12 liJxd4 'eb6 13 a4 as 14 i.b5 �b7 15 \WhS'!! lId8 16 lIdl (D)


16 ..:g8! 17 g3 ltg5! 18 lIYxh7 lXd5! 19 b3 .i.c5 20 i.b2 <l;e7 21 .:te2 tt:ie5 22 'iWh4 lixd4! 23 iLxd4 i.xd4 24 exd4 1!i'c6 25 dS llxdS 26 f3 �xdl+ 27 .i:txdl lDxf3+ 28 .ltxf3 'l'1'f3 29l:ld2 0-1 .•

White lost on time. His king is hunted to extinction after 29..,ti'ht+

30 'Wt>f2 't'kg2+ 3 1 Wei ft'gl+ 32 'it'e2 .lia6+ 33 Wf3 �hl+ 34 'it'g4 'ii'e4+ 35 <;i;>h3 iLfl+ One intcresting point about Svesh­ nikov's use of the Semi-Slav is move­ order. With a few exceptions, Sveshni­ kov has generally prCfell'ed the order I d4 d5 2 c4 e6 3 itJc3 c6. After 4 CLlO, Black has the additional option of playing 4 ... dxc4 ('If6 is the Semi­ Slav), leading to the complicated and very interesting AbrahamslNoteboom Variation, the main line of which runs 5 e3 i.b4 6 a4 b5 7 .1i.d2 .th7 8 axb5 .txc3 9 il.xc3 cxbS 1 0 b3 as I I bxc4 b4 1 2 i..b2 lfJf6. However, the draw­ back of Sveshnikov's move-order is that White can offer the sharp Mar­ shall Gambit with 4 e4. After the fur­ ther moves 4...dxe4 5 ttJx:e4 .1i.h4+ 6 �d2 'iWxd4 7 .txh4 'iixe4+ 8 �e2 practice suggests that White has good compensation. For this reason, most Semi-Slav practitioners (induding Mikhail Gurevich, discussed above) prefer to enter the opening via a Slav move-order: 1 d4 dS 2 c4 c6 3 lllf3 CLlf6 4 lbc3 e6. This allows White the option of the Exchange Slav with 4 cxd5 (or 3 cxd5), but although this variation is rather tedious if Black is trying to win, it certainly is nothing like so dangerous as the Marshall Gambit. .

Mark Hebden White: 1 d4 - Barry Attack, Torre At­

tack Black 1 e4: L.e5, Closed Lopez, etc. Black 1 d4: King's Indian



The tinal player whose repertoire we are going to examine is the Engli:->h GM Mark Hebden_ Mark plays virtu­ ally all of his chess in open events, and in particular in weekend Swiss events in England. When discussing Mikhail Gurevich's repertoire above, I made the point that success in open events requires the ability to score heavily against weaker players, regardless of colour. This is even more true in f\ve­ round and six-round weekend events, where a player can rarely afford to drop more than one draw if he i� to

6 Jte2 (Capablanca played the slower

cOllnt on first place. Hebden ha'> been

6 h3) 6. b6?! 7 QJeS ! �h7 8 h4!. Over

the most con�istent1y successful week­

the years, Mark has gradually refined


end tournament player in England over

this so-called Barry Attack to the point

the past 20 years, and his repertoire is

where he has been able to claim many

perfectly suited to such events.

grandmaster scalps with it, culmi­

Mark's greatest strength is his abil­

nating in victories against 10hn Nunn

ity to take relatively little-known and

in two consecutive Hastings Premier

apparently hanniess lines, and tum


them into something much more fear­

Hebden - Nunn Hastings 199611

some than they appear. In his youth, he did this with the 2 f4 Sicilian (now generally called the Grand Prix At­ tack), which had long been regarded as a quiet and tame variation. When he

1 d4 iDf6 2 "'f3 g6 3 1bc3 d5 4 -".f4 �g7 5 e3 0-0 6 �e2 cS

switched to I d4 in the late-1 980s, he

The slower 6...b6 allows White to

alighted on the line I d4 t;}f6 2 "Df3 g6

initiate a strong kingside attack after 7


lbe5 .i.b7 8 h4, one brutal example of

3 �3 d5 4 .ltf4

This had been known for decades

which went 8 .. .'!bfd7 9 h5 'Dxe5 1 0

t I dxe5 c5 1 2 \Il¥d2 e6 1 3

(Capablanca used it to beat Yates at

.i.xe5 ..t.xe5

New York 1924, for example), but was

hxg6 fxg6 1 4 ..tg4 'ue8 J 5 f4 1i'd7 16

generally played in quiet fashion,

0-0-0 liJc6? 1 7 QJe4

similar to the London System. Mark's

Likavsky, Cappelle l a Grande 1992.

innovation was to interpret the line

1-0 Hebden­

7 ll:leS

much more aggressively, often castling

This move was Hebden'g improve­

queenside and launching a kingside

ment over the older 7 dxc5, after

pawn-storm if Black develops too

which 7 ...4Jbd7! is a good reply.

slowly; for example, 4.. kg7 5 e3 0-0 .






One year later, Nunn tried 7...cxd4 hut lost to a neat tactic after 8 exd4 .!bfd7 9 liJf3 liJf6 1 0 Qk5 .!bfd? 1 1 liJf3 li:lf6 12 'iWd2 i.g4 1 3 tDe5 iLxe2 1 4 �xe2 lt'lh5?! 15 St..e3 qx6 16 0-0-0 J:l.c8 17 f4lLlf6 18 g4 �a5 19 a3 lLlxeS 20 fxe5 1itxc3? 21 cxf6 11fc8 22 Wbl l ltxc2? 23 litd2! 1-0 Hebden-Nunn, Hastings 1997/8.

8 0-0 cxd4 9 exd4 (D)


9...�b6 In Hehden-Farndon, Isle of Man 1999, Black preferred 9 .. .'BaS, but af­ ter 1 0 "@d2 .ttd8 1 1 a3 a6 1 2 �f3 �fS 13 gfe l 'ith6 J 4 lba4 �a7 15 lOxc6 bxc6 1 6 c3 J:.e8 1 7 h3 l2ld7 18 g4 .i:.e6 1 9 .ig2 1!fb7 20 b4 a5 21 .!Des White was well in command.

10 It'lxc6 bxc6 ll lba4 �a5 12 c3 ilJd7 13 b4 1!Vd8 14 �d2 e5 IS .Jth6 i.xh6 16 ifxh6 Ue8 17 :Mel nb8 18 dxe5 llxe5 19 'i'd2 a5 20 i.fl axb4 21 cxb4 'iWf6 22 lbe5 ttJxe5 23 neI 'li;g/ 24 1ilc5 White's grip on the dark squares and outside passed a-pawn give him a clear advantage.

Another home-brewed system which has brought Hebden great quccess is

his interpretation of the Torre Attack after 1 d4lDf6 2 ttJf3 c6 3 kg5 cS. The usual continuation now is 4 e3, hut af­ ter the sharpest reply 4...�b6 White is forced tochoose between the ungainly 5 �c 1 and the speculative pawn sacri­ tlce 5 .'bbd2. Neither of these lines suits Mark's style, and he instead pre­ fers 4 c3, after which 4...�b6 can be mel by 5 .'bbd2!, when it is very risky to take on b2, while 4.. .cxd4 5 cxd4 'm'b6 can be answered more naturally by 6 �c2. Even so, the Torre is hardly the most fearsome of openings, hut Hebden has shown that it possesses an unexpected drop of poison. A key element in Mark's repertoire is his clever use of move-orders. Every possible scenario and move-order is carefully worked out, and he is well­ nigh impossible to catch out . Since he always answers I d4 .'bf6 with 2 lLlB, he is able to avoid the Benko by play­ ing the line we discussed in Chapter :"i (1 d4 .'blo 2 "tJf3 cS 3 d5 b5 4 ..tg5), while against 2."c5 3 d5 e6, seeking II Modem Benoni, he has the option of4 tbc3, producing yet another position on which established theory has little 10 say, but which Hebden has played and analysed extensively, As Black, Hebden uses far feWCI home-made systems, generally stick iug to main lines after 1 e5 and in 11lL" King's Indian, However, his use III move-orders to keep opponents on balance can be seen here too. In 111< main-line Lopez, Hebden started 011 his career as a faithful practitioner (,I ..•


the Marshall Attack ( I e4 e5 2 ti.JD tbe63 i.b5 a64 .ta4 tbf6 5 0-0 .i.e7 6 �el b5 7 .ltb3 0-0 8 c3 d5), but in re­ cent year� has tended to prefer normal Closed Lopez lines. However, he still uses the Marshall move-order 7 ...0-0, thus forcing hi� opponents to consider whether they are prepared to allow the gambit. Many players avoid the Marshall with 8 a4 (or other moves, such as 8 d4, 8 h], 8 d3, etc.), but in


general these Jines probably offer White less advantage against a well­ prepared player than do the main lines of the Closed Lopez. Thus, by feint­ ing at a Marshall, Heixh:n is fre­ quently able to bluff his opponents into avoiding 8 d, and the fact that he does still occasionally venture the Mar­ shall proper means that opponents who are not prepared to face it cannot risk calling his hluff.

I ndex of Players

When the second-named player appear� in bold, that player had White. Other­ wise, the first-named player had White.

ADAMS - Dreev 1 1 3; Quillan 1 3 1 ADlANTO - Karpov 126




Khurtsidze 25

Dreev 1 1 3

ANDERSSON - Byrne, R. 37; Franco 40 ANDREEV, K.


Sakaev 6 1

ANDREEV, N. - Sokolsky 95 ANICHENKO - Sokolsky 90 BELIAVSKY - Stanee 1 2 BOTVINNIK- Matulovic 14; Tolusb BURGER


Alexander 3 1


BYRNE, R. - Andersson 37 BYVSHEV




DREEV - Adams 1 1 3 ; Anand 1 13: Sveshnikov 137; Svidler 1 14 EHLVE.'>T - Kramnik 128 FARAGO - Miles 46 FEDOROY - Shirov 5 1 FISCHER

Panno 1 20; Rubinetti 1 19; Uhlmann 1 5


FLOHR - Konstantinopol.,ky 35 FRANCO - Andersson 40 GELLER - Byvshev 17 GERUSEL - Miles 98 GUGORl('


Tukmakov 21

GUREVICH, M . - Short 108 HEBDEN


NUIlIl 1 3 9




Adianto 126; Kasparov 58; Short J 10; Timman 32 -

Karpuv 58; Miles 6 1

KHURTSTDZE - AJexandrov 25 KNOPPERT - Mikhalevski. V. 61



KONSTANTfNOt'OLSKY - Aohr 35: I.evenfish 34

KRAMNIK - Ehlvest 12&; Timman I I

EM. - Pillsbury 3 1 LAVTIER - Short 8 LEvENFISH - Konstantinopolsky 34 LASKER,

MATULOVIC - Botvinnik 1 4 MrKHALEvSKI, Y. - Knopperl 61

Mll.liS - Farago 46; Gt-rusel 98: lIIescas 100; Kasparov 61 NUNN - Hebden 139

PANNO- F"�her 120 PEI.TS - Sokolsky 97 PIIL'iBURV - La"ke... Em. ] 1 QUILLAN - Adams 1 3 1 RU6INE"!TJ - Fischer 1 1 9 SAKAEV


Andreev, K. 6 1

SAMARIAN ·- Sokolsky 96 SIUROV - Fedorov 5 l SIIORT - Gurevich, M. lOS; Kurpov 1 1 0; Lautier 8 SMACiJN - Sveshnikov 136 SoKOLSKY - Andreev, N. 95; Anichtnk.o 96: Pelts 97; Samarian 96; Strugach 96

STANEC - Beliavsky 12 STRUGACH - Sokolsky 96

SVESHNIKOV - OI"1!eV 137: Smagln 136 SVf()l.t;:R - Dreev 1 14

1'TMMA� - Karpov 32; Kramnik

TOLUSH - Botvinnik 1 7

TtJKMAKOV - Gligoric 2 1 UHLMANN - Fischer 1 5


Index of Openings

Caro�Kann Defence: B I O 35; B 16 33, 34; R 1 7 125 Dutch Defence: A90 67; A92 69 English Opening: A 1 1 70, 9J; A 16 39, 40; A 17 J28; A20 90; A29 74, 75, 76, 86, 90; A30 90 Flank Openings (Misc.): A(Kl 95, 96, 103; A04 120 French Defence: COl 108; CO? I U; C08 134; C 1 0 77, 78; C I S 14, 15, lOS, /06 Griinfeld: 07642; 087 21; DY4 42, 87, 89 King's Indian: E60 97; E90 62; E92 36, 37, f29; E94 55; E97 122 Modem Benoni: A65 65; A7066 Nimzo·Indian: E21 58; E35 /23 Open Games: C32 3 1 ; C39 51; C42 3{, 32, 80; C44 50, 105 Queen's Gambit: D23 63; D31 59, 138; D32 49; D33 130; D34 El7; 035 62; D36 10; 037 1 1 , 133; D43 [30; 047 23, 25, 137; 058 8, 57, 125 Queen's Indian: El5 126; E 1 8 38 Queen's Pawn: A40 46 47, 70, 98; /\41 101, 102; A43 60, 61; A45 48; A46 140; ,

000 48, 93, '39; 001 94 Ruy Lopez (Spanish): C67 43; C68 121; C78 131; e80 122: C88 141; C92 125; C94 125; C97 54 Scmi-Open Games (Misc.): BOO 100; B01 99, 1 14; B06 71; 807 70, ]01 Sicilian Defence: B30 136; B32 35, 136; 83363, 79; 840 120: 845 64; 863 110; 866 113; 870 Il3; B78 85; B87 119; 890 JJ8

How to build your opening repertoire  
How to build your opening repertoire