Opening Preparation

Mark Dvoretsky and Artur Yusupov

With contributions from:

Sergei Dolmatov

Yuri Razuvaev

Boris Zlotnik

Aleksei Kosikov

Vladimir Vulfson

Translated by Joint Sugden

В. T. Batsford Ltd, London

First published 1994

© Mark Dvoretsky. Artur Yusupov 1994 Reprinted 1994, 1996 ISBN 0 7134 7509 9

British Library Cataloguing-in-Publication Data. A CIP catalogue record for this book is available from the British Library.

Ail rights reserved. No part of this book may be reproduced, by any means, without prior permission of the publisher.

Typeset by John Nunn GM and printed in Great Britain by Redwood Books, Trowbridge, Wilts for the publishers, В. T. Batsford Ltd, 4 Fitzhardinge Street, London W1H0AH


Editorial Panel: Mark Dvoretsky, John Nunn, Jon Speclman

General Adviser: Raymond Keene OBE

Commissioning Editor: Graham Burgess


6 Building an Opening Repertoire

Mark Dvoretsky

There arc many different approaches to working on your opening repertoire. It is an individual matter, and every chessplayer has his own guidelines. I hope that those I am going to describe will nonetheless also prove useful.

Which openings to include in your repertoire

Your choice of openings should depend first and foremost on your own taste and style of play. This rule may sound obvious, but all the same it is quite often broken, even by strong players.

When I was teaching at the Institute of Physical Culture, one of the students, a Candidate Master whose tournament results were rather poor, showed me some of his games. He was a quiet, sober-minded young man, and I was astonished that he played sharp openings like the Sicilian and King’s Indian. Why was this? The answer turned out to be simple. In the Moscow Palace of Young Pioneers he had belonged to a group whose coach was keen on fashionable opening lines. In other words, the young player’s choice of openings had depended not on his own taste but on that of the coach. I advised him to change his repertoire - in particular, to switch to I d4 with White. Scon afterwards his results improved, since he started playing his own kind of chess.

That was an example from the experience of a Candidate Master. But it seems to me that an entirely similar mistake was made by Grandmaster Mikhail Tai in his preparation for the return match against Botvinnik. In their first match, Tai had had some problems on the White side of the Caro-Kann, even though he achieved a plus score from that opening. In the return match he resolved to ‘stun’ Botvinnik with the Advance Variation which at that time was rarely used. From a strictly technical viewpoint he may have prepared it quite well. He did, in fact, have some interesting ideas in this system, yet he registered a minus score with it: one win, several draws and two losses. The explanation is simple. Tai had an excellent understanding and feel for open positions with lively piece play, but e4-c5 leads to a closed position of the strategic type. In such a game Botvinnik had no trouble finding his bearings, whereas this was not at all Tai’s forte. He went straight into

112 Building an Opening Repertoire battle on Botvinnik's territory, just where the latter was more sure of himself. This being the case, opening preparation was rather beside the point. You may obtain a promising position, but if you have little Bair for the type of game in question, you are quite likely to make mistakes. That is just what happened - Tai repeatedly failed to make the most of his position.

It may sound like a platitude, but openings do have to be studied in accordance with your own tastes. Another point is perhaps a little less obvious: in constructing an opening repertoire, you need to take your powers of memory into account. I know chessplayers with brilliant memories (Gavrikov, Balashov and many others). It is worth their while to make use of this asset and equip themselves with complicated modern opening systems where there is a great deal of theory - where you have to know an immense quantity of games and memorise various refinements, since if your opponent knows a little more, you may suffer a disaster in the opening. There arc plenty of such variations - for example the very sharp Najdorf Variation of the Sicilian. In openings like the Griinfeld and King’s Indian, White has an extremely wide choice; he is the one who determines the opening formation, and Black has to be prepared for everything. You can only play such lines with Black if you have a good memory.

Another tiling that players with good memories can usefully do is to expand and diversify their repertoire. They can afford to play a variety of openings, since they are able to absorb and memorise them. This enables them to select whatever system is most unwelcome to a particular opponent.

For players with less good memories it is dangerous to follow the same path. I know from my own experience what an excruciating labour it is to memorise ‘theory’ before a game. You have it all written down in notebooks, you have gone through it ten times before starting play, and still you can’t remember it. If this is so, it may well be better to concentrate on what I would call ‘opening schemes’ - logical systems with a smaller amount of theory, in which it is more important to understand the position and know about typical ideas and resources than to memorise specific details and precise move-orders.

In general, openings might be formally divided into ‘opening variations’ and ‘opening schemes’. Of course this classification is relative, since the theory of any opening involves both a set of exact, concrete variations and elements of logic and planning - the question is only how these factors intcr-rclate. Thus, with a good memory, go ahead and learn ‘opening variations’; with an indifferent memory, concentrate on ‘opening schemes’.

Example of an ‘opening scheme’

A long time ago, when I was still a ‘first category’ player, I was interested in the question of how to combat the Closed Variation of the Sicilian Defence. People are sometimes worried by strange problems! How to play a normal Sicilian with Black -1 evidently knew that.

My coach at that time was A.Roshal, and I have to say that he was a good one. He is now a well-known journalist, though unfortunately somewhat lacking in principle. In one of his coaching sessions he demonstrated a system of play against the Closed Sicilian. I liked the system and it seemed logical. 1 saw that it was not only suited to that particular opening but would work against a number of analogous formations by White - for instance the King’s Indian Attack. In other words, this scheme is fairly flexible and versatile. I recommend it to you too; you will not regret it.

Right away I started employing this new plan.

Gorodilov-Dvoretsky Leningrad 1964 French Defence

Black’s plan is suited to many contexts, including the Chigorin Variation of the French.

Immediately after the game I discovered that in such situations Black has to reckon with the positional threat of c4-e5, as played in the beautiful game Pctrosian-Pachman, Bled 1961. (This game, as well as some others I shall mention, can be found in the Appendix to this lecture.) But when you are just starting to play a new system, there arc many refinements you don’t yet know about. Deep understanding comes with practice.

I had played this way in similar situations myself. I would retreat the knight, push my pawns to f4 and g4, and expect to give mate shortly. I was particularly interested in how Black can resist this sort of attack.

The advance of the b-pawn is Black’s basic plan. He creates counterplay on the queensidc.

Strategically I believe Black virtually has a won game. A short while ago the bishop on g7 faced a solidly defended white pawn on c3, but now it is pressing against a weak, vulnerable one. The other bishop has also occupied an excellent diagonal, attacking the pawn on d3. Black controls the open b-file and will increase the pressure with ...,Й‘а5. His well thought-cut scheme for deploying his forces has enabled him to work up a queenside initiative in a short time.

What is White to do? Look how flexibly the black knights arc positioned. They defend each other and at the same time the knight on c7 is controlling f5. Black has to give close attention to White's possible thrust with his f-pawn. If this breakthrough is threatened (for example after g3-g4), Black will prevent it with ...f7-f5. In so doing he will retain control of all the central squares; his position will remain solid and supple.



My opponent wants to develop his queen’s knight, but comes up against a tactical stroke typical of this set-up.'

14 ...          ФЬ4!

15 cb       Axal

There is no need to show any more - Black is a sound exchange up. and won easily.

My first outing with this opening scheme had been a success. Things went equally well in the following game.


Moscow 1964 French Defence



















By this time, as you can see, I understood the position better and didn’t allow e4-e5. It was Cicero who said that anyone can make a mistake but only a madman persists in his errors.

The position now resembles a Closed Sicilian (since the knight is on c3), but it is more comfortable for Black than that system normally is, since White doesn’t usually block his f-pawn with his king’s knight. In addition, it is hard to make sense of the move 'S'e? - it serves no purpose for White.

Black's plan is the same - to extend the diagonal of his king's bishop by advancing the b-pawn.

10 #62(78)

Here is an interesting problem



which is important for the whole variation. White probably wants to play ДЬб. Black can preserve his strong bishop from exchange by playing !0...Se8 and meeting 11 ДЬб with 11...ДЬ8, but a different reaction is also possible; Black may allow the exchange, then rearrange his pawns on dark squares (,..e6-e5, ...C7-f6) and play to confine White’s light-squared bishop which becomes ’bad*. Both strategies are feasible, and in each game Black needs to consider the specific reasons for adopting one or the other.

Why is this move dubious? At that time I didn’t realise that Black must watch out for 11 d4 as well as 11 Ji.h6. After the opening of the d-file, the pawn on d6 becomes vulnerable. In such cases Black usually needs to prevent d3-d4 by playing ...$5d4.

White wants to seize the centre with c2-c3 and d3-d4. But this plan is too slow; it would only make sense if Black had no time to ‘fasten onto' the pawn on c3 with his own b-pawn.

11 ...


12 c3


13 d4


14 be


15 Sei


16 cd


Black’s pieces arc very harmoniously placed - much more so than White's. Just as in the previous game, the black bishops arc sweeping the whole board, the rooks control the b- and c-filcs, and the white pawn on d4 is weak. After 17 ’^xa5 £jxa5, the black knight would go to c4.

17 £c3      Scc8

18 Seel     Wa3

Threatening to invade on b2 with a rook.

19 Sabi £>a5

When you are sure of your plan you don’t even need to think much -all your moves are natural. The game plays itself; Black could even have played this way in a blitz game.







22 £kll

Better is 22 £lb5.

22 ...


The harvest begins.



Ox a 2














White resigned since he is losing a second pawn.

In demonstrating this game I have not gone deeply into any variations. The first reason is that both opponents were only in the ‘first category*. Concrete details and nuances are best studied in games by stronger players. But secondly, we arc examining an ‘opening scheme* rather than an ‘opening variation*. In a case like this, the detailed analysis of variations is less important to us than the outline of the game, the plans of both players, the typical devices for conducting the fight

This comparatively easy method of play brought results not only in junior contests. I later continued to employ the scheme successfully against some very strong opponents.

Bronstein-Dvoretsky Moscow 1976

Sicilian Defence

1   e4

2  £c3

3  g3

4 Ag2

5  d3

6  Ae3

Why not 6...£d4 instead? That was what Denker played against Smyslov in the USSR-USA match in 1946. Take note, however: the black knight should occupy d4 only after the white king’s knight has moved to f3 or e2. White answers 6...Qd4?l with 7 Фсс2! followed by c2-c3 and d3-d4. The variation 7...Фхе2 8 4ixe2 ДхЬ2 9 Zbl favours White, since 9...'Й'а5+ 10 £d2 Wxa2 fails to 11 мхЬ2! &‘xb2 12 ДсЗ.

7 f4        £gc7

8 £f3

Now which move, 8...0-0 or 8...£ki4, do you think is more precise? We have already observed that Black has to take d3-d4 into account. It is not always dangerous, but it is better to prevent it all the same.

8 ...            £ld4!

9 0-0        0-0

One of the basic positions of the Closed Sicilian has been reached. White has a number of continuations: 10 Wd2, 10 £f2. 10 Ebl; even the pawn sacrifice 10 e51? has been played. In this situation, general considerations arc not enough; knowledge of the detailed theory is indispensable. You can easily find out about it for yourselves if you want, but at present we have other aims.

10 g4

What should Black play now? It would be very dangerous to allow a pawn sacrifice with f4-f5, breaking up the pawn cover in front of the black king. The standard response to such a threat, as we have said before, is the counterstroke ...(7-f5.

10 ...            f5! (79)

But how is Black going to take back on f5? In this kind of position there is a rule of thumb: recapture with the opposite pawn to the one your opponent captured with; in other words answer g4xf5 with ...c6xf5, and e4xf5 with ...g6xf5. I

к Aa к*

JXi 4 AA


sr амяд .■_• в мая д:дм вия to ишядэд

don’t know how to explain it logically, but my experience with the system tells me that this rule generally applies.

11 gf cf

12 ■•Hidl

In this position the standard plan of ...Zb8, ...b5 and ...b4 is rather slow. Black needs io complete his development and go into action in the centre. What is the best square for his light-squared bishop? In such positions it often develops on e6 (after a preliminary ...&h8, so that £jg5 may be met by... Д g8). But I decided to place it on c6 to oppose the white bishop on the long diagonal.

12 -       £d7

13 ttT2

White finally evicts the knight from d4.

13 ...         £xf3+

14 ДхГС     Дсб

Black wants to prepare ...d6-d5 (perhaps by playing ...b6 and bringing his queen to b7), after which he would win the battle for the centre.

15 Wg2 b6

16 Sadi Wc7

Black intends ...’ЙЪ7, ...Sad8 and ...d5.

17 £c2

Of course, taking the pawn on b2 is bad; White would answer c2-c3 and set about trapping the bishop. He new wants to play c2-c4 after first protecting the b-pawn. I guessed this plan and prepared an antidote.

17 ...          Sac8

18 Acl       ЙЪ7 (80)

Black’s position is preferable; his pieces arc exerting troublesome pressure on the opponent’s centre.

80 IV


а идйгщ


19 с4?!

Positionally this move is justified; it prevents the advance ...d7-d5, and prepares to bring the knight to d5 via c3. However, in carrying out his plan, Bronstein has underestimated the flank diversion I have prepared. It seems he should have played 19 £jg3, to which Black would reply 19...d5 20 c5 d4 with somewhat the better chances.

19 ...          >a6!

I have a special name for this kind of move - 'the strategic dual attack’ The pawn on a2 is threatened, but in addition Black wishes to exchange pawns in the centre. If White takes on c4 with a piece, my knight will acquire the tremendous square f5. If he takes with his pawn, I shall capture on c4.

20 ^c3 fc

21 de

Nevertheless 21 £}xc4 was better. In my opponent’s place I would not want to part with the pawn.

21 ...


22 ФЫ


23 Zfel?!




25 b3


26 £g3


Again Black has set up that battery on the long diagonal which he possessed a few moves earlier. Against the natural 27 Zxd6, I had prepared the highly unpleasant counter-blow 27...£}f5! (28 Sxc6 ®h4l).

27 “gl?

He should have supported his bishop with 27 Zf 1. Now I start forcing events to my own advantage.

27 ...          ^f5!

28 Qxf5     HxfS 29 .£b2 Zf7 is also good.

























37 Zfl <±>xg7