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


Preface (Mark Dvoretsky)

Black has a decisive advantage.

We shall come back to this ‘opening scheme' later, but for the moment let us continue our discussion of how to construct an opening repertoire.

Sonic remarks on the technique for opening work

The most inefficient method - which I know that many of you use - is to write down opening information in a notebook. Nothing worse can be imagined! You fill up the pages with games and variations - then new games, fresh ideas, additional variations come to light, and you don’t know where to put them. Some pages of information turn out to be unsound - they have to be revised or even discarded, and of course you can hardly insert clean pages into your book. A kind of aversion gradually develops - you feel how outdated your notes are, and how inconvenient it is to write down novelties.

Ail your information, especially on the openings, should be collected in a card index. The cards can cither be miniature ones or large sheets. When necessary you can always write out a new card, affix it to any other one, throw away one that contains mistakes - in short, do what you like with them.

Another piece of advice is to leave wide margins; there will always be something to insert in them. Leave gaps in places where you feel something new will crop up. Write on one side of the card only.

Artur Yusupov, Sergei Dolmatov and all other chessplayers who work with me possess large sets of files containing analyses of various openings or even individual variations. But today, of course, even this way of working is a little old-fashioned. Obviously it is much handier to manage your card index on a computer. In this form it is always just as good as new. You can easily amend it, add to it, correct it. You can use a system of codes whereby everything is neatly classified. Computerised handling of opening information is a topic demanding special discussion; wc will not go into it just now.

‘Your own theory’

Let us suppose you have made a good choice of openings and your card index is organised faultlessly. It contains the latest games and excerpts-from specialised articles. You are thoroughly acquainted with it and have committed everything to memory. 'Now I'm bound to get a plus from the opening’, you think to yourself. But you are wrong - because to achieve realty good results, it is not enough to know the ‘official’ theory. What is essential (as Botvinnik once remarked) is to possess ‘opening theory of your own’.

It is very important to include in your repertoire some systems and variations on which your opinion differs, even if only slightly, from that of the theorists. This contribution ‘of your own’ may be a novelty which entails the reappraisal of a whole variation or rehabilitates a scheme considered bad. Or it may be an unconventional assessment of a familiar position. The position may be to your own liking although it is supposed to be none too favourable. You evolve a plan of action suited to it, and decide that you arc willing to play it despite its dubious reputation.

In general you need to be keen on your own pet lines, systems that you have analysed and have a feel for. A player who only knows what has been played before can scarcely count on success. He will never gain the advantage against an experienced opponent, since the latter will know it all too. But with the aid of ‘your own theory', you can be a step ahead of your opponent in the opening; you can put him in an uncomfortable position, take him into territory where he doesn’t understand what is happening.

How an opening repertoire expands

It is not often that a player just gives his head a scratch, says to himself 'How about studying, er, the Nimzo-Indian?’ - then turns to the Encyclopaedia and learns it up. Such things happen, but only rarely. Usually a new scheme or variation is taken into your repertoire as a result of some particular stimulus. To many young players this stimulus is supplied by their coach. He will say ‘I’ve got some good analysis on such-and-such an opening system. I’ll show it to you - you’ll beat everyone with it.’ This often proves useful. But do not make a habit of working this way. Sooner or later the coach’s slock of ideas will dry up, and anyway you will attain a standard where he can no longer help you; you will have to think for yourself. That said, the help of a coach may indeed give you some good ideas now and again.

When Valery Chekhov won the qualifying tournament for the World Junior Championship in 1975, it was clear that the openings he was playing would not do for the championship itself. He had no active systems with Black, and even with White he played all kinds of rubbish although he had been coached by an openings specialist at the Palace of Young Pioneers before he started working with me.

Recognising where his weak points were, and what problems needed to be solved, I invited Grandmaster Sveshnikov to one of our training sessions. The set of openings that he could show us was well known - he has been playing them all his life. With Black he plays the Lasker-Pelikan or Sveshnikov Variation; with White against the Sicilian he plays 2 c3. This was exactly what we needed - a system against the Sicilian and an active way of combating 1 c4. At that time there wasn’t a large mass of theory on the Sveshnikov Variation; the only other player who constantly used it was Ti-moshchcnko. Sveshnikov helped us to assimilate these two openings, and Chekhov employed them successfully in the World Junior. They became part of his repertoire. What is more, the notes made during that training session were useful to me in widening my own stock of openings. I later showed the Sveshnikov Variation to Artur Yusupov and Sergei Dolmatov, and they too played it for a while. In other words, a few hours' consultation with Sveshnikov helped to shape the opening repertoire of a whole group of players for some time.

Another example is Dolmatov’s preparation for the World Junior Championship in 1978. The situation was a similar one: Sergei didn't have a reliable system with White against the Sicilian. I couldn’t help him myself, since I didn’t play anything respectable against it cither -1 would just play ДЬ5 at the first convenient moment. Before the tournament we invited Grandmaster Tukmakov to a training session. Tukmakov is a connoisseur of the Sicilian for the Black side. For such a specialist to demonstrate the basic ideas of White's play is not too difficult. Our consultation with him proved exceptionally useful to Dolmatov. In the World Junior he played normal Sicilian lines, confronting the Schcveningcn Variation with confidence; ever since then he has constantly been successful with White in the main lines of the Sicilian.

It often happens that information is ‘in the air' and reaches us by chance. Once, while still at university, I went to the Institute of Physical Culture to hear a lecture by Grandmaster Razuvaev on the Exchange Variation of the Spanish Game. He demonstrated some recent games by Fischer and explained their basic ideas. I enjoyed that one-and-a-half-hour lecture so much that after looking at the Exchange Variation for myself I went on to win some good games with it.

So a hint from a coach or specialist may come in very useful and stimulate you to include an opening in your repertoire. It is easy to see why. When you begin studying an opening, you have a huge pile of material in front of you - large numbers of games and some pages of small print in the Encyclopaedia of Chess Openings. You don’t know how much of this is actually needed, what the main lines are and which lines are of secondary importance. You look at variations without knowing what lies behind them. If your coach can explain the main ideas and help you to make your selection, this is of course a major step forward.

But then, it is not just from your coach that you can expect help.

Working jointly with one of your friends is especially productive. You both have your own ideas and collections of information, and it pays to exchange them and analyse them together. The drawbacks to this way of working are always outweighed by the advantages. Of course you cannot play an opening variation against the same friend who analysed it with you; and if he is the first to use one of your innovations, you can no longer rely on it for surprise effect. So much for the snags. On the plus side - first, you are acquiring fresh information, and secondly, the openings you already play will be more thoroughly worked out. Ultimately you are competing not against your friends but against the rest of the chess world. You are disarmed against your colleague but better armed against all other players - and that is more important. The co-operation between Yusupov and Dolmatov, extending over many years, is a very instructive example. Numerous variations were worked out by their joint analysis. Some ideas of Yusupov's were taken up by Dolmatov and vice versa; as a result they both improved their opening repertoire.

So the second stimulus io expand your repertoire is the exchange of information with a friend.

A third type of stimulus is the analysis of games. It is precisely in this way that strong players discover the ideas that arc most important to them. At the previous session of our school, in the course of demonstrating one of his games against Karpov, Yusupov explained how the Open Spanish came to figure in his repertoire. As you may recall, he had analysed a game Karpov-Savon dating from 1971, and found an improvement for Black. This novelty spurred him to study the Open Variation as a whole.

And here is one other means of improvement. Choose a player on whom to model yourself, one whose ideas and style of play appeal to you. You can copy the opening repertoire of this player and study the systems that your hero plays.

Working with chess literature

Chess publications are a very important source of new information. You should regularly look through the periodicals, books and Informator. You never know when you will come across an idea that will later come in handy. Even old publications can help.

Here is an example. Many years ago I was studying a collection of games by Rashid Nczhmctdinov. It is a remarkable book, with very attractive games and some superb combinative play. I took a close look at one combination which Nczhmctdinov played against an amateur in a simultaneous display. I liked it, and incorporated it into my stock of chess exercises.

A few years passed. Something in the Griinfeld Defence caught my interest, and I opened the Encyclopaedia of Chess Openings. Suddenly I noticed that the opening variation of Nczhmctdinov’s game, in which he found a forced win for White, appeared in ECO with the opposite assessment - better for Black. I realised at once that this was a mine that could be laid to catch some assiduous reader of opening books.

Nezhmetdinov-Amateur Kazan 1951

Griinfeld Defence





















This very move was recommended in the first edition of ECO.

(In the second edition the mistake is corrected; 7...^d7 is now given as the main line for Black.)

In addition to this sortie, 8...a6 should be considered. A forced exchange of blows now ensues.

' 9 Va4!

What is Black to do? Nezhmetdi-nov has defended his bishop; after a queen exchange. Black loses a piece.

lO—’i^xal is met by 11 de.

White has already won a piece, since capturing the rock on al is wholly bad for Black.

In ECO this position is assessed as better for Black. The white rook is en prise. Do you see how Black answers the natural move 13 Ebl ? Quite right - 13...ШЗ+1! 14 £xd3 Axc6+, and Black emerges with an extra pawn. This occurred in a game Isakov-Nikitin (1947).

You may suggest 13 £if3. That is an interesting move. It is not considered in ECO, so let us take a look at it. What if Black captures the rook? White replies with 14 Sdl, or the even more accurate 14 Axd7+ Zxd7 15 3dl. Excellent - with 13 ?)f3 White has developed a piece and stopped the rook from being taken. What should Black play? Of course, 13...Wd3+! again, but this time White doesn't have to take the queen. After 14 Фе 1 ^c3+ he can repeat moves -which is something - but is entitled to play on for a win with 15 Ad2 Wxal-h 16 Фс2. Black cannot then take the other rook, on account of 17 £)c5. He must play 16...Ж>2, keeping e5 under control. Let us say White continues with 17 Sdl or 17 Zcl. Is his attack enough for the exchange? It would be interesting to ponder this position.

But before immersing yourself in complex analysis, you must always ask: ‘Have I missed something earlier, right at the start of my calculations?’ It makes no sense to study lengthy variations which the opponent can simply sidestep. In fact, after I3...’fird3+ 14 Фс1, Black has the excellent reply 14...^g71, which seems to refute 13 £lf3 outright.

I suggest you study the diagram position for yourselves and try to solve the problem: how did Nezh-metdinov win the game? (solution p.138)

I showed the opening variation to Yusupov and Dolmatov. It was of more interest to Yusupov, who constantly opened with 1 d4, than to Dolmatov who did so only from time to time. Well, here was this opening trap. But could it be used? There were two questions that first had to be answered.

To begin with - what if Black chooses a different move-order and plays 6...Ag7 instead of 6...С5 ? In this case White must cither prepare for the main lines of the Griinfeld or find some means of sidestepping them. We began analysing 7 ДаЗ.

Theory docs consider this move, and Dolmatov proceeded to win an excellent game against Bagirov with it. Still, the bishop on a3 somehow doesn't 'fit in’ .with the Griinfcld. We finally concluded that White can hardly lay claim to an opening advantage in this line. So it seems he cannot do without studying the ’normal’ Griinfcld lines.

The second question arises after 6...c5 7 ДЬ5+. What happens if Black avoids 7...£3c6 ?. According to theory, 7...4id7 8 £jf3 gives White the better chances. After all, what is one of White’s problems in the norma] Exchange Gninfcld? Black attacks the white d-pawn with his own c-pawn, his bishop and the knight on c6. White defends it with his knight and bishop. In principle he would like to have his knight on f3, but he then has to reckon with the pinning move ,.JLg4. Therefore While more usually develops the knight on c2. If it does go to f3, he usually plays Zbl first - not all that useful a move - so as to take the rook off the al-h8 diagonal. But after 7 ДЬ5+ £d7 8 43f3, White has no worries about his centre - his opponent can neither pin with ...ji.g4 nor bring his knight out to c6.

The most natural move for Black is 7...Ad7. After 8 £xd7+ «’xd7 9 $3f3, While has achieved something - he doesn’t have to fear a pin on his knight, and can solidly protect his centre. But arc these substantial gains? On the basis of some game played long ago, theory stated that Black had equality. We began studying the resulting positions and unearthed some ideas for White; we even concluded that he could count on an advantage. Then we came across an article in a foreign magazine, which demonstrated in detail that with precise play Black docs equalise. We couldn’t refute this analysis, and hence in the end we lost interest in the variation and its trap.

All the same, in the World Junior Team Championship at Graz in 1981, Yusupov lured Morenz into this very line and won exactly as Nczhmcldinov did. So studying that old book did bring some profit, albeit on a small scale.

More often, of course, new ideas come to you from more recent games and articles. Let me tell the story of another successful discovery.

In 1984 in Estonia, Aleksei Dreev and I were preparing for the World Junior Championship together with another participant in the tournament - Lembit Oil. Internationa] Master Nei, Oil's coach, brought to the training session a whole suitcase full of chess publications including many different foreign journals. I had not seen them before and started going through them at my leisure. In one Bulgarian magazine I found an article on the Exchange Variation of the Queen’s Gambit Declined.

This is one of various possible continuations, and is quite old (at present 11 h3 is more popular). The move was first played by Marshall against Rubinstein in the Moscow international tournament of 1925. Theory considers that it leads to equality.

When consulting an article I usually look at the writer’s conclusions first. In this case the author maintained that White can always acquire a plus in this line (rather an implausible conclusion, incidentally). I then turned to his main variation.

11 ...


12 Axc7


13 Axe4


14 £d2


15 ГЗ


16 £xf3


17 e4


18 Sxe4 (82)

This move is stronger than 18 £xc4 Af5.

18 •••             <=.ad8

19 Sfel

In the game already mentioned, Marshall scored a quick win with 19 3e5?! h6 20 £c4 Wb4? 21 a3! 'Йс4 22 tff2 Af7 23 ЬЗ! «ГхЬЗ 24

£fd2 ’tfa2 25 £c3, but Black could have defended better with 2O...‘«Zc7!, threatening 21...Sxd4.

19 ... h6

At this point the game Tal-Vagan-ian, Moscow 1975, continued 20 3c5 tff7 21 Йс4 3d6? 22 Wc3 £d7 23 3a5, and Black was unable to defend his pawn on a7 in view of the terrible threat of 24 Фе5 £xe5 25 Sxc5. However, not even this game proves that White has an opening advantage; by playing to simplify with 21...Ad7!, Black could have counted on equalising.

The author of the article suggested an interesting build-up for White:

20 Sle3J? ЙТ7

21 He!

White masses his forces in the vicinity of the kingside. He can avoid exchanges in the e-filc by occupying e5 with his knight when necessary, and then bringing his major pieces across to the adjacent kingsidc files.

The idea seemed to me to be promising from both the purely technical and the practical viewpoint. No one can follow every periodical; the article in a Bulgarian journal would be known in Bulgaria but might not be noticed elsewhere. So we would have some ideas that our rivals would not be familiar with.

Actually this variation had little to offer Dreev, since at that time he opened with 1 c4. I simply copied out the analysis on the assumption that it would come in useful sooner or later.

When preparing with Yusupov for the Candidates Tournament in 1985. I suggested that he should investigate this system. I showed him the variations given in the article, and he liked them. First we analysed the resulting positions, then we played a training match to a time control of fifteen minutes for the game.

Incidentally, I strongly recommend playing speed games to consolidate the opening information you are studying. (Of course, analysis should precede the games, and supplementary analysis should follow them.) They don’t take up much time, but they generally serve to bring new problems to light, even if the players have done some work on the opening already. In Yusupov’s view you should take alternate colours, facilitating a more objective view of the position.

In our series of games we un-_ earthed a large number of fresh nuances and acquired a much better feel for the opening than before. Yusupov incorporated the Exchange Variation into his repertoire for White. In the Candidates tournament he won a hard-fought game against Spassky, crushed Nogueiras, and had further success with (he system later.

Of course, Yusupov’s successes did not at all result from any special strength of the variation in question. On the contrary, wc came to the conclusion (which was not hard to predict) that Black can secure equality with precise play - just as in any other sound opening. The point was simply that wc were a little ahead of the opposition - wc had investigated the position more deeply and built up a stock of ideas that were generally unfamiliar.

Since Artur doesn’t just play the White side of the Queen’s Gambit, our analyses also helped him to strengthen his defences for Black. In particular, playing Black in the eighth game of his Candidates Match with Jan Timman, when it was imperative for Timman to win, Yusupov played a move wc had prepared in our training session -18...h6!? (instead of 18...3ad8) -and easily equalised.

Both White and Black

We have briefly discussed the basic principles for constructing an opening repertoire. I shall now describe one particular precept in greater detail. It will find favour with players of the white pieces who don’t go all-out for the maximum - for an advantage by any possible means - but merely seek to play ’their own’ game, their own type of position.

It sometimes makes sense to play with White a system that suits you with Black - that is, to approach the White side of an opening by playing Black with an extra tempo.

Shortly after taking up my effective plan for combating the Closed Sicilian - a plan which brought me wins in the majority of games - I naturally conceived the idea of using the same formation with White. What did this entail? Obviously I had to take up the English Opening. Of course, in that opening Black has a wide range of systems at his disposal, and White must be prepared for any of them. But if Black incautiously plays 1 ...c5,1 have the chance to draw him into my favourite scheme.

Grandmaster Razuvaev once told me about a specific method he employs in his (highly productive) work on opening theory. He always selects and concentrates on certain key games which he treats as paradigms of the opening he is studying. They must be games in which both players (or at least one of them) played logically, constructively, and in which valuable ideas and typical resources were employed. Such games enable you to understand the opening formation more deeply and commit it to memory more easily.

You may regard the following game as a paradigm of its system.

Dvorctsky-Timoshchenko USSR Team Ch, Moscow 1966

English Opening

Black plays a Closed Sicilian with colours reversed.

An interesting point - I consider Black’s last move rather weak. In this variation, the knight is better placed on f6 or even h6. The reason is purely tactical and not at all obvious. It will come to light a few moves later.

Remember how we answer this?

Of course! We have to prevent ...d6-d5.

I can now explain why the knight on c7 is worse placed than on f6 or h6. Black is at present unable to play ...Ah3, for after exchanging I take his pawn on c7. With the knight on f6 or h6, the move ...Ah3 would be possible, since after the exchange on h3 White couldn’t take the pawn in view of ...^g4 forcing mate. A small tactical detail with a great deal of significance. If Black could exchange the light-squared bishops unhindered, he would stand quite well. Incidentally, the position with the knight on f6 ought to be familiar to you from my article The 'Superfluous'Piece, published in Training for the Tournament Player.

A game Dvoretsky-Veselovsky, Moscow 1967, went 10...a5 11 a3 Eac8? (Black should not just abandon the quccnside to its fate) 12 b4 ab 13 ab €)d8 14 Ь5 c6 15 be be 16 Фхе7+ Sxe7 (83).

83 W

What should White do in cases like this?

13 e4!

A standard plan - rearranging die pawns on light squares once the light-squared bishops arc off.

13 ...         Axg2

14 <±>xg2      f5

15 f3

I think White's position deserves preference. He has more space, and the black bishop is confined by its own pawns. If Black undermines the centre with ...c7-c6, the reply is ФсЗ; then at some future stage White will exchange on c6 and meet ...b7xc6 with b4-b5, securing the central outpost d5 for his knight. Black for his part can station his knight on d4, but the two outposts are unequal in value. A knight on d4 can be attacked by White’s bishop, while the black bishop will not be able to exchange itself for the white knight. Here we sec the advantage of the ‘good’ bishop over the 'bad' one.

15 ... ФГ7 (84) next move. I had only just gained my Master title at the lime, but I rate it as a true Grandmaster move. I found it by a purely logical process, and we will now examine this logic.

A little exercise for you to work out for yourselves: What is White's most accurate continuation from the diagram? (solution p.138)

11 b4 £xd5

1 l...c6 is answered by 12 Фхс7+ 'Й'хс? 13 b5. After a pawn exchange on c6, the bishop comes out to a3 and the queen to a4, etc. White has an easy, comfortable game.

Black’s last move leads to a closed type of position.

12 cd Ah3

84 W