The big Dvoretsky interview, part 3

Today we present the third and last part of our big interview with Mark Dvoretsky, which we conducted during the Tal Memorial based on questions from visitors of this site. In this section we spoke about training tactics, the usefulness of endgame technique, aspects of computers and chess, chess in the future, the World Championship cycle and some miscellaneous topics. Enjoy!

On November 11th, 2010 we interviewed Mark Dvoretsky in the press room of the Tal Memorial, in the GUM department store in Moscow. In total we spoke about two hours on the different subjects that were suggested by our readers.

The first part was mainly about Mr Dvoretsky himself, and about books. In the second part we spoke about the young talents of today, the 'Russian/Soviet chess school' and strong endgame players, before moving on to the big subject of chess improvement. In this third and final part we spoke about training tactics, the usefulness of endgame technique, aspects of computers and chess, chess in the future, the World Championship cycle and some miscellaneous topics.

Please note that in the transcription we have taken the liberty to correct Mr Dvoretsky's English a bit here and there, but we tried to stick to his own words and sentences as much as possible.

The Dvoretsky interview, part 3

How should a player train when he is in the process of stagnation/bad form and the like? It's the same. People differ, so you should analyze the reasons of your troubles in chess and try to eliminate them. It's another principle, and the normal work of a chess trainer when he starts to work with somebody: start from diagnosis. He should understand the personality of his student and to decide what to do.

Do you recommend repetitive cycles of tactical problems, or that a student look at as many varied tactics as he/she can? Why does he speak about tactical exercises only? I prepared a lot of different kind of exercises for the development of various skills, not tactics only. And not only chess skills, but also psychological skills. Professional trainers should be ready to help his students in many areas. Nobody can be strong everywhere so; I told already about it. If the trainer has a wide repertoire, he can prepare in many areas.

But I think this specific question is basically: is it better to learn with specific problems of the same theme, or is it better to mix the themes? Again, there's no clear answer; it depends on the aims you have. For example, say, some player is always underestimated the opponent's counter chances. OK, for him it's natural to concentrate on this direction so for a trainer it's good to suggest him exercises which develop exactly this skill. But sometimes a player should just make a general training, for example prepare himself to be in good form before competition and in this case it makes no sense to concentrate on a specific topic. So there are a lot of situations and a trainer should select the exercises in accordance to the situation.

Every time a strong GM seriously misplays an endgame, e.g. Kasparov losing the R+4 v R+3 or Kasparov-Short (g9) where both players blundered in a R+2 v R ending, commentators say something like 'this shows the importance of understanding endgames'... This question is an example of wrong logic. First point: if a commentator says this, he's saying nonsense. Kasparov and Short made mistakes not because they don't understand the endgame; Kasparov was quite good in the endgame; Short perhaps as well. It's just because they blundered, so it's a tactical mistake and such a mistake can be made in any stage; in the endgame, in the opening, in the middlegame. So they missed a tactical opportunity; it's not a matter of lack of knowledge in some specific position. Kasparov against Piket is another story. OK, this is a theoretical position, it was analyzed by Averbakh in the past and so on. But it's impossible to remember everything. Kasparov lost this position not because he didn't know the theory, it wasn't the point. It was one of his drawbacks, and I wrote about this: Kasparov, when he was a defender, he was not a good defender. There were various reasons and one reason was that he had no good practice in defence. His opening were too strong to have enough bad positions and practice in this area. Another reason is that Kasparov always, just according to his nature, tried to defend by the most aggressive way. Very often this is the right and correct approach, the most promising approach, but not always. Sometimes it's necessary to keep patience, to wait, but Kasparov was unable to do it. He lost this endgame because he already had a good defensive set-up and he started to change the set-up. It was the same story when he almost lost the fourth game in his match against Kramnik. He already had a drawn position, a fortress, and then he started to improve the position of his rook. He wanted to bring the rook from one good position to another good position. Kramnik didn't exploit this and the game ended in a draw but the moment when he started this plan, Kramnik could have improved his position and win. So the problem is not endgame knowledge, but some features of his style. So it's not necessarily about mistakes based on bad endgame technique, or endgame knowledge and so on.



...But it seems to me these mistakes show exactly the OPPOSITE: they show that a player can become a very strong GM without mastering the endgame. Maybe mastering the endgame is much less important than other chess skills? As I told you Kasparov and Short are quite good in the endgame. One or two mistakes demonstrate nothing, absolutely nothing. We cannot decide that Kasparov is bad in the endgame because he made this mistake. Everybody can make a mistake, it happens. We can say it when we have analyzed many examples and it's typical for him, he made several of them, he 'usually makes a mistake in such a situation', but that's not true. So we cannot make a judgement based on a single example. Besides, top players of course differ. Some of them are great in the endgame, some of them are weaker. Not everybody is strong enough. But if a player is really weak in the endgame, it harms him a lot. I already mentioned the example of Bronstein, or Topalov, I can give more examples... Well, Tal wasn't good in the endgame when he was young. Fortunately for him at some moment players couldn't use it but in his second match against Botvinnik, Botvinnik used it several times. No, well, Tal was a genius, so sometimes a chess player can develop very well even without, but it's an exception. Basically a chess player shouldn't have a handicap in any area in chess, in the endgame, in tactics, in understanding, and so on.

We now move on to a number of questions about computers... A couple of questions are about the way of using the computer for a chess player. I can say that of course I use it but I'm not a great specialist here. Again it's better to recommend some better specialist. I very much like the book of John Nunn: Secrets of Practical Chess. It's a small book, but quite good. In the second edition of this book he devoted one chapter exactly to this topic so I believe it may be better to study Nunn's recommendation because he's a better specialist than me in this area.

Another kind of computer question is about how I use it. Well, I don't think I'm doing it too professionally but in some areas computers help. The computer helps me to check my exercises and improve them, but not to arrange my examples, exercises, instructive topics. On the other hand, for the last decades I haven't travelled with my students to the tournaments. Because of the computer and the Internet I could analyze their games during the tournament and when a student came after the tournament I was ready to discuss his game. It was by the way a funny thing, I was even a little bit surprised: I trained some strong players like Inarkiev for example, a strong grandmaster who played a game with another strong opponent. He perhaps discussed the game afterwards with the opponent and then he also checked his game with the computer, but still I was surprised that every time when we discussed a game I had many instructive moments which I could demonstrate him, which he didn't know about. So the computer helped me a lot when I prepared this analysis, but I didn't work too deeply, but still I could demonstrate some instructive thing. It means that some experience of a chess player, some general understanding, is also important. Of course without the help of the computer this kind of work would have been much more difficult to do.



Another thing that is asked is: what is the disadvantage of using a computer? It's also quite obvious. Most important is that if a chess player trusts a computer too much and takes the recommendation mainly from the computer, he doesn't develop the practical skill which he will need to use during the game. For example the skill to quickly see short tactics. The computer helps you to avoid this work. The computer demonstrates short tactics instantly, so when you watch in on the screen you don't need to think about it yourself so you don't train this skill. But in a practical game you need to do it every time, in every game, at almost every move. Without training you will do it much weaker. So you should somehow try to replace this work with the computer with some intentional training in this area; you should develop the skill to be a good tactician. Another thing is that a computer understands a lot of things, but of course not everything. Just recently, somewhere in May and June of this year, on the ChessCafe site my article about the positional sacrifice was published. [Download it here as a PDF - CV. Update: that should probably be this one.] I demonstrated examples of modern strong players who sometimes ignore, don't see or don't understand this element of the positional sacrifice, which was more obvious to the experienced players from the past. I gave some examples which I believe are convincing. Maybe the computer influenced this, because the computer has the tendency to underestimate the positional sacrifice. In the past people just understood; they analyzed a game and saw it's dangerous, that perhaps they shouldn't play such a position. But if the computer evaluated it profitable for you, OK you missed such an opportunity, you don't concentrate on them. So computers can also spoil some areas of chess understanding. A trainer who understands this problem can again eliminate these deficiencies by suggesting some instructive example in such topics, giving some good exercises and so on.

Which are, in your opinion, the most important developments in chess strategy and general understanding after the Kasparov era? How is the game improving? What have the world champions after Kasparov added to chess? There has been practically no general development. The computer became stronger and so the analytical devices work better, so the opening analysis became more massive and more deep and so on, but it's not a development of some general understanding, or something like this, nothing.

What do you think of Chess in future, say in 2050. Today calculators are being used for even simple mathematical calculations which were once done just mentally or by oneself. Will future generation think, 'my chess engine can play or solve chess then why should I spend so much time on it? I can use my creativity and intelligence for some other thing which a machine can't do!' About chess in the future I can mention something. At the end of 2007 and the beginning of 2008 on the ChessCafe site my big articles about the chess problems was published. [Download it here as a PDF: Polemic Thinking, Part 1 and Polemic Thinking, Part 2 - CV.] It's mainly about the very harmful influence of opening theory and some other aspects. Maybe somebody who wants to know my opinion about some important chess problems can read this article. I analyze the influence of opening theory, I analyze Fischerrandom/Chess960 and make another suggestion for how it's possible to change the chess rules and so on. I don't know whether it will be used in the future; people are very conservative. I believe that some problems exist, it's true, but it's better to read about it because it's not easy to explain. You know, in the past I wrote some big article which analyzed another big chess problem, less important but still maybe essential: the problem of short draws in unclear positions. I analyze this problem, I analyze various approaches to solve this problem and then suggested a solution which looked very reasonable to me and discuss this. It was exactly what we now call 'Sofia rule'. This article was published in several sources two years before the first tournament in Sofia. Danailov of course read this article - he studies chess sources - and he used the rule for the Sofia tournament. Of course he didn't mention me, the real source of this suggestion, and know this rule became more and more popular.

I think even in 2001 or 2002 the Corsica chess circuit started with this rule and also claim to have the 'copyright'... Maybe. I'm not sure, I don't know exactly about the Corsica rule but I have the impression that there is some difference with the Sofia rule. But Danailov imitated my suggestion mainly.

Who do you think is a stronger player if we have one person who is World Champion and other person who is No1 on ELO rating list? Which is the best way to know who is the strongest player in the world? What do you mean by 'the strongest player'? Today there's one strongest player, tomorrow it will be another strongest player, in one tournament one player plays stronger... We have some players of about the same level. Just recently Topalov was number one on rating, the Carlsen, Anand, again Carlsen, maybe Aronian will soon be number one, who knows. A chess player has better and worse periods in their development and it's sport. In one tournament one player can be stronger and in another, another. Following this reasoning, the World Champion is not necessarily the strongest player in the world; he's the strongest player in the World Championship. An honest World Championship with a normal system, without big privileges for anybody, OK the winner of this tournament becomes the World Champion, it's OK, he's the strongest player in this tournament. Maybe somebody else will win the next tournament, so what? We have it in any sport, it's normal. Sometimes big champions, who keep leadership for several years, and people respect them very much, but even such champions can lose their title, temporarily or forever, so... we shouldn't decide who is the strongest.

Do you agree that playing chess at the highest level (top sport, under pressure from the parents) for children aged five till seventeen actually is an unhealthy activity because the total surrender and fixation costs a lot of individual development and spiritual/physical joy? It can be dangerous. If the young player has smart parents and a good trainer he can develop reasonably and there's no harm. We have examples of both: chess players who received a very limited, one sided support and who have psychological problems, and on the other side very well developed, very smart, very interesting persons. So it depends on how a player develops. On the other hand, as was proven in many experiments studying chess helps to develop some essential, general skills for kids. They don't have to become a professional player but they get a new source of pleasure if they can play chess and if they can follow games of other players. Professional chess can be harmful, but with an inaccurate approach, a primitive approach.

Do you believe that blitz (or bullet) chess in prejudicial for one's game? Don't you think that it could also tell something about the person sheer chess talent, as we can see that the best classical time players often 'happen to be' the best chess blitz players too? I don't take blitz too seriously. Blitz is a pleasure; all chess players like to play blitz, on the internet, against some friends or in some tournaments. There's nothing wrong with it as long as a chess player doesn't do it all the time and has no time for other activities. To play blitz for pleasure is OK, and sometimes it's possible to use blitz for training, to play some kind of positions or some specific openings. But we cannot judge a player by blitz; it's not too harmful, it's not too useful, just some source of pleasure, sometimes for spectators also.



There's not necessarily a relation to the strength of a player in blitz and normal chess. Intuitive chess players are usually much more successful in blitz games, it's natural. On the other hand, in chess, generally, both ways of thinking, intuitive decision-making or logical, have equal chances to be successful. We had World Champions who were great intuitive players like, say, Capablanca, Karpov, Anand, Tal, and some other, and we have logical World Champions, like Botvinnik for example, or Kasparov. So all approaches are possible. Kasparov is a great player and also he was very good in blitz. Some players, like Grischuk for example, he's a great blitz player so we can suppose that he will play quickly in tournament games but in tournament games he always has horrible timetroubles. So there is no direct connection between blitz play and tournament play.

Maybe it works the other way around: he knows he will play good moves also in timetrouble so he knows he can think longer... Still, everybody understands that if he would have more time, he would be able to avoid some mistakes, so there's another reason, some psychological reason.

What is your opinion on Magnus Carlsen's decision to withdraw from the Candidates? You know, everybody can make any decision. I don't know his motivation, his real reasons and so on, so it makes no sense to discuss it not with Magnus himself. On the other hand of course this decision was made because he had some problems with the modern World Championship. It's true, there are really serious problems which are very interesting to discuss, but it's a big topic, a separate topic, perhaps we shouldn't do it now. For example he mentioned the great privileges of the World Champion - I absolutely agree with him. I know that Kramnik, Gelfand and some others disagree, Kasparov, Karpov. But many players agree with this position and I also agree. On the other hand he told that the World Champion shouldn't have any advantage, any privileges, and this is also wrong. When we play a World Championship it should be a system, not a single match or tournament, it's a system. So everybody starts at some stage and it's natural that some players came to the next stage by winning or keeping some results in previous stages and some of them get the right to play just because of their previous successes, it's absolutely natural. The win of the previous World Championship is also something we can consider the win of some previous tournament, so the winner should have some privileges, but of course not so fantastic as he has now. Also in the case of Carlsen: why should he play in the Candidates, he should start in the semi-final of the Norwegian championship, because maybe some younger generation can beat him. He should also play several steps and don't have privileges.

He got in because of rating of course... Rating is also a previous result, it's not 'this set-up of competitions for this World Championship', it's previous results, it's also a success like winning a previous World Championship, so it gives some privileges but not absolute privileges, like now. But it's a topic for a serious discussion and perhaps we have no time for it. Some other problems he mentioned are also connected to modern FIDE and their policy, their strategy... In many areas he is absolutely right - FIDE is a horrible organization now but again it's a topic for a separate discussion.

Which players that you either trained or not, of the top chess world do you consider that they are at least equal (or maybe even better) in natural chess pure talent to the very best (players that made it to WCH or the top ranks), but for other reasons (laziness, personality, etc) haven't or wont ever been able to be either WCH or even the top 5. In my view, the most notable example is my favourite genius Vasil Ivanchuk. Ivanchuk is of course a good example and there are many others. I will mention one of my former students, not so well-known like Ivanchuk, it was certainly my most talented student and he had all practical qualities to become World Champion: Alexei Dreev. I started to work with him when he was eleven years old. When he was young, during several years he had much better results than players of the same age, like Ivanchuk or Anand. He had much better results and he had perhaps even better prospects. But he didn't even became a player of the level of Ivanchuk, I don't speak about Anand the World Champion. The main reason was bad environment - bad family environment, bad environment in the city where he lived and bad environment in the country where he lived because a lot of harm was done to him by his country. This was the reason why he couldn't use his talent but when he started his quality was absolutely fantastic.

Looking backward into chess history, what do you think about Bobby Fischer's non-computer-era endgame technique in comparison to other world champions and modern top grandmasters? Fischer was fantastic in the endgame. He was fantastic in many areas but in the endgame he was absolutely great. I remember there is the story about when he was like sixteen years old, something like this, and in a tournament he played an endgame against Taimanov - bishop and pawn against bishop - and he knew how to make a draw. The analysis of this endgame was published in some magazine in the Soviet Union and he had studied this analysis. So he was fantastic in the endgame, he played brilliant endgames.

When learning from your book on endgames, should I use a board or try to do the calculations in my mind? What will improve my chess more? You can do both, you can analyze some complicated line at the board and simultaneously you can try to do something in your mind. Actually, to train yourself in blindfold chess, to see the board, I believe that it is important, I didn't do such kind of training myself but I know for example the opinion of grandmaster Igor Zaitsev, he believed that it's very important for a chess player and he recommended to train blindfold chess and to make blindfold analysis and so on. So I have no certain opinion about it but I believe that perhaps it's useful for a chess player. Logically it should be useful. I can recommend one good book, the book of Jonathan Tisdall, Improve Your Chess Now. There is a chapter in this book the development of this skill. I can recommend to read it and to think about it.

Links

• The big Dvoretsky interview, part 1
• The big Dvoretsky interview, part 2
• Archives of Mark Dvoretsky's articles at ChessCafe