Candidates Round 4

Live video on YouTube

Anand – Kramnik 1/2-1/2 (1.42.52)
Karjakin – Topalov 1/2-1/2
Mamedyarov – Andreikin 1-0
Aronian – Svidler 1-0

I only have two small points to add to today’s games. The first is in Mamedyarov – Andrekin, which was essentially decided by time trouble. Obviously it is stupid to be down to seconds at move 37, but still this happens to all of us at times.

The main point is that the position after 37.Rd8+ is a classical example of elimination, the main defensive calculation technique:

Andreikin chose wrong. He should have played 37…Kh7 with more or less even chances, while after 37…Kf7 38.Qd6 Qa6, White does not only win back his piece, but won the entire game after 39.Rd7+!. Had the king been on h7, this would not have been possible. Andreikin had seen this, but said he had missed that 39…Kg8 loses to 40.c8=Q (as well as everything else). It is all a bit bizarre to me, but in time trouble people can often get confused.

In Aronian’s game Svidler could early on have gone for a slightly worse opposite coloured bishops ending, as he mentions in the video. This would not have been as simple as some would think.

But more importantly, I think he made a big mistake at this point:

Rather than suffering in an unclear position with 34…Kf8, which is without doubt easier to play for White, Svidler played 34…Qd7 to go into an opposite coloured bishops ending. He is only somewhat worse, objectively. But I have noticed that when we have positions with opposite coloured bishops, the stronger player scores much better than he otherwise would. If you look at Kramnik’s games in London 2013 and also his game with Karjakin here, his only game with White so far, he aims for positions early on in the game for positions with opposite coloured bishops. Carlsen also has a fantastic score in endgames with opposite coloured bishops.

Obviously, with this in mind, it is quite interesting that Svidler and Aronian both praised 34…Qd7. In my experience things that hang on very long calculation in defence are very unreliable. This is what Svidler was counting on. Do not miss the press conference; the player’s comments are very interesting.

So, although it looked dangerous to play on with an extra piece with queens on the board, probably this was statistically the better choice. If nothing else, it also gave some chance that he would win the game as well.

31 thoughts on “Candidates Round 4”

  1. Hello Jacob,

    I guess one interesting thing is why Peter Svidler did not go for Nd4 on move 27 when according to the commentators this was a safe way for a draw rewarding a good preparation.

    Was there any reasonable chances to play for a win with the choice he made ?


  2. Well, according to El Svid himself, it was because he was out of his preparation and he thought that the ending reached by 27…Nxd4 was significantly worse for Black and the game unclear. I doubt Jacob is going to be able to improve on that explanation.

  3. Hello John,

    Correct, I missed the press conference and I replayed the end of the broadcast where indeed he explains that .. So altogether my initial post can be removed :).

    That being said, Peter Heine Nielsen made the interesting comment at that point of the game that he would go for the Nd4 line because the other options looked really hard to play as Black before discussing the engine suggestion.

    The conclusion of the tournament round is that Aronian is back on track and unfortunately for Svidler he is not rewarded in terms of points for the spectacular chess he is displaying.

  4. I must say that the press conferences have led me to find Svidler quite “sympa” as the French put it.

  5. “In Aronian’s game Svidler could early on have gone for a slightly worse opposite coloured bishops ending, as he mentions in the video. This would not have been as simple as some would think.”
    I don’t see any problem for black at all there, what am I missing?

  6. Someone has to ask dumb questions. Can someone explain the compensation Aronian “saw” with 22. Bxf7+?? OK there are white’s two center pawns, BUT if 27…. Nxd4 ( which he would have “seen”) this seems a slim thread given black’s a & b pawns. Also, 25. Qxe6 did not seem bad with the Knight snatching the h7 pawn possibly. How many people would have considered Qxf7+ seriously?? Yes, a check to be looked at— but a hard sac for a pretty complex/ uncertain continuation.

  7. @Jacob Aagaard, I would like to ask about Svidler’s decision-making in this game and how that fits in the scheme of things which is your ongoing series on GM Prep. Chessvibes has this statement: “Svidler summarized the game as follows: “A very interesting and theoretically important game I think. I seem to be choosing wrongly as in when to go for a principled fight in this tournament.””
    This type of decision-making, it is not purely a calculation, or is it? strategical? Is it a totally different decision-making/skill? will you be treating it in “Thinking Inside the Box”?
    It seems to be a type of decision-making which is very important in one’s success and improvement as a player.

    Perhaps similar is Anand’s decision in the game against Kramink in the same round on move 21, whether to play Nb3 or Nc3. Are they same/similar decisions or different?

    1. This is quite an interesting question and very difficult to answer. TITB will deal to a great extent with simple questions with simple answers, though it will also go a bit beyond this.

      Svidler decided to take risks and then decided to try to force a draw by long calculation. He was in my opinion over-reliant on calculation, which is probably a common weakness for him.

      If we look at the 2013 candidates in order to make a plan for the tournament, we saw that Kramnik drew his first 7 games, but still could have won the tournament at the end. If we have learned anything from it, it is that you should not get too desperate early on. Both Kramnik and Aronian ruined their chances there from losing games, rather than from not winning enough games.

      I would go out on a limb and propose that Peter did not think about a tournament strategy in advance and for this reason was deciding a bit randomly what to do in the situation, rather than going for draws against 3-4 players with Black and then just play the rest of the games. Obviously he could have had any tournament strategy he wanted, but it was important to have one, in my opinion.

  8. >Both Kramnik and Aronian ruined their chances there from losing games, rather than from not winning enough games.

    I’m not sure about that! Kramnik lost one game playing va banque in the last round. You might just as well say that if he’d won more game earlier he wouldn’t have needed to play the Pirc in the last round. Let’s face it, losses with this opening shouldn’t really count for rating purposes.

  9. @John Cox
    You could rephrase it to: Kramnik did not ruin his chances by playing 7 draws; which is essentially the same argument, and it won’t be open to the argument you have here :-).

  10. Thanks Jacob. I know you are busy editing and producing books for us chess books addicts …… But if I can follow up with with another question ……

    I hear what you are saying about a tournament strategy and I agree.
    However, I would like to come back to Svidler’s thinking (really interested in this). You said he was probably over-reliant on calculation. Do I take it that you think this is a thinking which is not calculation?I don’t know what it is and would like to know. More importantly is a question, can a chessplayer trains for this?

    Phenomenologically, I see it as partly calculation, ie, you calculate the candidate variations, you pause and you evaluate the position (whether you use Tisdall’s stepping-stone (great book! btw) or otherwise) and you keep doing this to the best of your ability. At the end, you have to evaluate which variation you would like to proceed. Here I suppose the tournament strategy as well as the game strategy has a place.

    Does the above phenomenological description makes any sense or is it mere fantasy?
    Eg, the game today between Kramnik and Aronian and most commentators say that if Kramnik had more time, he might have been able to see the win (35 Rg1) but he was short of time. In the press conference, Kramnik said (chessvibes’ paraphrase): ” Kramnik had in fact seen 35. Rg1! Bxh6 36. Rde1! Re4 about which he said “maybe it’s winning for White but it’s not so simple. I was really running out of time and my pieces are so bad that I didn’t like it.” ”

    [Evaluation of position whilst calculating a variation, do you think it is more of an intuitive thinking, ie, intuition based on thousands of patterns and positions one has played/learned in training, ie, Kramnik saying he does not like Black’s two bishops and the pawn on e2.]

    If it does sort of make sense, then I suppose part of the training is the ability to calculate and evaluate positions at certain length from the current position?
    The part of the training is, I suppose, mental training? The literature I am reading in this area calls it, mental toughness. As a chess trainer, do you think this is what is missing from most chess training books/programmes or coaches/trainers?

  11. @weng
    There is a book I am not allowed to mention by John that says that modern chess consists of opening preparation, intuition and calculation. Furthermore it refers to a study stating that grandmasters know about 100,000 games (or positions; sorry, cannot be bothered to look it up).

    Certainly there are a lot of people who play like this; also strong grandmasters. But it is by no means the deciding factor.

    At the Olympiad I had a longer chat with Ponomariov, who had trained with PRACTICAL CHESS DEFENCE with before playing really well in Dortmund a few years back. His opinion was that it was very helpful, but the effects were not permanent, which I think not many would expect. After all, solving exercises is like going to the gym for a chess player :-). I would have loved my work in the gym last summer to have an effect now; but sadly it does not seem to be the case!

    Ruslan thought that the main strength of the top players was positional; not calculation. The pattern is actually far more fragmented than that if I have to give my opinion. First of all there are different types of positions. We know that Anand with the initiative plays very easily. He hardly calculates, he simple sees all the options. But Anand in a worse endgame sees nothing. Surely this is not lacking skill in calculation; and indeed, neither case has anything to do with calculation! Moves that appear in your head without effort are not calculated.

    Then we have general principles; finding out the mechanics of the position we are looking at, evaluating long term compensation, defining weaknesses and targets, schematic thinking, finding out what pieces to exchange, evaluation of instabilities. The list goes on.

    Calculation surely is an act of thinking. But to call thinking an act of calculation is nonsensical. I know it has been done, but it is a sort of violation of language and makes it difficult to understand what is really going on and how you can advance various skills.

    About training material: One place to start is Grandmaster Preparation – Positional Play. Look for example on the 3Q blog post from two days ago to get a hint of that technique. But it is really just one of many.

  12. At 6:07 (and a few seconds) you can see a few moments of non-calculation thinking from both players. (

    Kramnik evaluated the Bxe4 dxe4 d5 sequence as winning. He asks what the computer was saying.

    Aronian did not fear it because he believed that in such positions he would advance the pawn and bring in the rooks (as in the game).

    The computer does not give White a winning advantage.

    Aronian says it looks very strong.

    The first and the last of these are intuitive statements. Guesses based on optics. Bringing the rooks in is sort of a schematic plan. The computer evaluates the position based on a numbers system, which I would not recommend copying – as well as maybe a billion possible positions (most of them ridiculously irrelevant).

    This is at a moment where variations are being exchanged, rather than actual thinking being performed. And still they are exchanging other considerations to a great degree. Then look at Svidler at the press conferences: all moves, moves, moves. Maybe he relies too much on brute force!? At least this was my suggestion, though I hope I was careful enough to say that I don’t know, of course.

  13. Not sure if this is the right place, but my post is related to today’s round. I was very surprised to see Aronian’s 32.a4 where I was certain he would go 32.Ra4, subsequently trying to establish a6+Bb7, instead of moving the rook to the side and immediately afterwards exchanging Bishop for Knight?! Any thoughts on this, anyone? Is my understanding of the endgame that poor? 🙂

  14. Now I’m flabbergasted, at least I suspected Aronian to prove me wrong and show some Rook Engame Technique++ which I’d never seen before, but he just drew! A pawn up with R+B against R+N!

  15. How did Svidler choose h6?? It was such fun to see the Leningrad Dutch—but it did not end well. What to make of Kramniks’ walk off, reportedly sans handshake, skipping news conference with Topa?? I must admit to prejudice—I am not too fond of the Topa/ Cheparinov/ Danilov clan. I am open to being corrected on this.

  16. I’m amazed by the way that Svidler didn’t play his beloved Grünfeld. Maybe he was afraid of 3.f3 as are some visitors of this blog?

  17. @John Cox
    I am pretty sure the feeling is mutual. I found the Elista match far more complex morally, but John has banned me from bringing it up on the blog permanently.

  18. Thanks Jacob for your reply. Give lots to think about. Much appreciated. My interest is very much in the insights in training for excellence and results and I have enjoyed your posts very much. I have read quit a number of training books which have been published. Looking forward to your final book in the GM Prep series.
    Re the book you are not allowed to mention perhaps I will send you an email.

  19. I don’t know under which topic to put, so I’m putting it here: yesterday I received the latest issue of New In Chess magazine, and to my pleasant surprise I saw a nice training column by Jacob on improving your worst placed piece, including three exercises. I like it a lot! @ Jacob, will this be a regular column or just this one time? I hope the first!

  20. @weng siow
    Happy to reply to things in private. I should say that I have great respect for most chess authors and that author in particular. Unfortunately, when I disagree with ideas he might have, it is taken as an insult and I want to avoid that.

  21. @Ray
    I am back with an unorthodox book review in no. 4. I am not sure beyond that. They seem interested to keep this a regular thing, but they also demand a lot of me and I am very busy :-).

  22. @Jacob Aagaard
    I understand – by the way, very interesting book review by Sadler of Playing the French (and Judith Polgar part two as well). He is even sharing some further analysis on the Tarrasch with 12…h6 with the readers, nice stuff! And Flear’s review on Playing the French and Berg part 1 in NIC Yearbook is also very positive 🙂

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top