The semifinal round of the World Cup had a somewhat surreal aspect to it. Looking at the games, it seems clear that the players were exhausted. The first two “classical” games were drawn in 14 and 16 moves respectively, as the players took a break from the pressure of the preceding rounds.
The second pair of classical games were also both drawn, although Vladimir Kramnik versus Maxime Vachier-Lagrave was a real fight. At one point the computer showed Kramnik as being +4, but even though it may be technically true, that assessment needs to be taken with a grain of salt. It was an extraordinarily difficult endgame of R+N+P versus R, where Kramnik’s one pawn was hard to defend. In essence, what Houdini’s evaluation meant was that by a long series of intricate tactical maneuvers, Kramnik could have held on to the pawn and won. Kramnik did make about three of the perfect moves, but went astray somewhere around the fourth. Really, from the human point of view, it was just unclear all the way. Vachier-Lagrave defended nicely, won the pawn and then suffered through the ordeal of the 50-move rule before he could claim a draw on move 125.
The playoff round took place today. The first playoff game between Vachier-Lagrave and Kramnik was as short and tragic as the previous one was long and thrilling. Vachier-Lagrave played the Scotch Game for, as far as I can tell, the first time in his life. Maybe he didn’t want to face Kramnik’s Berlin wall in the Ruy Lopez, but Kramnik doesn’t even play that so often any more. In ay case, Vachier-Lagrave’s choice of the Scotch was a disaster. It was more like the Botch(ed) Game, as Vachier-Lagrave didn’t know where to put his pieces and eventually walked into a combination that simply won an exchange. He resigned on move 22.
Memo: When you’re playing the most important games of your life, don’t play an opening that you’ve never tried before.
Actually, I was struck by how unusual this kind of game was for this tournament. Throughout the World Cup, if you tuned in around move 20 you would find that Houdini’s evaluation of most of the positions had barely budged from 0.00.
Out of curiosity, I looked over the entire archive of games with Houdini’s evaluations, to ask: How often does it happen that one of the players “loses the game in the opening”? I decided to define “losing in the opening” to mean that the other player got an advantage of at least 1 pawn (according to Houdini) no later than move 20, and maintained that advantage for the rest of the game.
In the first round, which had a greater disparity in strength of the players, 13 out of 218 games (6 percent) were decided in the opening according to this definition. That’s not very many, but the percentage is even lower in the later rounds. From round 2 on, only 8 out of 210 games (4 percent) have been decided in the opening. (By the way, Vachier-Lagrave-Kramnik does not quite count because Vachier-Lagrave made his losing blunder on move 21.)
Here are the eight opening debacles since round two:
- Andreikin-Nguyen (round 2) — White was +1 from move 15 on.
- Li-Giri (round 2) — Black was +1 from move 18 on.
- Vitiugov-Ragger (round 2) — White was +1 from move 18 on.
- Svidler-Radjabov (round 3) –White was +1 from move 18 on.
- Hammer-Kamsky (round 3) –Black was +1 from move 17 on. (White played an unsound piece sac.)
- Gelfand-Moiseenko (round 3) — White was +1 from move 19 on.
- Eljanov-Karjakin (round 3) — White was +1 from move 18 on.
- Caruana-Granda Zuniga (round 4) — White was +1 from move 20 on.
Interestingly, none of these except maybe Andreikin-Nguyen were early opening debacles; in all the other cases the crucial mistake arguably happened during the transition from opening to middlegame. In my opinion there has been only one game in the entire tournament (428 games!) where it looks as if one player may have fallen into a book trap. That was Adams-Wan from round 1, a Tarrasch French where Adams sacrificed a piece on e6 on move 12. I have a feeling that a sacrifice that comes that early must be known to theory, although I haven’t checked for sure.
Moral: At the elite grandmaster level, players just don’t lose the game in the opening. It sort of makes you wonder why players bother to amass this huge knowledge of opening theory, if the games are going to be decided later.
You could argue that the reason so few games are decided in the opening is because all of the players know the entire opening book. You could also say that the point of studying opening theory is not to get a +- advantage, but just to get a ± or +/=. Both of those points are probably true. However, I think it is even more true that the top players become top players by winning from equal or even inferior positions, and by grinding out the 80-move (not to mention 169-move!) endgames.
In the other semifinal, the “Cinderella vs. Cinderella” match of Dmitry Andreikin versus Evgeny Tomashevsky, “Cinderella” Andreikin won the second playoff game with a very late-developing kingside attack after move 40. So Andreikin goes on to meet Kramnik in the championship round. As I said before, this match could be a lot more interesting than you’d expect, because Andreikin has a positive record against Kamsky in their previous encounters.
Reader Juande Perea posted an interesting comment to my last entry. According to Chessbase, Andreikin has reached the championship round while actually losing rating points. The reason is that he has been playing the “Grischuk strategy” of drawing all of his classical games and winning in the playoffs. Because only the classical games are rated, he has lost rating points.
In his 12 rated games through the first six rounds, Andreikin won only once and drew 11. By the way, you might have noticed that I already mentioned his lone win — it was his game against Nguyen, where according the computer he was essentially winning by move 15! So his strategy so far could be summed up very easily — if you’re winning in the opening, great. Otherwise, settle for a draw.
However, I hope we will see a different strategy from Andreikin in the championship round. I hope he’ll resurrect his strategy from the 2013 Tal Memorial, where he drew everybody else but defeated Kramnik. He is now just one step away from completing the same feat at the World Cup.