I watched the World Championship playoffs from 3 a.m. to 8 a.m Chicago time.: exciting stuff!
Viswanathan Anand first won the unified World Championship in 2007: this is his third successful defense (Kramnik 2008, Topalov 2010, and now Gelfand 2012). Understandably, Boris Gelfand's Chicago fans must be very disappointed, his childhood teachers Tamara Golovey and Leonid Bondar chief among them. But Tamara can be very proud of her student's playing Anand on an absolutely level footing.
The third game of the playoffs was incredibly bizarre, even if one overlooks the tactic Gelfand missed on move 26. Here's the rook ending:
The PGN has been cut and pasted from the
official website, with Black's 51st move corrected
per Dennis Monokroussos (and confirmed by me from
rewatching the video).
(Hmm: the official website appears to be wrong in another respect. Ian Rogers includes a repetition of position on moves 55-57: see below.) Black's 51st move was indeed the blunder 51...Kf5? and not the correct 51...Kf4!: had 51...Kf4 52.Rc8 been played, I can't imagine Anand missing the elementary intermezzo 52...Rc2+ 53.Kg1 Ke5.
It's amazing (and amazingly painful) to see a supergrandmaster blunder in this position:
What in the world was Gelfand thinking when he played
61.59.Rh7?? (My best guess is that Gelfand was trying to prevent Anand from playing ...Kb7 and setting up the
Vancura position with a subsequent ...Rg5+, ..Rg6, and ...Rc6. But there's no time for this: White's king simply runs immediately to the h-pawn and relives the Rh8 of guard duty. You can check this yourself with the
Shredder endgame database. And as long as White has the rook on h8 and is threatening to push the pawn to h7, Black's king can't come any closer than the c-file because of the beginner's skewer trick (
explained by Matt Pullin here).
Of course, when one is playing on a ten-second increment, as Gelfand was, there isn't any time for "thinking." What a painful way to toss away the win! (Time pressure blunders in world championship matches are nothing new, of course; and the mutual blunders in this rapid game are not at all representative of the overall high level of play in this match.)
If world champions and their challengers can get confused in positions with only five or six pieces on the board, we should be more forgiving of our own errors. (This past weekend, I managed to lose a queen for rook and pawn on the White side of a Catalan in only
eleven moves: hmm.)
Chess is hard.
***
P.S. May 31
Gelfand had more time than I thought, and the hallucination was mutual. GM Ian Rogers
quotes Anand in discussing the above position:
Despite having built up almost a minute on the clock through four quick
moves, Gelfand returns the favour. The obvious 61.Kg3-g4 wins, whereas
the text move (i.e., 61.Rh8-h7) is too slow by one tempo. “I thought I would get a Vancura
position,” said Anand, “but I don't.”
And the official website appears to have an incorrect game score in another respect, omitting a repetition of position: the actual final moves were
55.Kg2 Re3 56.Kh2 Ra3 57.Kg2 Re3 58.h5! Re5 59.h6 Rh5 60.Rh8 Kxc6 61.Rh7 Kd6 62.Kg3 Ke6 63.Kg4 Rh1
