Maybe time trouble, or just a blunder, which happens to everyone?

After <37 ... axb5> black can't hold onto the pawn for long. White plays <38 Rc1> and the threat of Rc5 means either the b-pawn or e-pawn will soon drop.

Eyal: Position after 40...Rxb7:

click for larger view

An example of 4 vs. 3 rook endgame which is quite difficult to defend due to Black's weak pawn formation, but perhaps Boris could have still saved it by activating his rook after 41.g3 with 41...Rb3, e.g. 42.Kg2 (42.gxf4 Rxf3 43.fxe5 Kg7 44.Rf1 Rh3!) Rb2+ 43.Kh3 Rb3 44.Kg4 h5+ 45.Kxh5 fxg3 46.hxg3 Rxf3, or 44.Rf1 fxg3 45.hxg3 Kg7 46.Kg4 Kg6 47.Rh1 (otherwise ...h5+) 47...Ra3.

<Eyal> It seems to me that your analysis after 41....Rb3 offered better practical chances of saving the game. The game continuation was straightforward enough that even a weak player like myself could have found it.

Also, as he mentions, <14.Ne2!?> is a very interesting move in the opening (offering the pawn on e4 - which Gelfand didn't accept)- instead of moves like Ng5, Nd4, and Qe2 which have been analyzed to death. Eljanov said he saw it played in a few correspondence games, so strictly speaking it isn't a novelty, but it's a new idea in super-GM practice. Second time in this tournament (after Aronian's 12.Nh4!?) that Gelfand faces a tricky novelty which he doesn't manage to handle well.

11 exd5
12 exd5 Nxd5
13 ?

See something that violates our rules? Blow the whistle and inform a moderator.