help analyzing GM's game

10...0-0 : if 11.exd6 ? then 11...Re8 ends up either winning the queen, or having a very strong attack after 12.Be3 Nf4. I wouldn't play it though, but I guess some analysis could convince you the piece wasn't free for White.

Yet the sacrifice is unsound, as the following of the game shows.

The line you mentioned for move 11 for Black gives him indeed 'one more pawn'... but one less piece - Black wants pressure along the e file.

Move 26, Black resigns because White has one more rook, the attack, and wins the Ba2 after ...Ke7 Qh4+.

Irontiger, you have too much respect for grandmasters (or maybe they weren't grandmasters at the time?). 11 exd6 is perfectly fine; true, black has ...Nf4, but it doesn't do anything concrete. I looked at it with a computer, and it pretty much seems like, after white defends g2, black doesn't actually have an attack -- how many threats could black really create? My inability to find so many is what leads me to believe that life is not so difficult for white. Don't assume that just because a very strong player played a move, that it must be good.

Hey, in a real game I would probably just take the piece and run away. But I'm by far not sure to survive. I would probably not go for 11...exd6, but White's way to take the piece in the game looks fine. BTW, players from that time are well-known for unsound sacrifices, but also for not being so bad at calculation.

I looked with the analysis board on it, and the least I can say is that I wouldn't calculate it through on the board in a 15 min game.

In your line with 13 Qf1, and 17 Kd2 (0-0-0 is better, but Kd2 will also work) Be4, white can just play 18 Qg3 and he's up three pieces for a rook and pawn, and none of white's pieces are dropping off. Moves like 18...Rad8 for black are too slow to be effective; white can reply 19 Rg1, and can follow even with Nd4, when suddenly the bishop is also attacked. Another obstacle for black is the discovery potential of the e5 knight: although it may not work at every move, black constantly has to be worried that something he does might suddenly allow the discovery to become effective. There are way too many distractions.

In your second paragraph, you are suggesting that it would be sacry to think about taking the piece, even if it works. Maybe it is scary, depending on the player, but objectively the sacrifice is plain bad! Even if your opponent is a grandmaster, always go with your gut -- if you think your opponent is bluffing, don't assume he isn't just because he's strong -- such players do bluff sometimes! Challenge him on it; call his bluff! Even if your gut is wrong, you will learn so much from it because you will get to directly see the refutation of your judgment, instead of just assuming it was bad. If you just get scared because your opponent has a few ideas with an otherwise questionable idea, you're letting him control you, and even when the winning move is in front of you, you'll be too scared to play it.

I hope you don't take any offense to the above; it's so far from my intention. It just irritates me a little bit sometimes when people are partial to a strong player, and as a result don't give that player's moves quite as much scrutiny as they normally would. Every player's moves deserve scrutiny!

I am far from taking offense from correct remarks ! Your analysis is more accurate than mine, and the sac could be accepted with your line.

What was my point is that even if Black is indeed throwing away a piece, it's not obvious at first sight, at least for me. Hell, I missed a move in the 7-move away line on the analysis board, so how could I ever be sure of my calculation if I had to do this in my head in limited time ?! I'm more or less forced to compensate my lack of calculation skill (which I am training in unserious games) by the 'feeling'.

I played a couple of masters (mostly in simul) and each time I got faced with an unclear looking 'bluff', I first looked the lines other than 'calling the bluff', and if none of them was looking great (with a strong advantage), then I jumped in the 'gutsy' stuff. Of course this is part of the 'simul' tricks - forcing the master to calculate a lot by playing the lines he hopefully will not have looked at.

But precisely here, even in 1 v 1, it cannot be bad to first take the g6 knight by 11.Bxg6, that's why it's the move I would play. And again, I would probably have settled for taking the piece after that, but it looks way safer then.