Samadov vs Karpov, 1966

There is a natural tendency to change pawn structures...that has to be resisted. One of the differences between a master and an amatuer is the experience that tells a master when he holds a superior pawn structure...

I was just playing this game with Guesss the Move feature and when you begin to think on every move of the game, you're starting to feel it's flow. I of course didn't find the 19.f3! move and was amazed what was the reason of it. Only later during the game I began to feel how strong it is, as whites now can freely move his N, R and Q without watching for that e4 pawn.

Kasparov's <18.Nh2!?> and <19. f3!> were an important innovation, killing Black's counterplay against e4.

<19...Qd7?!> 19...c5! looks stronger, immediately challenging White's central structure. Kasparov: "This move was very bad. It's from another planet. He needed to attack the center, but this move did nothing. Then with each of his next four moves, he worsened his position." (from Fred Waitzkin, Mortal Games)

<24...Ng8?> This is just bad. 24...Nxg4 25.hxg4 Be6 is better for White, but Black is still very much in the game.

<25.Bxh6!?> This is a strong sacrifice, but the simple 25.e5 was also good for a big advantage. Kasparov: "I felt it. I didn't calculate, I played it by intuition."

Commenting live on the game, Yasser Seirawan said the sacrifice was one of the worst oversights of Kasparov's career. (Waitzkin) This only proves Seirawan doesn't understand Kasparov's chess.

15..bxa4 certainly does look ugly the knight gets b4 (for the moment) but white has ready made pressure against the a-pawn and it will probably fall.

20...Qb5 also seems bad, taking the queen from the kingside where all white's pieces are heading.

23..c6, maybe planning 24..d5 is too slow with the pressure white already has.

<24...Ng8?> This is just bad. 24...Nxg4 25.hxg4 Be6 is better for White, but Black is still very much in the game.>

I agree. I think AK did not want to exchange knights when GK's Nc4 can replenish it in a couple of moves, whereas black Nb4 is very restricted by his own pieces and a long way away.

25. Bxh6 is a classic GK attack, seizing the moment and exploiting the dynamics of the position. However, my chess computer (20 MHz 32-bit) sees it after only a couple of seconds (8_2) and believes it cannot be accepted.

<Commenting live on the game, Yasser Seirawan said the sacrifice was one of the worst oversights of Kasparov's career. > LOL he was never really up to commentating on AK or GK's games and that proves it! Although I enjoy his colourful comments on other games.

The question though, is the position after 30..Qxe8 lost by force? We need Rybka analysis for that! There is chaos in the black camp and not even AK's legendary defense can stop the white attack after a few moves.

I think so. I know Kasparov thought so- and Rybka assesses it as a full pawn better for White.

<Kasparov's 25. B:h6 is a very deep sacrifice: in the end Karpov has small material advantage (2 knights vs. a rook and a pawn), but open d-file kills him because white rook and queen can use it virtually unopposed.> Indeed - as well as the generally much better coordination of White's pieces, and the weakened position of Black's king, which succumbs to the second wave of attack, led by f4-f5. If Black tries to stop the invasion on the d-file by 31...Bd7, 32.Rc5 seems very strong - winning the a5 pawn with Black gaining no counterplay.

An alternative elegant way to win on move 36 (instead of Qf2) would have been Rg3! followed by Rg5 (the rook is immune - 36...Qxg3 37.Rh8+! Kxh8 38.Qxh6+ Kg8 39.f6; or 37...Kg7 38.Qxh6+ Kf6 39.fxg6 fxg6 40.Re8).