16.Bh5 Rh7 17.Qg4+ Kh8 18.Qxe6 wins the e6-pawn, but at the cost of dissipating White's initiative. Black would then have the chance to activate his bishop by 18..c5 with good prospects, since 19.d5 Re7 costs White the e5-pawn.>

Nunn, "Understanding Chess Move by Move"

<Hesam7> Nunn considers the piece sac entirely sound and 11..a6? a mistake that allows it: <Anand falls in with Kramnik's plan and soon finds himself in a critical position.> Instead of 17.Ne4?! he suggests 17.Qc2 as <a dangerous alternative, after which Black would face serious difficulties. 17..Rh7 is bad, since after 18.Qg6+ Kh8 19.Bxe6 Qg5 20.Qe4 White avoids the exchange of queens, and without his e6-pawn Black is bound to be struggling. One line might be 20..Rb8 21.Rf5 Qe7 22.Bxd7 Qxd7 23.Raf1 Kg8 24.e6 Qd6 25.Rf7 Bg7 26.Kh1, when White has a winning position. Other moves are little better; for example, 17..Bg7 18.Qg6 Re8 19.Ne4 or 17..Qe8 (preventing Qg6+) 18.Ne4! Rh7 19.Rf6 and again the e6-pawn falls, since 19..Re7? loses to 20.Nd6.>

17..Rh7 is given a ?! mark while he gives detailed analysis of 17..c5 that seems to lead to a draw.

Instead of 19.b4? (<This is the wrong move. Kramnik reasons that Black's only way to free himself is to play ..c5, so he attempts to clamp down on this move. It is certainly an unexpected move, but in fact it allows Black to slip out of the net. In a situation where one has a strong bind, it is often hard to decide between slowly trying to increase the pressure and cashing in with immediate action. Here White should have chosen the latter course. Perhaps the main defect of the text-move is that it doesn't succeed in its main ambition. Black can often play ..c5 despite White's pawn grip, since the activation of the light-squared bishop is usually worth more than a pawn. Moreover, if White meets ..c5 by bxc5 then Black obtains connected passed pawns, while White's pawn-chain e5-d4-c5 can be blockaded on the light squares>) Nunn suggests <19.axb5! cxb5 (if 19..axb5, then 20.Ra7 is good for White) 20.Nxb7 Rxb7 21.Rxa6 Rb6 (21..Nb6 22.Rf6) 22.Rxb6 Nxb6 23.Rf6 with a very unpleasant position for Black, since White obtains a third pawn for the piece and acquires two connected passed pawns in the centre of the board. The game Ward-Grabliauskas, Copenhagen 1998, played shortly after Kramnik-Anand, tested this assessment.>

Some more of the annotation, very far from all since virtually every move is commented:

23..Ba8! <The strongest move. Black frees his rook to move to f8, and preserves his bishop ready for the eventual breakout with ..c5. This is in fact the last passive move Black plays in the game. Move by move, the circumstances for a breakout have been gradually improving - Black has organized his kingside pieces and activated his dark-squared bishop, while White has made scant progress.>

25..c5!! <A brilliant idea, the point of which is revealed next move. Black cannot play 25..Be3 as 26.Rxf8+ Kxf8 27.Qh8+ wins at once.>

26..Bd5! <Remarkable. Black simply ignores the fact that White has taken a piece and focuses on the real priority: supporting the crucial e6-pawn. 26..Rxg5 is bad as 27.Bxe6+ Kg7 28.Qh4 leaves Black in a deadly pin, and White wins after 28..cxd4 29.Bxd7 Rxf1+ 30.Rxf1 Qxe5 31.Be6! After the text-move, Black is genuinely threatening to take the knight. White decides simply to retreat it, but then Black aquires two connected passed pawns on the queenside.>

27..cxb4 <The material balance has shifted so that instead of being a piece down, White is now a pawn up. However, everything else has changed in Black's favour. All his pieces are now on active squares, he has two well-advanced connected passed pawns and White's pieces (especially his queen and the h3-bishop) are out of play.>

17 Qc2 is a dangerous alternative - with three(!) pages of analysis, concluding "Thus 17 Qc2 would have been good for White, but there is no reason to criticize the text move."

19. b4? "What on earth is this move, I hadn't even considered it. After the surprise faded. I realised that unless Black takes drastic action he is going to be squashed. Incidentall, it is difficult to imagine that this pawn move is going to be the cause of White's defeat!" Anand recommends 19 axb5!

25 c5! (without further comment)

26 ... Bd5!! After the text-move the bishop finally gets out. Unbelievably, Black is already better. Vladimir was short of time and now missed his last chance.

27 Nf3? 27. Bxe6! is recommended instead.

My Best Games of Chess (Anand). Nunn contributed to this book as well, so the coincidences in text/analysis may be unsurprising.

27. Bxe6 Bxe6 28. Nxe6 Qxe6 29. bxc5 b4 30. Qe2 Rxf1 31. Rxf1 c3 32. Qd3 Qd5 33. Qxa6 Qxd4 34. Qe6 Kh7 35. Qh3 Kg8 36. Qe6

Indeed.

Semi-Slav Defence (D43)
(Notes based on Anand’s annotations in the Informant CD)

1.Nf3

A flexible move characteristic of Kramnik. It reserves all options for White.

1...Nf6 2.c4 e6 3.Nc3 d5 4.d4 c6

The Semi-Slav Defence.

5.Bg5 h6

Heading for the Moscow Variation.

5…dxc4 6.e4 b5 7.e5 h6 8.Bh4 g5 9. Nxg5 hxg5 10.Bxg5 Nbd7 is the Botvinnik System.

6.Bh4 dxc4 7.e4 g5 8.Bg3 b5 9.Be2 Bb7 10.e5 Nh5

This move gives White a free hand. 10...Nd5 is preferable.-NSH

11.a4!

The tempting 11.Nxg5 is good for Black. 11...Nxg3 12.Nxf7 Kxf7 13.fxg3 Kg8 14.0–0 Na6!.

11...a6 12.Nxg5! 12...Nxg3 13.Nxf7 Kxf7 14.fxg3 Kg8! 15.0–0 Nd7 16.Bg4

16.Bh5 is innocuous on account of 16...Rh7 17.Qg4+ Qg5.

16...Qe7 17.Ne4

17...Rh7!

Not 17...c5? 18.Nd6 Bd5.

Now the tempting 19.Rf7 fails to 19...Qd8! 20.Qc2 Qg5 21.Bf3( 21.Rd7Qg4 or 21.Raf1 Bxd6 22.exd6 Qxg4 23.R1f4 Qg5 24.Rxd7 cxd4) 21...Bxd6 (21...Nxe5 22.dxe5 Bxd6 23.Rf4) 22.Rxd7 Bxf3 23.gxf3 Bf8.

So White should play 19.Bf3!Bg7(19...cxd4 20.Bxd5 exd5 21.Qg4+ Bg7 22.Rf7± ) 20.Nf5± !( not 20.Bxd5 exd5 21.Nf5 Qe6 22.Qg4 Rh7).

18.Nd6

If 18.Nf6+ Nxf6 19.Rxf6 Bg7 20.Rxe6 Qg5.

18...Rb8 19.b4?

Missing 19.axb5! cxb5 (19...axb5 20.Ra7±) 20.Nxb7 Rxb7 21.Rxa6 Rb6 22.Rxb6 Nxb6 23.Rf6± (Kramnik).

If 23...Qd8 24.Bxe6+ Kh8 25.Qg4! Bg7 (25...Be7 26.Rg6 Bg5 27.Rg8+ Qxg8 28.Bxg8 Kxg8 29.h4 ) 26.Rf7 .

Or 23...Kh8 24.Bxe6! 24...Bg7 25.Rg6±.

19...h5!!

A brilliant counterattack. Now Kramnik does not get a second chance.

20.Bh3

Not 20.Bf3 h4 21.g4 Bh6.

If 20.Bxh5 Qg5.

20...Bh6! 21.Kh1 Bg5 22.Qc2

22.Ra3 c5! 23.bxc5 Bd5 is good for Black.

22...Rg7 23.Qe2 Ba8! 24.Qxh5 Rf8

25.Ne4

25.axb5 cxb5! - Anand.

If 26.Rxf8+ (26.Rxa6??Rxf1# ) 26...Nxf8 27.Rxa6? Bd5–+ -NSH

25...c5! 26.Nxg5 Bd5!

The greedy 26...Rxg5? fails to 27.Bxe6+ Kg7 28.Qh4± !.

Or 28...cxd4 29.Bxd7 Rxf1+ 30.Rxf1 Rxe5 31.Qxd4 .

27.Nf3?

White could have put up more resistance with 27.Bxe6+!, although after 27...Bxe6 28.Rxf8+ Nxf8 29.Ne4 Rh7 30.Qd1 (30.Qe2 cxd4) 30...Rxh2+ ! 31.Kxh2 Qh7+ 32.Kg1 Qxe4 , he would still have lost.

27...cxb4 28.axb5 axb5 29.Nh4 Qg5 ! 30.Rxf8+ Nxf8 31.Qe8 Rf7! 32.Nf3 Qg6

Alert as ever. 32...Qe3? 33.Bg4!.

33.Qxb5 b3 34.Rf1 Qd3 35.Kg1 Qe3+ 36.Kh1 c3

36...Rb7??37.Bxe6+ .

37.Bxe6!

A last brave attempt.

37...Bxe6 38.d5 Rxf3! 39.gxf3

Or 39.Rxf3 c2–+ 40.Qf1 Qxf3 41.gxf3 b2.

A picturesque position.

Analysis Diagram: after 41...b2

39...Bh3 40.Qc4 Bxf1 41.Qg4+ Kh7 42.e6 Ng6

0–1

Hopefully a preview of the Bonn WC match this fall. :-) >