<33. Kd1 Kf7 (33... Rc4 34. Kc2 Kf7 (34... Rxa4 35. Kxc3 and 36.Kd3 is bad) 35. Kd3 Rc8 36. Rxc3 Rxc3+ 37. Kxc3) 34. Kc2 Nc5 35. Rf1+! Kg6 (35... Kg8) 36. Rf8! Rxf8 37. Bxf8 Nxa4 38. Be7!>>

Not sure. Hopefully <Mateo>, <Eyal> or someone with rybka takes a good look at this line and see if black can save. I mean, if L'Ami gave this variation after the game, he probably spent a little time on it and maybe white missed a win.>

<yalie: <yalie: am too lazy to go through 30 odd pages - but is it true Anand missed a possible winning advatage by playing 33.Rc2 instead of 33.Kd1? any good analysis? when i play myself, white just win everytime with Kd1 - but I may have missed a key resource somewhere.>

I am increasingly convinced Anand missed a clear win here. Kd1 wins in almost everyline I play against comp.>

"There are interesting lines after 33.Kd3 c2? 34.Rxc2 Rxc2 35.Kxc2 Black's knight has no moves. White would like to set up a zugzwang with King on d4, pawn on e4 and bishop on e3. I thought this could not be organised without allowing Nc5 but I wonder if that matters. 35...Kh6 36.Kd3 Kg6 37.Kd4 Kh6 38.Ba3 Kg6 39.e4 Kh6 40.Bc1 Threat h4 40...Kg6 41.Be3Nc5 (41...Nf8 42.Kd3 Nd7 43.Kc3 ) 42.Kc4 Nxa4 43.Kb3 wins. However Black plays 33.Kd3 Nc5+ as in the game and draws."

<you vs yourself> Well, I'm not sure... L'Ami's idea is certainly interesting; the 33...Rc4 line leads to a position where the winning plan described by Har-Zvi can be utilized. However, in the 33...Kf7 line I'm not sure what's the plan in case of 35...Kg8 (which appears in brackets with no continuation following), since then White cannot play Rf8.

You are right but Ivanchuk and Morozevich are very imaginative players and being very good tournament players makes them flexible to adjust their games strategy almost instantly.

Kramnik is only good in a match play, he can be of equal or beat anybody if they get tired of a draw and if they play as they were as Kramnik with his strong chess program studied thoroughly and prepared counter-moves. If Anand only plays almost the same openings like in Mexico, Kramnik has a better chance. His preparation with his very strong chess programs can find atleast how to counter up to draw the previous games of Anand which were all e4 openings. But unfortunately Anand plays d4 and even Kramnik is a d4 player, those opening variations he memorized for almost a year that his chess programs had found were almost all wasted in the current WCC match. It becomes like a tournament format for Kramnik that he is playing with another strong player and like in the tournament, his scores are not so impressive but always draw due to he needs almost a year for each of his opponents to study each of their styles and games. If he is playing with 10 opponents in the strong tournament, he needs 10 years to be the 1st place. So expect his worst results. But if the mathc will be stopped now and give him another year to study d4 again then he will have a little bit better chance.

He is not as genius as Ivanchuk or Morozevich who could play in different formats with different opponents. In my opinion, if Kasparov did not pick Kramnik as his personal challenger in 2000. Kramnik will never be a world champion due to he needs to topped each qualifying rounds for WCC before he can become a challenger and he would of course lose like what happened with him against Shirov in 1998 who was supposed to be the official challenger.

Kramnik's problem is that his playing style is designed not only just for matches, but <just for matches when he has the lead>.

His playing style is keep drawing, until his opponent gets tired and make some risky moves and if he spotted it, he might get a chance to win.

<INTRODUCTION>

Anand vs Kramnik, 2008 is the <seventh game of the 2008 title match>.

Quantitative mapping of this game between these players is below. Figures in brackets immediately after each move are the corrected engine evaluations generated on the forward slide that followed the initial reverse slide originating from the last move of the game. This smoothed out many, but not all fluctuation in the engine’s evaluations. The complexity of some variations, especially in the opening, was very likely too great to enable a fuller reconciliation from the initial reverse slide.

<General methods used are described in the bio at the <Bridgeburner chessforum.>

The evaluation values in the opening come at the end of initial reverse slide back to the starting position from the last move of the game. Engine preferences are included throughout the game where they differ from players preferences except in the well trodden opening and in the ending where evaluation figures became artificially inflated.

Some analysis is included to provide some idea of the reason for the engine preferences where they didn’t coincide with the moves played, except in situations where the evaluation differences was very minor or otherwise irrelevant.

<Summary>

After the subtle complexity and excitement of the sixth game, most of the excitement in this game focuses on the endgame from move 31 when many engines suddenly reported highly inflated evaluations that did not reflect the completely drawn nature of the game. For the main part, this is an uneventful game with neither player making significant errors, and ends in a quiet draw in a position which is not only impossible to win, but impossible to lose unless White chose to suicide by shifting the h-pawn at an inopportune moment.

<THE GAME>

<GAME MOVES 1 – 9>

<1. d4> (=0.17) <1…d5> (=0.17)

<2. c4> (=0.12) <2…c6> ( 0.27)

<3. Nf3> (=0.13) <3…Nf6> (=0.17)

<4. Nc3> (=0.13) <4…dxc4> (=0.20)

<5. a4> (=0.13) <5…Bf5> (=0.13)

<6. e3> (=0.13) <6…e6> (=0.13)

<7. Bxc4> (=0.13) <7…Bb4> (=0.13)

<8. 0-0> (=0.13) <8…Nbd7> (=0.25)

<9. Qe2> (=0.18) <9…Bg6> (=0.18)

GAME MOVE 10:

<10. e4> (=0.00): <10…0-0> ( 0.28)

In Vaganian vs D Marovic, 1990, Black played <10…Bh5> on this move…and was crushed. Alternatively, Black can take on the challenge and the proffered pawn (and the <engine’s preferred move> with <10…Bxc3 11. bxc3 Nxe4>, but White’s <12. Ba3> prevents Black from castling king side. This may not be an insoluble problem, eg: <12…Qc7 13. Rfc1 0-0-0 14. a5 a6> with even chances (=0.00).

It’s not surprising that Kramnik did not buy into this volatile line.

GAME MOVE 11

<11. Bd3> (=0.23)

Preferred by theory over the <engine’s preference: 11. e5> ( 0.28): <11…Nd5 12.Bg5 Qc7 13.Rfc1 N7b6 14.Nxd5 Nxd5 15.Nh4 Be7 16.Nxg6 hxg6>. Opening Explorer shows the practical breakdown between <11. Bd3> and <11. e5>.

GAME MOVE 11 to 13

<11…Bh5> (=0.23)

<12. e5> (=0.23) <12…Nd5> (=0.23)

<13. Nxd5> (=0.00)

<Engine preference> <13. Bd2> (=0.23)

<GAME MOVES 13 & 14>:

<13…cxd5> (=0.04)

<14. Qe3> (=0.00)

Still theory, with this move being invariably played, apart from one experiment with <14. h3> that went badly wrong for White (in V Gagarin vs Wojtkiewicz, 1995)

GAME MOVE 14:

<14…Re8> ( 0.44)

Intriguingly, none of the <engine’s first three preferences>, namely: <14…f6> (=0.00); or <14…f5> (=0.08) or <14…Qb6> (=0.13) appear in the database, with <14…Be7> being the most common response: Opening Explorer .

<GAME MOVE 15>

<15. Ne1> ( 0.27) Appears four times in the database.

<Engine preference> <15.Bd2> ( 0.44). This appears once in the database; it was won by White: F Berkes vs Portisch, 2003

GAME MOVE 15

<15…Bg6> ( 0.29)

The other alternative in the database is the <engine’s first preference>: <15…Rc8> ( 0.27), played twice: in Topalov vs Kramnik, 2006, and in P Karthikeyan vs V Diepeveen, 2009 played in the Dutch Open in 2009, in which White played the startling <16. Bxh7+> and won.

GAME MOVES 16-17:

<16. Bxg6> ( 0.29) <16…hxg6> ( 0.29)

<17. Nd3> (=0.25)

<Engine preference> <17.Bd2> ( 0.29): <17…Be7 18.Nd3>

GAME MOVES 17-20:

<17…Qb6> (=0.25)

<18. Nxb4> (=0.25) <18…Qxb4> (=0.25)

<19. b3> (=0.25) <19…Rac8> (=0.25)

<20. Ba3> (=0.20)

<Engine preference>: <20. Bd2> ( 0.25): <20…Qe7 21. Rfc1 Nb8 22. b4 Rxc1+ 23. Rxc1 Qd7 24. Qa3>

GAME MOVES 20-28:

<20…Qc3> (=0.20)

<21. Rac1> (=0.20) <21…Qxe3> (=0.20)

<22. fxe3> (=0.20) <22…f6> (=0.20)

<23. Bd6> (=0.20) <23…g5> (=0.20)

<24. h3> (=0.20) <24…Kf7> (=0.23)

<25. Kf2> (+0.23) <25…Kg6> (=0.23)

<26. Ke2> (=0.23) <26…fxe5> ( 0.51)

<Engine preference> <26…Rh8>

<27. dxe5> ( 0.51) <27…b6> ( 0.51)

<28. b4> ( 0.51) <28…Rc4> ( 0.79)

At this stage the game becomes extremely drawish, belying the subsequent strangely inflated evaluation figures that many engines, including the Shredder 11 used in this analysis, have produced. The evaluation figures are included out of academic interest only, and for completeness for anyone to make of what they will. Functionally, they should all be =0.00.

GAME MOVES 29-37:

<29. Rxc4> ( 0.79) <29…dxc4> ( 0.79)

<30. Rc1> ( 0.79) <30…Rc8> ( 0.79)

<31. g4> ( 0.79) <31…a5> (=1.70)

<32. b5> (=1.70) <32…c3> (=1.91)

<33. Rc2> (=0.96 and shifting) <33…Kf7> (=1.71)

<34. Kd3> (=1.71) <34…Nc5+> (=1.71)

<35. Bxc5> (=1.71) <35…Rxc5> (=1.71)

<36. Rxc3> (=1.71) <36…Rxc3+> (=1.87)

<37. Kxc3> (=1.87)

Drawn. Final position:

click for larger view

<CONCLUSION>