The Chess Mind

So said Anand in a post-match interview with The Hindu. I hope that's another lesson Kramnik (and everyone else) learned: if one won't be a good sportsman out of respect for one's opponent or a golden rule ethic, he should at least avoid taunting out of self-interest. (I can certainly confirm this from the point of view of the tauntee: few things have motivated me more strongly than a trash-talking opponent: I train harder, I'm more concentrated, etc.) Learn from Kramnik, by not doing what he did.

HT: Brian Karen

Right here, for your viewing pleasure.

It's ready to read, at the top of this post.

It was a very long press conference, and it will take me a while to type in what I can. But it happened and I'm working on it - check back to the "live update" post in a few minutes.

2:30: Post-Press Conference Update:

This was a long press conference, and a very neat one - I hope those of you who didn't get to watch it on Foidos take the time to see it later on ChessVibes or elsewhere.

Anand's opening comment was fairly short, and he rattled off the lines faster than I could write them down. The gist of what he said, though, was this: he was surprised by the Najdorf, and played 6.Bg5 which he doesn't usually do. Kramnik's 9...Qc5 may have surprised him a little; he was (more) familiar with 9...Nc6, as played in a Radjabov-Grischuk game earlier this year in Sochi. Kramnik's 12...exf5 was a critical moment, and if Black can finish his development and get his bishops firing White can be in trouble. So he liked his 13.Qe3, and later felt that 16.Qf4 was correct as well. He noted that 17...Be6 instead of 17...f5 was the only other try, but gave a line at the end of which he said he was escaping. After 20.Kb1 he knew he was out of danger, as Black could not do much with his broken kingside pawns.

Kramnik too churned out variations in his statement, but added a more general comment or two. He half-jokingly said he had a hard day yesterday, trying to find a forced win with Black against both 1.d4 and 1.e4. He was happy when he got a Najdorf, because then he could at least get a game. The only disadvantage was that he had no clue about the theory (of the particular line they played). He found 9...Qc5 over the board, and didn't know if it was a novelty. [It's not.] It looked interesting to him and keeps the game sharp. He also acknowledge that 12...exf5 was a critical move, but without it he felt he'd simply be structurally worse. So he decided to "mess things up" with that capture. 13.Qe3 was a good reaction. He felt that his 13...Bg7 was a good move and that 13...Be6 was not. (Anand had mentioned 13...Be6 as OK at first, but Kramnik rattled off some line at the end of which White is the exchange up for inadequate compensation.) Later, Kramnik thought that 15...O-O was objectively better than 15...Rg8, but that his move was the only way to fight. Anand's 16.Qf4 was a good reply. Kramnik mentioned 19.Bd3, and noted that 19...Be6 20.Nxf5 Qb4 leads to a position where "everything's hanging" and that turns out well for Black, but after Anand's 19.Nxc8 Rxc8 20.Kb1 Black has no tricks and ...Nb4 doesn't work. The final position may be holdable for Black but it's better for White.

Q & A:

Q for Anand: Relieved or happy?
Anand: Happy, but more relieved than happy.

Q for A: What now, will you continue to be active?
A: Have no thoughts of retiring, will certainly keep playing. For now, my team and I will celebrate and relax.

Q for K: Disappointed or relieved?
K: Relieved? No. You're relieved when you win, disappointed when you lose. But life goes on. I made mistakes in preparation. I am eager to improve. I will make serious changes to my preparation for tournaments and even to my play. It was a harsh lesson, but I will work. For the moment, I will relax. I hope someday to fight for the world championship again. I'm happy to have played the match and to play with Anand. It was a very interesting chess experience. It's not over [in that] I will continue to make use of the lessons here.

Q for K: How will you continue - what's your place in the next cycle?
K: You're misunderstanding. If I have another chance, I'll take it; if not, not. For now, there's less pressure, so I can rest, relax, improve.

Q from Campomanes for A: 21 years ago, you won the World Junior, and you have had a bumper crop of titles in the intervening period. What do you see for your next 21 years?
A: Hard to see ahead - I couldn't have seen this from Baguio. But I'm still excited about chess, and such a result gives you wings.

Q for K: After 1.e4 you seemed unsure for 20-25 seconds. Were you, or was it a bit of a show?
K: (Smiling/laughing.) I was trying to get a bit of concentration, since I had just gotten to the board. I expected 1.d4, but had ideas for 1.e4 as well. I wasn't unsure, just getting concentrated.

Q from GM Rogers for K: Do you feel you were always playing a moving target with Anand switching openings?
K: Not really. We were both preparing. Anand's team's prep was maybe more clever and we couldn't keep up the first half of the match. But in the second half, I got more of what I wanted. Changing openings is just modern chess, everyone knows a lot, and I switched too. Maybe repeating the same line in game 5 was a mistake but we're human, we make mistakes.

Q for A: Is it different winning a match vs. a tournament [Mexico City]?
A: Tournaments, broader; matches, deeper. Kramnik has an almost lethal match style, so I'm very happy to have gotten a lot of work done. Didn't expect to win by two points, so I'm very proud of this result.

Q for both: 12 game match OK, or 16 better?
K: Depends very much on your position in the match. (Everyone laughs.) I'd prefer a bit more, maybe 14 games. But the format of a match is great, of very high value, and draws public interest. I hope it continues even if I'm not in it. 14 games is just right.
A: I've played a 20 game with Kasparov, a 12 game match here, and 6 game matches in the k.o.'s which is like a Candidates match. [Ironically, some of Anand's Candidates matches were longer than 6 games.] 12 games is fine; the difference between 12 and 14 is not enormous.

Q for K: How did you cope with your difficulties early in the match and come back to fight? And how did you come to the press conferences in a good mood even when you lost [Kramnik said "looked like a good mood"]?
K: I was in a fighting mood from the beginning, but everything went wrong the first 6 games. After that, I could fight. As for press conferences, we're professionals. We don't feel good, but it's not journalists' fault. And it's only a game - yes, an important game for the world title, money - but only a game. It's not a reason to be depressed. And people lose games, especially against such an opponent as Vishy. I have a motto: you're responsible for the quality of your work, not the results. Vishy is a great player and was better in this match. It's no reason to shout at journalists.

[And with that, the press conference came to an end, and so now does this post. Stay tuned later tonight for the game on a replayable board with fuller annotations, and for my ChessBase show at 9 p.m. ET in which I'll recap the match from game 6.]


1:20

I assume there's going to be a press conference, but they sure are taking their sweet time about it. Maybe they're engaging in another round of drug testing, brilliantly ensuring that the players aren't who they are today on account of anabolic steroids. (Maybe the geniuses who thought this up are the ones who should be tested, but it might be too late to rescue them.) Meanwhile, for posterity's sake, here's the final position of the match:



1:00

MATCH OVER!

Yup, that's it! After 21.Nc1, the game concluded 21...Ne7 22.Qd2 Qxd2 23.Rxd2 Bh6 24.Rf2 and now I'm not sure what happened. I was away from the computer and 24...Bf4 was allegedly played, but as that simply throws away a bishop for nothing it's rather unlikely. (I believe the move was 24...Be3.) In any case, whatever happened on Black's 24th move, they agreed to a draw. It's over, and Anand has won the match by a 6.5-4.5 score.

Back after the press conference.

12:40

Latest moves: 19.Nxc8 Rxc8 20.Kb1.

I think I know why Kramnik refrained from ...Be6 on moves 16 and 17 now, but it looks like it was just a cheapo. Had Anand played the natural move 20.Bd3, then 20...Nb4 trapped the rook, though the position would remain unclear afterwards. After Anand's strong 20.Kb1, taking care of the back rank by allowing Nc1, Black doesn't seem to have very much. For example: 20...Qe1+ 21.Nc1 Qb4 22.Nd3 Qxf4 23.Nxf4 Nb4 24.Rxf5 Rxc2 25.Ne6+ Ke7 26.Nxg7 Rd2 27.Bxa6 bxa6 28.Re1+ White is slightly better and enjoys a position that's practically impossible to lose.

Kramnik has just played 20...Qe1+ and Anand replied 21.Nc1, and if the game continues down that line I think the game will end in a draw - unless Anand feels like playing for a win. I think the match is as good as over.



Times: Anand 40 minutes, Kramnik 41 minutes and counting.

12:30

Anand chose 16.Qf4, sidestepping the discovered check possibility, eyeing the d6, e4 and f6 pawns and keeping the Bg7 plugged up. Now exf5 is a serious positional threat, so Kramnik's choices were 16...Be6 followed by 17...fxe4 or else the immediate 16...fxe4, and he chose the latter. 16...Be6 seemed to gain a useful tempo for development, but for whatever reason he declined to use it. Anand played 17.Nxe4, and Kramnik again forsook 17...Be6, preferring 17...f5, unleashing the dark squared bishop's diagonal. The obvious 18.Nxd6+ Kf8 ensued, and now Anand has some choices and 19.Nxc8 and 19.Bd3 come to mind.

Times: Anand 46 minutes and counting, Kramnik 48 minutes.

12:05

After a long think, Kramnik chose the ambitious 13...Bg7. The point is that someday, after ...fxe4 and ...f5, that bishop can become a monster on the long diagonal, especially in conjunction with the queen on e5. Achieving that isn't going to be so easy, however, especially after White's reply, 14.Rd5. Kramnik could have prevented this with 13...Be6, when 14.Qb6 Bh6+ 15.Kb1 O-O 16.exf5 Bxf6 17.Bd3 gives White nice compensation for the pawn.

After 14.Rd5, Kramnik played the normal 14...Qe7, and Anand has replied with 15.Qg3, forking the units on g7 and d6. Black has a choice between three moves here, 15...Bh6+, 15...O-O and 15...Rg8, and there's something to be said for each of them. 15...Bh6+ is a materialistic approach, allowing Black to save his extra pawn and maybe grab another one on e4. 15...O-O takes care of king safety issues (relatively speaking), and 15...Rg8 tries to activate another piece, with tempo. White can meet that move with 15.Kb1 or play something like 15.Qh4 or 15.Qf4, trying to restrict Black's bishop; after each reply (to 15...Rg8) the position is in some sort of balance. And Kramnik has just played it: 15...Rg8.



It's a very complicated position, strategically and tactically, and it's not surprising that the players are burning serious time on the clock.

Times: Anand: 58 minutes and counting, Kramnik: 59 minutes.

11:30

Only one half move has been played, but it's an interesting one: 13.Qe3. This clears the d-file and the f1-a6 diagonal while giving the queen two beautiful diagonals (b6 is an especially wonderful square for the queen). Kramnik has been deep in thought since this was played, having used 24 minutes on it thus far. It's not over, but Anand fans should be pretty happy here.

Times: Anand 1:17, Kramnik 1:12 and counting.

11:00

Oh boy...

The next few moves were "normal". After Kramnik's 9...Qc5, Anand played 10.Qd3, preparing to castle long. Kramnik continued developing with 10...Nc6, Anand avoided the possible swap of knights and queens by playing 11.Nb3, Black played the thematic centralizing move 11...Qe5 and then White continued normally with 12.O-O-O. In the few previous games Black had played 12...Bd7, and there may be other normal moves as well (e.g. 12...b5).

But here Kramnik played 12...exf5. Sure, it wins a pawn, but it's strategically disgusting. His d- and f-pawns are isolated, the e-file is open and d5 beckons a white knight. Maybe this is an "I must play to win" kind of risk, but it looks pretty gruesome to me. Kramnik does have extensive experience with the Classical Sicilian, meeting the Rauzer (which is what this really is, strategically - ignore the Najdorf move order), so it's not as if he doesn't know any of this, and he must feel that this is an acceptable risk.

Times: Anand 1:24 and counting, Kramnik 1:36.

10:25

Yes, 1.e4 has been played! Kramnik chose 1...c5, and now after 2.Nf3 d6 Anand avoided the comparatively drawish 3.Bb5+ and went for an Open Sicilian with 3.d4. 3...cxd4 4.Nxd4 Nf6 5.Nc3 ensued, and now Kramnik played an unusual move for him: 5...a6, the Najdorf Variation. Kramnik used to play the Sveshnikov regularly, and in the mid to late 1990s he was a consistent Classical player (5...Nc6).

With the Najdorf, I can't help but think of the end of the 1969 World Championship match between Boris Spassky and Tigran Petrosian. Through game 16 (of 24) the match had been all square, 8-8, and Petrosian had drawn one game after another with the Petroff, and with ease. For no obvious reason (a death wish of sorts?) he switched to open Sicilians in games 17 and 19, got butchered and lost the match. It didn't play to his strong points and played into Spassky's, and I think the same is true for Kramnik and Anand. But if you have to play for a win, then what can you do?

Returning to the game, Anand played 6.Bg5, formerly the main line and a move that leads to a huge number of forced draws, but after 6...e6 7.f4 Qc7 Kramnik had found his way to a line without many or even any forced draw variations. Here Anand avoided typical Najdorf-like approaches and chose 8.Bxf6 gxf6 9.f5, giving the game the character of a Classical Rauzer. Kramnik played 9...Qc5, and that's where we are at the moment.

Kramnik has achieved a playable position where no forced draws are lurking, where he has trumps that could lead to a win someday, where Anand doesn't have some mega-prep lurking and in the sort of position he had some experience with. It's miles away from a win, but he has achieved everything he could reasonably hope for from a black opening. Not a bad start!



Times: Anand 1:45 and counting, Kramnik 1:51.

9:55:

Once again, this could be the last game of the match, and since Anand has the white pieces, it's extremely likely if not quite a mortal lock. A few commentators have noted that Kramnik has never beaten Anand with Black, and that's not an encouraging factor for the challenger. However, while I agree that Kramnik's chances remain pretty low, I think that particular argument is overrated for several reasons.

First, Kramnik has defeated Anand with Black, just not in a classical game. He has done so in rapid games. Second, he generally hasn't tried to win with Black in slower games. That doesn't mean he would have had he wanted to, but that fewer attempts will naturally result in fewer successes. Third, Anand had only defeated Kramnik once with Black in classical chess before the match (and that from a dodgy position), but that didn't stop him from winning twice here with the black pieces. And finally, Anand has more pressure on him here than in any of the previous games, and that can have an effect as well.

Nevertheless, I think Anand will keep control and look for ways to dry up the game as soon as possible. Maybe it's a cynical approach, but it's better to do that and be the champ! To that end, I think 1.e4 has a good chance of appearing, as there are more chances to control the opening and steer it in a drawish direction than with 1.d4. The game starts in a couple of minutes, so buckle up: it's time for a ride.

Here it is.

1:30 Post-Press Conference Update:

Kramnik: We played a Nimzo with 4.Nf3, a fashionable line. 18.Re1 was a novelty; nor crushing, but it offers Black a certain choice and it's not so easy. It looks like I [Kramnik] am losing tempi with all the bishop moves, but I managed to get Black's pawn on e5 and his pieces lose coordination. 22...Bg4 was not an accurate move. Maybe 22...f6 was better, and then he'd play 23.h3. After 22...Bg4 23.Qa6 was best; taking on c5 [23.Bxc5] was not clear. Anand's 23...f6 was virtually the decisive mistake; 23...Be6 24.Bf1 Qf3 gives White a nice edge but it's far from over. But in the game after 26.Rab1 it's just over, because Black can't blockade c4. One nice line is 26...Kh8 [instead of 26...c4] 27.a5 Nc4 28.Rb7 Qg8 29.Bh6! gxh6 30.Bxc4, winning [30...Bxc4 31.Qxf6+ and mate next move]. It was a pleasant surprise; I didn't do anything special and then the position was basically winning.

Anand: 18.Re1 is a tricky move, because it's unclear where Black should place his pieces. He mentioned 22...Nc4 as a possible alternative to his 22...Bg4, but suggested that White still had an edge after 23.Qa6 Nxe3 24.Rxe2. As for the game, he said he "just missed something" after 23...f6, and didn't see what to do after 24.a4.

Q & A Highlights:

Not much of interest here. Anand was asked what exactly it was that he missed after 23...f6, and for whatever reason avoided answering it.

Kramnik was asked by GM Ian Rogers how he assessed his chances in the match (and as a joke, if FIDE will rush in to stop the match); he answered that his chances improved. His approximate response: "I just have to play. I'm happy to play normal chess, not under pressure in the opening. I will try to play well, use whatever chances I have. It's better for me not to think what are my chances. They're still less than 50%."

Someone else asked Kramnik if he had noticed the similarities in the opening with the game Grischuk-Gelfand from Mexico City in 2007. He answered that there are lots of similar games [as noted above, it's a fashionable variation], but that there are a lot of nuances; it's a very delicate position, difficult to understand even for top chess players.

The game with annotations will be posted later today. Remember that tomorrow is a rest day for the match; game 11 is on Wednesday.

12:56

It's over! Not the match - yet - but the game, and Kramnik has won it. The remaining moves were 27...Na4 28.Rb7 Qe8 29.Qd6 and Black resigned. White's control of the board is overwhelming and Black is probably losing a piece - Re7 is threatened, but there's also Qb4 followed by Ra1.

A decent game by Kramnik, but Anand helped with the serious inaccuracy 23...f6 and the outright mistake, maybe even blunder, 26...c4.

The match score is now 6-4, Anand, and I expect that in game 11 after the rest day he'll play to dry up the game and finish the match.

One more update later, after the press conference, and later today I'll post my analysis of the game. Stay tuned.

12:50

And again, as in the previous game, Kramnik slipped a little, but Anand returned the favor with interest. After Anand played 25...Be6, Kramnik replied 26.Rab1 almost immediately. The problem with using the a-rook instead of the e-rook is that after 26...Rcb8 27.a5 Black has the jump 27...Na4.

White would still maintain a serious edge, but it would have been reduced a bit from where it was. Unfortunately for Anand, he chose to move a completely different unit to c4 - 26...c4, and now 27.a5 gives White a winning position.

Times: Kramnik 56 minutes, Anand 19 minutes and counting.

12:30

Kramnik's advantage is growing. This is reminiscent of Kramnik's own match victory over Kasparov, where almost as soon as the match seemed over, once Kramnik went up 2, Kasparov got pressure in game after game (though without ever managing to win a game).

After 22...Bg4, Kramnik played 23.Qa6, as expected, and now Anand might have come up with a stinker: 23...f6. The move has a certain logic to it. Black wants to put the knight on c4, but 23...Be6 24.Bf1 leaves him unable to do so. By playing 23...f6, Black wants to play ...Qf7 there, to win the c4 square, but there's a problem. After 24.a4! Qf7 25.Bf1 Black can't overwhelm White with 25...Be6 because after 26.Reb1 Nc4?? fails to 27.Rb7 and 26...Bc4?? to 27.Bxc4 Nxc4 28.Rb7 or 27...Qxc4 28.Rxb6, winning a piece in every case. Black can try to prepare this with 26...Rcb8, but 27.a5 Bc4 28.Bxc4 Nxc4 (again, not 28...Qxc4?? because 29.axb6 wins a piece) 29.Rxb8+ Rxb8 30.Bxc5 with a clear extra pawn.

At the end of the day, then, White has won the battle for c4, and Black's pieces will be in disarray for a while. White is clearly better, and the situation on the clock favors Kramnik as well: 1:01 for Kramnik, :32 and counting for Anand.

12:00

Sorry, but only one more half-move to report, and it's 22...Bg4. This does inaugurate ...Nc4 as a sort of threat in such a way that he won't lose his light-squared bishop in return (e.g. ...Nc4 Qa6 Nxe3 Qxe2), but now White can stop it with 23.Qa6 or 23.Qb5.

You might wonder about the immediate 23.Bxc5, but then Black has 23...Nc4 followed by ...Nd2 and ...Nf3, when White will have some kingside worries on the light squares. So I'd expect 23.Qa6 or 23.Qb5.

Times: Kramnik 1:12 and counting, Anand :50.

11:32:

"Long line, wrong line" indeed: I guessed wrong from the very first move! After 11 minutes, Kramnik chose 22.Be3. Times: Kramnik 1:28, Anand 1:03.

11:30

"Black shouldn't play 21...e5 now". Needless to say, he did, apparently believing that the strength of his c-pawn suffices against White's assets in the position after 22.Bxe5 Nc4 23.Qa6 Qxe5 24.Rxe2 Qxc3 25.Ree1 Ne5 26.Rec1 Qb2 27.Rab1 Qd2 28.Rd1 Qc3, but it looks like White has a serious edge after 29.Qd6 (29.Qe2 is only equal: c4 30.f4 Nd3 31.Qe3 Qc2 32.Qd2 Qa4 33.e5 Rab8=) 29...h6 30.Rbc1 Nf3+ 31.Kh1 Qa3 32.Qd3 Qxd3 33.Rxd3.

"Long line, wrong line" holds true for the above, I'm sure, but it seems representative. It looks like Kramnik is going to have his chances in this game, too.

11:15

18.Re1 got Anand thinking, but he seems to have done just fine against it. One idea of Kramnik's novelty was to impede typical maneuver 18...Be2, as now 19.h3, threatening g4, will make Black's bishop beat a hasty retreat. So after 15 minutes, Anand chose 18...c5, and after 19.Qa5 Rfc8 20.Be3 Be2! appears at least close to equality, if not already there. He has achieved the thematic move and threatens ...Nc4, and it's not clear what White really has.

21.h3 is useless here: 21...Nc4 22.Qa6 Nxe3 23.Rxe2 Nxg2 24.Kxg2 Rd8 with equality. Kramnik played 21.Bf4 after his first think of the game (12 minutes), and it looks like the best move in the position. The bishop avoids exchange via ...Nc4, keeps rooks off of b8, and aims to embed itself on d6, supported by e5 (though only when ...Nc4 isn't in the offing).

Black shouldn't play 21...e5 now. It doesn't lose a pawn, but White has a nice edge after 22.Bxe5 Nc4 23.Qa6 Qxe5 24.Rxe2 Qxc3 25.Ree1. A more likely choice is 21...Bd3, which echoes 21.Bf4 in keeping the opponent's rooks off of the b-file.

Current times: Kramnik 1:39, Anand 1:10 and counting.

10:30

Here we go with what could well be the last game of this world championship match, as any result but a Kramnik win gives Anand victory in the match. This doesn't mean that Kramnik needs to play crazy chess, and there's no real likelihood that he will in any case, but he does need to keep the tension in the game until he achieves something concrete.

The game has begun with a little surprise: 1.d4 Nf6 2.c4 e6 3.Nc3 and now Anand played his first non-Queen's Gambit family opening in the match with 3...Bb4, a Nimzo-Indian. It's not a big surprise, though, as Anand has played the Nimzo pretty regularly for a long time.

Now Kramnik offered a small surprise in turn with 4.Nf3, the "Flexible Variation", so called because depending on Black's replies are met by significantly different rejoinders. The pictures of the players we get to see on Foidos are a little small, but it seemed to me that Anand had a little smile on his face after seeing this move. I interpreted it as an acknowledgement that Kramnik had finally taken the upper hand in the pre-emptive preparation battle, getting in his surprise first, but that's just a guess. (Maybe it will get confirmation in the press conference.)

After a moment, the game continued apace: 4...c5 5.g3 cxd4 6.Nxd4 O-O 7.Bg2 d5 8.cxd5 Nxd5 9.Qb3 Qa5 10.Bd2 Nc6 11.Nxc6 bxc6 12.O-O Bxc3 13.bxc3 Ba6 14.Rfd1 Qc5 15.e4 Bc4 16.Qa4 Nb6 17.Qb4 Qh5, all of which has happened on many occasions and on the highest levels.



The two known moves here are 18.Bf4 and 18.Be3, and just to pick a tiny bit from the cream of the crop, there are the games Grischuk-Gelfand, Mexico City 2007; Bacrot-Carlsen, Biel 2008; and even Kasparov-Anand, Wijk aan Zee 2000.

But that's history: Kramnik uncorked a novelty, 18.Re1, and now Anand has started to think (13 minutes on the move and counting). We'll discuss the move later; for now, let's get this posted.

It was a close one, but Anand eked out a draw. With his 6-3 lead he is one draw (or win) away from retaining the world championship title. Today's game was a big hurdle, and you can see just how close it was by checking out the game, with my notes, here.

The note: they're starting an hour later than usual today.

The ad: I'd like to make an e-book of the match for sale through here. I'm thinking of the following price structure: $10 for the games alone in .pgn and .cbv format (with expanded notes from what I've already shown in the replayable boards, with separate text files introducing the games), $15 for a series of comparatively short ChessVideos.tv-style videos, and $20 for a combined package. If you're genuinely interested, not only in the idea in the abstract but willing to buy at those approximate prices, drop me a note here. If there's enough real interest, it'll happen; if not, not.

Going into this game, Kramnik's chances to save the match were already very slim, and with the draw today slim has left the building. Anand leads 5.5-2.5, and needs just one point in his next four games to keep the title. He failed to cash in, but Kramnik did have some chances today, as you can see here.

1:30 Post-Press Conference:

Kramnik: This was the first time I had a slight advantage, so it was quite pleasant. Anand's 10...Bxb5 was an interesting novelty (or at least a novelty to him). 11.Nb3 was the principled move but he didn't like it. He got nothing special in the game, and it was equal after 12...Nc6. 14...Rd8 would have been equal, but after 14...Rg8 White was better, and after 19.Re1 Black's position was dangerous. White has at his disposal the e5 and f5 breaks, for example, but Anand found a few accurate defensive moves - 19...a6, 20...Kf8 and 21...Rg6. His 22.g3 was slow and not in the spirit of the position, but he couldn't find a way to attack. Black's Qd4 + Rd3 construction was shaky, but it also interfered with White's coordination, and then all Black needed was to get the other rook into play and then he's OK. That said, he thought that 26...Qc5 was better than 26...Rd8, as the former would not allow Qh5. In the final sequence, he wanted to bring his pawn or king to h6, but never found a way to make it work - he's never in time - so he took the perpetual.

Anand: After 14...Rg8 15.f4 he knew he had done something wrong, and then had to choose between various setups where he's worse. He thought bringing the king to the queenside would be too slow, and felt that 19...a6 was forced. 20.Nd5 was very interesting and dangerous, but because only major pieces would be left maybe he can survive. White was better the whole game. Anand was happy when he found 21...Rg6, noting that if 22.f5 exf5 23.exf5 Rg4 Black would be in fine shape. He discussed various attempts to exploit the Qd4/Rd3 construction, including in positions where White's rook is on e1. So (my particular move reference, not his) 21.Na2 Qd8 22.Nc1 Rd7 and White hasn't made progress, because to achieve anything positive he'll need the knight back on c3. So Black's position was worse but defensible, and the last accurate move was 32...f5, after which it's a draw.

Q & A:

A couple of people asked Kramnik about playing on at the end with Kh3 with the idea of Rg2. Kramnik replied that it was an option but that he'd probably lose. (The second time, understandably growing tired of suffering fools, said approximately "my situation in the match is not so good but not to lose like an idiot". Having vented, he let the questioner save face, adding that he understood the question and had of course considered the ideas.) Specifically, he tossed off the following lines: 39.Kh3 Rd3 and tehn Black's king runs out, shedding the h-pawn but gaining big counterplay. Or Black can defend with ...Qc5-e7. Also, if instead 39.h4 Rd3 40.Rf2 Qc5 and ...Qe7, or 40...Kf8-e7 with counterplay for the pawn. Or at the end, in response to the suggested 40.Kh3 Rd3 41.Qg5+ Kg8 42.Rg2 Qc5 43.Kh4 Qe7 44.Kg5, Kramnik noted 43...Ne7 instead "and maybe White will get mated". (Around here came the "lose like an idiot" remark.)

A substantive question for Anand was to explain why he chose the Rd3 + Qd4 plan. His reply was illuminating: I felt the position was worse but manageable. I need activity because the structural threat of f5 is so strong, and at some point you have to pick some plan and stick with it.

And so he did, and it worked out: Anand leads 5.5-2.5. Tomorrow is a rest day, and then he'll have White on Saturday in what could be the final game of this match.

More analysis later in a separate post.

12:40

Kramnik is trying, but it looks like there's not going to be anything more than a perpetual check here. Yes, there it is: draw agreed. The final moves were 32...f5 33.Qf6+ Kg8 34.Qg5+ Kh8 35.Qf6+ Kg8 36.Re2 Qc4 37.Qg5+ Kh8 38.Qf6+ Kg8 39.Qg5+ Kh8 Draw.

The problem for White is that tries like running the h-pawn up the board, moving the king up to h6 or trying to utilize the g-file don't quite work, generally because of Black's counterplay along the d-file. More on this later, but here's one quick variation: 36.h4 Qe3 37.h5 Rd2 38.Rxd2 Qxd2+ 39.Kh3? Qc1 40.Kh4 h6 41.Qxh6 and White would be very happy now, were it not for 41...Nxe5! and Black wins.

Back after the press conference.

12:15

Kramnik finally got a little chance, and immediately let it slip. After 22.g3, Anand played 22...Kg7 (the ugly 22...f5 was possible, but it's not a move humans want to make, especially without a really good reason), and after 23.Rd1 Rxd1 24.Nxd1 he chose the very slow 24...Kh8. The motivation is a good one, to re-centralize the rook with ...Rg8-d8, but now White could have achieved a genuine advantage with either 25.Qc2 Rg8 26.Rd2 Qb6 27.Qc3 Kg7 28.Ne3 or 25.Rf3 Rg8 26.Rd3 Qc4 27.b3 Qc5 28.Qb2.

Unfortunately for Kramnik fans, he chose a couple of routine moves, 25.Nc3 Rg8 26.Kg2, and the position was equal. Anand continued his plan with 26...Rd8 (26...Na5!?), and now Kramnik finally made an active move with 27.Qh5 (27.f5 might have given an edge). Now we've seen some repetitions, but fear not: this is almost surely to gain time on the clock:

27...Kg7 28.Qg4+ Kh8 29.Qh5 Kg7 30.Qg4+ Kh8 31.Qh4 Kg7 and now 32.e5 has been played.



At a glance, this doesn't seem to give White more than a draw, but we'll see. More later. Times: Kramnik 19 minutes, Anand 20 and counting.

11:45

Kramnik played 21.Ref1, as suggested below, and yet after Anand's fine reply 21...Rg6 Kramnik still can't successfully blow open the f-file. Kramnik can play 22.f5, but after 22...exf5 23.Rxf5 is verboten, as after 23...Rd2 24.Qf3 Ne5 25.Qh3 Ng4 or 24.Qh5 Rdxg2 White's position collapses. White can play 23.exf5, but it promises nothing special after 23...Rg4. Besides, what are White's rooks doing on the f-file in that case?

Kramnik has therefore chosen another waiting, building move: 22.g3. It's committal in its own way, though, opening the a8-h1 diagonal and further exposing the second rank, so Kramnik definitely isn't getting something for nothing here.

Times: Kramnik 31 minutes, Anand 39 minutes and counting. Stay tuned!

11:30

Anand replied with 20...Kf8, and now it's important to note that the immediate 21.f5 isn't so good. Or rather, after 21.f5 Ne5, 22.fxe6 isn't good: 22...Ng4 followed by 23...Rd2 forces White to give up the queen for insufficient compensation, e.g. 23.Rf4 Rd2 24.Rxg4 Rxe2 25.Rxg8+ Kxg8 26.exf7+ Kxf7 27.Rxe2 b5 with a clear advantage for Black. White could play 22.h3 instead, but after 22...exf5 23.Rxf5 (perhaps not best) Rd2 24.Rd1 (forced) 24...Rxe2 25.Rxd4 Rxb2 26.Rd8+ Kg7 27.Rxg8+ Kxg8 Black is slightly better. So Kramnik should wait before making the break - 21.Ref1 looks like a smart move here.

Current times: Kramnik 40 minutes and counting, Anand 49.

11:20

Very interesting: after 17 minutes, Kramnik has uncorked 20.Kh1, keeping his powder dry. This breaks the pin on the Rf2 and prepares Ref1 followed by f5 - the original strategic plan behind f4. Happily, this keeps the tension in the position, which Kramnik must do if he wants to have serious winning chances in this game.

11:15

Kramnik chose 19.Re1 instead of 19.Nb5 and the endgame mooted in the last post, and now the position is much sharper. Anand replied 19...a6, which keeps White's knight off of b5 and, believe it or not, gives his king a flight square.



Kramnik is deep in thought now, and at least part of what he's considering is the leap 20.Nd5. Black grabs the knight, 20...exd5, and after 21.exd5+ Kd7 22.dxc6+ Kxc6 White's best might be 23.g3, and now Black can play 23...Kb6 with the idea of ...Ka7 if necessary. White may be a little better here, but Black's well-centralized forces offer good compensation for White's superior structure.

Times: Kramnik 52 minutes and counting, Anand 54 minutes.

10:45

A very slight surprise has just happened. After 15.f4, mentioned last time, the subsequent series of moves were all predicted long ago: 15...Rd8 16.Qe1 Qb6+ 17.Rf2 and now instead of 17...Kf8, squirreling the king away to h8 as soon as possible, Anand played 17...Rd3. The problem with this aggressive-looking move is that White can play 18.Qe2 (just played), and what now? It's hard to see what the point to ...Rd3 was if Black doesn't play 18...Qd4 now, so let's consider that.

Here's a long, forcing (but not forced variation) to contemplate as we wait to see what happens: 19.Nb5 Qe3 20.Qxe3 Rxe3 21.Nd6+ Ke7 22.Nxb7 Rxe4 23.Rc1 Rb8 24.Nc5 Rd4 25.Nxe6 fxe6 26.Rxc6 Rd1+ 27.Rf1 Rxf1+ 28.Kxf1 Rxb2 29.Rc7+ Kd6 30.Rxa7 with an extra pawn at least partially compensated by Black's active king.

Times: Kramnik 1:09, Anand 1:04(!)

10:00

The game has proceeded along quiet lines: 13.a3 Bxc3 14.Nxc3 Rg8 15.f4. White's aim is positional: he'd like to play f4-f5, leaving Black with some questions about how he'd like to keep his kingside together. Black's king is stuck in the center too. That probably won't be a king safety issue, but it does interrupt the coordination of his other pieces, especially the rooks.

Does this mean that White is better here? Perhaps surprisingly, the answer seems to be no, he's not. Black's slight lead in development seems sufficient to keep the balance (but not more) after something like ...Rd8 and ...Qb6+, possibly followed by ...Kd7-c8, taking care of his only remaining weakness. It does seem that White's position is easier to play, so practically speaking Kramnik has reached one of his best positions in this match. (Admittedly, this isn't saying much.)

The times are getting closer: Kramnik 1:14; Anand 1:32 and counting.

9:30

Anand has offered some new opening surprises. After 1.d4 Nf6 2.c4 e6 3.Nf3 d5 4.Nc3 Anand eschewed his usual Semi-Slav with 4...e6, and chose instead the Vienna Variation with 4...dxc4. Kramnik continued to move at a normal rate: 5.e4 Bb4 6.Bg5 c5 7.Bxc4 cxd4 8.Nxd4, and now a second surprise: 8...Qa5 rather than the usual 8...Bxc3+ 9.bxc3 Qa5. Now White has the option of 9.Bd2, keeping the queenside structure intact, but tried instead to return the game to normal 8...Bxc3+ channels with 9.Bb5+ Bd7 10.Bxf6.

Now 10...Bxc3+ 11.bxc3 gxf6 (not 11...Qxc3+ 12.Kf1 gxf6 because of 13.Rc1, as Rc8+ is threatened when the queen moves away and 13...Qxd4 14.Qxd4 Bxb5+ 15.Kg1 Nd7 16.h4 favors White) 12.Qb3 returns to a main line Vienna, and 10...gxf6 11.O-O Bxc3 12.Bxd7+ Nxd7 13.bxc3 is a less common but still normal Vienna position. But here Anand produced a third surprise with 10...Bxb5, a move which, as far as I can tell, was played only once before in a non-elite correspondence game. That game, Magallanes (2209) - Hitzegrad (2303), ICCF email 2006, continued 11.Nb3 Qb6 12.Bxg7 Rg8 13.Bd4 Qc6 14.Qf3 Nd7 15.Rc1 Qd6 16.a3 Bxc3+ 17.Bxc3 Ne5 18.Bxe5 Qxe5 19.Rc3 Rd8 20.g3 b6 21.Qe3 Ke7 22.Nd2 Rd7 1/2-1/2.

Perhaps calculating some of this and seeing that White doesn't achieve much, Kramnik opted for 11.Ndxb5, and after 11...gxf6 12.O-O Nc6 we've reached the current position. (Given below.) White has 1:37 and counting; Anand 1:56.

Here it is.

12:30, Post-Press Conference Update:

I missed maybe a sentence or two at the beginning, but got just about everything else.

Anand opined that the position after 23.Bd6 was slightly better for him for the obvious reason that his bishop was better than Black's knight. He agreed with Kramnik (from their mini-post mortem at the board) that Black should have played ...fxe5 more quickly than he did, but noted that 31...a5 came just at the right time. Whenever White plays Kd3, Black has ...Nc5+, so to try to win White must play e4, Ke3-d4, but then Black has ...Ke8 and he is again in time to hold.

Kramnik's summary added a few interesting points. He recalled that he played the same line against Topalov in a rapid game in Elista, but played differently at some point. He felt that after Anand's 18.Nxb4 Qxb4 19.b3 Rac8 that 20.Bd2 would have been better, or at least a better winning try, keeping some tension and aiming for play on the kingside. After 20.Ba3 he felt the position was completely fine and he got too relaxed. He should have played ...fxe5 earlier (I'm assuming he meant on move 24), and after (25.)dxe5 b6 when he's up two tempi compared to the game. He thought during the game that all was well, that he would just play ...g5, put the king on g6, and then see what was happening, but he then started to realize that it wasn't so easy. Maybe he could draw with passive play, but that would be dangerous; it would be better to get active. He felt that 30.Kd2 (rather than 30.Rc1) might have been better. It would still probably be a draw, but it would be more difficult. (He gave a line that started 30...a6 31.b5 axb5 32.axb5 Rc8, but unfortunately I didn't catch the rest of it.) After Anand's 30.Rc1 he believed it was a forced draw, even if it required some "only moves" from Black.

Of the questions, a couple shed further light on the game. One asked about the draw offer on move 21, why Kramnik offered and why Anand refused. Taking the answers out of order, Kramnik offered because he thought that the position was objectively drawn and that if Anand had wanted to play for a win, he'd have chosen 20.Bd2. Anand reported that he was a little surprised by the offer. He didn't think he had much, but that Black would still have to make some accurate moves.

GM Illescas asked the players' opinions about the possible knight ending that could result if White had managed to win the pawn when it was on c4 with a trade of rooks. Both shot off some long variations, and I was only able to write down some of them. Here then is the gist: Anand believed Black would be lost, and gave a line like 1.Ba3 Kg6 2.Bc1 Nc5 3.Be3 and noted that White would be able to achieve a winning zugzwang. He added that this approach wouldn't work if Black's king were on e8. On the other hand, in a position where Black plays 1...c3 2.Kd3 a6 3.Rxc3 Rxc3+ 4.Kxc3 b5 he wasn't sure how to win that one. At some point he'll have to play Bc5 and Kc3, but then ...Nb8-c6 should hold. And he wasn't even sure he could get that. Kramnik hastened to add that there was no need for this, but summarized nicely by saying that if the pawn gets to ...b5, it's a draw; if the pawns are on b6 and a5, it's lost. He added some illustrative lines/ideas, but I wasn't able to record them.

That concludes this post; full annotations later.

12:00:

Draw. The game continued 33.Rc2 Kf7 34.Kd3 Nc5+ 35.Bxc5 Rxc5 36.Rxc3 Rxc3+ 1/2-1/2. (Those unfamiliar with DGT boards might think White played the preposterous 37.Ke4??????????? here, but that's just the usual incompetence by the referee in using the idiotically designed DGT system. (To indicate a draw, White's king is put on e4 and Black's on e5, but if it isn't done just right Ke4 gets interpreted as a move. This has happened literally hundreds if not thousands of times, and despite this arbiters still haven't learned how to do it right and DGT still can't be bothered to create a different system. Something really complicated, for example, like a switch on the side of the board.) What actually happened was that after Kramnik took the rook, Anand looked up as he finished writing down the move, gave the "Draw?" look, and that was that.)

Back to chess. White might have been able to keep a little life in the position with 34.e4, with the idea of Ke3-d4, but that might lead to a probably drawn rook ending instead after 34...Rc4 35.Kd3 Rxa4 36.Rxc3 Nc5+ 37.Bxc5 bxc5 38.Rxc5 Rb4. Still, it's a try.

Anand leads the match 5-2. Game 8 is tomorrow, my fuller annotations are tonight, and the press conference is coming up now. Unfortunately, Foidos is having some problems, so I might be out of luck. (In that case, we'll all have to watch it later on, on Chess Vibes.) Back with the final update soon.

11:45:

Alright, forget what I just wrote. After Kramnik's excellent 31...a5 and Anand's 32.b5 (maybe he didn't have to play that?), the position immediately became a draw. Kramnik played 32...c3, and only now it's dawning on Anand that Black has a perfect fortress: 33.Kd3 Nc5+ 34.Bxc5 Rxc5 35.Rxc3 Rxc3+ 36.Kxc3 and that's it, thanks for playing, time to shake hands. White can bring his king to the kingside and play h4, but Black ignores that as well, meeting h5 with more napping and hxg5 with g6. Kramnik lives to see the weekend.

11:30

Yep, things are getting grim here for Kramnik. To catch you up to speed, what we've seen since 28.b4 is 28...Rc4 29.Rxc4 dxc4 30.Rc1 Rc8 31.g4. Black's c-pawn is more a weakness than a strength and three of his pieces are passive. One possible continuation is 31...a6 32.b5 axb5 33.axb5 c3 and now 34.Rc2!, preparing Kd3, endangers the life of the c-pawn. (Note: not 34.Kd3 right away on account of 34...Nc5+ 35.Kxc3?? Nb3+.) So it's not clear what Kramnik has in mind, but it's looking increasingly like 5.5-1.5. Wow.

11:00

Kramnik has started burning up time on the clock, and that's usually a dangerous sign for him. At this point he can afford it, but it might be a slight worry. After 18 minutes, Kramnik played 25...Kg6, which both activates the king and gets it off the f-file. Anand played 26.Ke2 almost immediately, and after another 9 minutes Kramnik played the very committal 26...fxe5. Opening a second file makes the position more lively than it would otherwise be, and after 27.dxe5 he made another somewhat surprising move, 27...b6. Normally Black wants to keep his pawns on light squares, where they're safe from the bishop's prying eyes - imagine all the rooks are off and White plays b5, for instance. Then Black will always have serious worries about his a-pawn.

The move does have a point, of course: Black would like to play ...Nc5, so 28.b4 (as yet unplayed) is fairly obvious. One question then is whether Black can get away with 28...Rc4, and that's highly unclear: White trades and plays Kd2-c3, when the c-pawn is weak. Trying to defend it with ...a6 and then ...b5 is easily met by b5, so we'll have to see what Kramnik has in mind. To paraphrase an old SNL bit, he's Kramnik and we're not, but it looks like he's getting himself into a bit of trouble, and again it's all self-inflicted. Weird.

Time to post: White has just played 28.b4, as predicted, and has 1:07 left. Kramnik has 45 minutes (and counting).

10:30

The game has continued along largely predicted lines: after 19.b3, we've seen 19...Rac8 20.Ba3 Qc3 21.Rac1 Qxe3 22.fxe3 f6 23.Bd6 g5 24.h3 Kf7 25.Kf2.



Some comments:

(1) Anand waited for a while before playing 22.fxe3. Since 22.Rxc8 is just bad (22...Qxf2+ and 23...Rxc8 gives Black a pawn and the crucial c-file), the thought is that Kramnik offered a draw. Objectively, it's not a bad offer, but Anand rightly rejected it. He can play on without any real risk, and given the pain Kramnik has suffered the past few days, he should strive to heap on some more while he can.

(2) 22...f6 isn't bad, but it does involve forsaking another important plan, one that looks like it would even more easily achieve a draw: 22...Nb8 23.Bd6 Rc6 24.Rxc6 Nxc6 25.Rc1 Rc8 26.Rc5 a6 followed by 27...Na7 and a rook swap (maybe prefaced by ...b5) with an essentially dead draw.

(3) Kramnik's 22...f6 doesn't look any less stable. Black's shell looks completely solid, and with only the open c-file for the rooks, it's very hard to see any real possibilities for White. Barring a meltdown by one of the players, the position should crawl to a draw sooner or later.

9:45

Back to the diagram. It's altogether worth noting that Kramnik has had this position at least twice before, and in the highest of high-profile games: it happened against Topalov in their 2006 world championship match. The first time was in the infamous and amazing game 2, when Topalov played 15.Ng5, which is the most aggressive approach and that tries to exploit Black's not playing 14...Be7. That game continued 15...Re8 16.f4 Bxd3N 17.Qxd3 f5 18.Be3 Nf8 19.Kh1 and now Kramnik played 19...Rc8 and got into serious trouble, though he won after many adventures. According to Bareev in From London to Elista, Kramnik's second Rublevsky (and everyone else since, with great results) played the right response later that year: 19...Be7 20.Nf3 Rc8 21.h3 Qc7 22.Rfc1 Qd7 23.Rg1 Qc7 24.Rgc1 Qd7 25.Qb3 and draw agreed, Jakovenko-Rublevsky, Moscow 2006.

So Anand chose 15.Ne1 instead, and this too follows an earlier Topalov-Kramnik game. In their first rapid playoff game, Kramnik chose 15...Rc8 and equalized after 16.f4 Bxe1 17.Rxe1 Bg6. Bareev notes that 16.Bxh7+! would have given Black trouble with a capital T, so he notes 15...Bg6 instead, and that's what Kramnik chose in our game. This closes the diagonal and swaps off the problem bishop.

Anand swapped the bishops - 16.Bxg6 hxg6 - and then reactivated the knight with 17.Nd3. In the only other game to see 15...Bg6, Black retreated his remaining bishop to e7, but Kramnik preferred to activate the queen with 17...Qb6. White swapped: 18.Nxb4 Qxb4 and then played 19.b3, preparing Ba3-d6, aiming to give the bishop its best prospects.

Generally speaking, Black is in good shape. His king is safe, he has swapped off his worst pieces, has just about finished his development and achieved the right minor piece imbalance. There is one remaining problem for Black, though, and it's that his knight is on a poor track. It has nothing to do on d7, and there's nothing for it on b6 either. Ideally, he'll try to maneuver it to c6, but that will take a little time. First he'll want to make sure he gets his fair share of the c-file, so something like 19...Rac8 20.Ba3 Qc3 21.Rac1 Qxe3 22.fxe3 Nb8 is possible, with a fairly equal position.

9:15

Anand opens the second half of the match (though it could turn out to be a very short "half") with White, and continued his "left handed" chess. Kramnik deviated from his previous black games with a Slav, and the following moves were all played a tempo:

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 dxc4 5.a4 Bf5 6.e3 e6 7.Bxc4 Bb4 8.O-O Nbd7 9.Qe2 Bg6 10.e4 O-O 11.Bd3 Bh5 12.e5 Nd5 13.Nxd5 cxd5 14.Qe3 (1:55) Re8 (1:54), and now Anand slowed down, as 14...Re8 is comparatively rare (14...Be7 is the usual move). Here's the position:



More discussion of the position next time.

Here it is, Anand's third win in four games.

2:15 Post-Press Conference Update:

This is a little later than usual, both because the press conference took longer than usual and because I've replying in the comments. The press conference was delayed because the players had to undergo the silly doping tests, so contrary to their usual practice of having the players out together and starting with the white player's response (Anand this game), Kramnik came out first and did a solo act, followed several minutes later by Anand.

Kramnik: Called it a "strange" opening. White had a choice to not go for the ending but did, and it was equal. Or was at first, but then suddenly he had problems. Maybe, he said, he should played 17...Ba8 rather than 18...Ne7, but he wasn't completely sure. After ...c5, White was always just in time to keep the pawn or at least positional pressure. He spent a great deal of time at that point, and things got worse from there. He believed 33...a4 was the decisive mistake, but that he was much worse already.

Someone asked him about 18...c5. His response was that after 18...Re8 White plays 19.Ne5, giving Anand a slight edge in a position where Black has no counterplay. It "looked like" his counterplay should have been in time because of White's bad coordination, but since it didn't work, then it's 17...Ne7 that should have been avoided - he pretty well tied those two moves together.

Two questions were asked about the trend of the match: one about what he and his team thought to do after yesterday's game; the second about what he'd do from here. In reply to the first Kramnik's answer was understandably brief: he just said that it wasn't pleasant but it's over, so they simply looked forward. As for what next, he said that it's time to rest, try to play well, and if he can win a game, he'll see from there.

Anand: He had the 9.h3 idea in the opening, and felt that it seemed interesting and had the advantage of quickly forcing Kramnik to think on his own. After Kramnik's 9...b6, he saw nothing better than the ending that arose but believed it was a little awkward for Black on account of the pressure against c7. Like everyone else, he didn't like Kramnik's 17...Ne7. Instead of 18...c5, he noted the same line that Kramnik did but elaborated a bit: 18...Rfe8 19.Ne5, the bishops get swapped and then White puts the knight on c6. It'll get swapped off too, most likely, and White will have some pressure on Black's position. After 18...c5 Black is close to having compensation but he doesn't quite have it, and Anand felt he did a good job of keeping control after that.

Towards the end, he saw the 42.Nf6+ (semi-)trap and considered the rook ending (discussed in the updates) and the pawn ending (discussed on Foidos but not addressed in the updates; it arises in a variation after 45...Nxd2 [instead of 45...Rxd2+] 46.f7+ Kxf7 47.g8Q+ Rxg8 48.Rxg8 Kxg8 49.Kxd2), but once he noticed 42.Rd3 that settled everything. He also considered 41.Rxg7+, but liked the idea of forcing the king back to g8.

From the more substantive Q&As: He was asked about drug testing and said "It's completely pointless, of course." He did acknowledge the point, which is that it's part of the attempt to become part of the Olympic movement, which would be beneficial for chess if it occurred. But doping tests are "for another sport". Also, the slightly weird 29.Ke1 came up, and it turned out that he had (of course) seen the normal 29.e3, and saw the follow-up 29...Nc4 30.Re2 Rd8 31.Kc2 Rc8 and thought that this forced a draw by repetition with 32.Kd1, because 32.Kb3 Na5+ 33.Ka4 Nc6 was dangerous. What he had missed, as the questioner suggested and he immediately acknowledged, was 33.Ka2.

That concludes the press conference wrap-up. Later today I'll post analysis of the game for a replayable board (and will probably add some other posts), and then tomorrow is a rest day in the match. Game 7 is on Thursday, and Anand will have White again - they're doing a color switcheroo for the second half of the match.

1:20 Update:

It's over. After 41.fxg7(?) Kramnik played 41...Kg8, and now Anand had a little trap to avoid: 42.Nf6+? Nxf6 43.exf6 Re8+! 44.Kd1 Rd8, when 45.Kc2 Rxd2+ 46.Kc3 Rf2 47.Kxc4 Rxf6 is at the very least not a trivial win for White. But here Anand slowed down and found the correct 42.Rd3!, with the idea of bringing the rook to d7 and only then to play Nf6+. The rest of the game was fairly easy: 42...Ndb7 43.Bh6 Nxe5 44.Nf6+ Kf7 45.Rc3 Rxc3 46.g8Q+ Kxf6 47.Bg7+ and Black resigned, as 47...Kf5 48.Qf8+ Ke4 49.Qb4+ wins serious material.

The score is thus a crushing 4.5-1.5 in Anand's favor, and he now needs just two points in the remaining six games to win.

[Foidos note, for those who are interested: They're offering a package deal for viewers: the second half of the match can be viewed for 29 euros for the remaining games. That's a good news/bad news scenario, of course. The good news is that it's a 20 euro discount, but the bad news is that it may not be all that many games. Still, even if it's just three games, that's enough to pay for it.]

OK, back with more after the press conference.

1:00 Update:

After 39...Kf7 40.Ne4! Nc4 the time control has been made, and now it's time for Anand to work out any one of the many very easy ways to win. 41.Rxg7+ Ke6 42.Ng5+ Kxe5 43.f7 was one of them, but Anand has chosen 41.fxg7 instead. This looks clearly inferior to 41.Rxg7+, but it really doesn't matter. Still, it's not a move that forces Kramnik to resign (maybe that's the point, to make Kramnik suffer), so maybe there will be a few more moves.

12:50 Update:

Resignation can come at any moment, though I suspect Kramnik will wait until after the time control at move 40 to throw in the towel. Here are the moves since the last update:

35.Rg4 Nc4 36.e4 Nf6 37.Rg3 Nxb2 38.e5 Nd5 39.f6

Black is just dead here. After 39...g6, for instance, White can play 40.Ne4 with threats galore. A line: 40...Re8 41.Rf3 Nd3+ 42.Rxd3 Rxe5 43.Kf2 with annihilation.

12:30 Update:

We left off after Black's 27th move; here's what has happened since then:

28.f3 Nd6 29.Ke1 a5 30.e3 e5 31.gxf5 e4 32.fxe4 Nxe4 33.Bd2 a4 34.Nf2 Nd6

There's rather a lot of content packed in here, and it will take too long to elaborate everything now. Here are some highlights:

29.Ke1: This was rather bizarre! The idea was to very slowly trudge the king to f2, bring the rook to the first rank and from there probably to d1, then Bd2, e3, Rc1 and it's easy from there. It seems that simply 29.e3 would have been far simpler.

30...e5: A fascinating idea, giving up a second pawn to achieve a blockade. Black's knights get good squares and White's bishop remains stupid for a while.

33...a4: This was in keeping with Kramnik's light-squared blockading strategy, but it was a mistake for two reasons. The first is that 33...Re8! was very strong, probably regaining a pawn while maintaining a blockade. The second point is that after Anand's 34.Nf2, Kramnik's blockade is collapsing.

Both players are getting short on time, so I'm going to publish now, as every time I explain where we are, it's becoming where we were.

12:00 Update:

Here's the current position after Anand's (White's) 26.Bc1:



At first glance, Black's position looks terrific. Sure, he's down a pawn, but all his pieces are active, while White's are passive and the rook looks silly on g2. In fact, Black is pretty nearly lost. White will play Kd1 next, then f3, then perhaps e4 (if possible), and then he's up a pawn with a better position to boot. Tactical tricks like 24...Nc5 don't help: 25.Kd1 Nxd3 26.exd3 (but not 26.Kxc2?? Ne1+) give White a technically won game.

Kramnik just played 26...f5, trying to at least make e4 difficult (after Kd1 and f3), but in the long run Black is toast. Final move update: 27.Kd1 Rc8; times: Anand: 40 minutes, Kramnik 19 minutes and counting.

11:55 Update:

I suspect people watching on the chess servers are going crazy wondering what in the world Anand has just done. The game continued 21...bxc5 22.Rxc5 Ne4 23.Rxc8 Rxc8 24.Nd3 Nc5, and now for no obvious reason the display board picked up White's move as 25.Bc3(??). Of course he played 25.Bd2, which has finally been corrected as a result of Kramnik's reply, 25...Rc2. Anand immediately responded with 26.Rc1. I'll comment about this momentarily.

11:30 Update:

Hmm, maybe 18...c5 wasn't so good. (Serves me right for trusting Seirawan, who was still thinking about the Karpov interview, while I was busy writing about it and trying to catch up.) He's not losing or anything yet, but the sac was unnecessary and now Kramnik has to suffer.

After 18...c5 19.dxc5 Black has several options, including at least 19...dxc5 and 19...a5. Let's look at each move in turn:

(1) 19...a5 20.Bd2 Ne4 and now 21.Ng5! is best. White has a slight but annoying advantage after 21...Nxd2 22.Bxb7 Rc7 23.Kxd2 Rd8+! 24.Kc2 Rxc5+ 25.Kb1 Rxg5 26.Rgd1 Rxd1 27.Rxd1 g6.

(2) 19...bxc5 is another try. Now 20.Rxc5 is poor on account of 20...Ne4, so 20.Bxc5 is right. Black can regain the pawn with 20...Ne4 21.b4 Nxc5 22.bxc5 Bxf3 (not 22...Rc7 23.Nd2 Bxg2 24.Rxg2 Rfc8 25.Nb3 and White keeps the pawn) 23.Bxf3 Rc7 24.Kd2 Rfc8 25.Ke3 Rxc5 26.Rxc5 Rxc5 27.Rd1 g5 28.Rd7 Nc8 and White still has an enjoyable edge.

So Kramnik chose door #3 after 16 minutes and played 19...Rfd8, but after 20.Ne5! White is clearly better. Now 20...Bxg2 21.Rxg2, and Kramnik has been thinking about this position for another 17 minutes. (Current times: Anand 54 minutes, Kramnik 37 minutes and counting.)

There are still chances, but Kramnik's in at least a little trouble again.

11:10 Update:

Now back to chess. The last move mentioned here was Anand's 16.Rg1, since then we have 16...Rac8 (16...Rfd8 first might be a slight improvement, trying to induce 17.e3 first, which might soften White up on the long diagonal while partially hemming in the Bd2) 17.Bg2 Ne7 18.Bb4 and now the dynamic 18...c5. This temporary pawn sac was not forced, but it looks pretty good. Anand played the obvious 19.dxc5, and that's where we are now. Anand has 59 minutes left, and Kramnik has been eating up a lot of time the last several moves and is now down to 56 minutes and counting.



I'll offer some commentary on this in a few minutes.

11:00 Update: Karpov review

Here's a summary of what Karpov had to say.

After discussing what he's doing (going around here and there representing Russia and starting various chess schools), he offered various opinions about contemporary chess.

* He doesn't like knockout events either for the world championship or for world championship cycles.

* Thinks 12 games is too short; 16 is about right.

* Thinks chess has changed over the last few years, but not so much because of the computers but the time control. It makes the game younger. Also, the loss of adjournments has hurt the quality of the game.

* He doesn't see anyone dominating chess in the foreseeable future like and Kasparov did. Maybe, he thinks, this is due to computers, but he's not sure.

* On current ratings, he thinks it's clear that there's inflation. He also thinks that ratings are generally overrated.

* Here's something bound to be controversial: he thinks Anand has become more mechanical thanks to his work with computers. Maybe he plays more accurately, but he has lost some fantasy, some creativity.

* Asked about creative players of today, he only referred to Morozevich, and this after prompting from Seirawan. Noted Morozevich's willingness to risk, and that this allowed him to sometimes have very good results and occasionally very bad results.

* About young players, he opined that Rajdabov had "missed his best time" already, by which he seemed to mean that his development had reached a premature stagnation, such that it's unlikely he'll develop to the very highest level of play. He was very positive about Carlsen, but felt that he is maybe missing some parts of his chess education. Had Carlsen been in Russia from an early age, Karpov feels that he'd have already become world champion.

* Karpov doesn't like what's going on in FIDE. The time controls and various rules and regulations (like the "stupid" mobile phone rule) are awful. He thinks that on some topics, world champions should have a right of veto. He further thinks there should be different rules for swisses, round-robins and world championship events. Certain blitz rules and their enforcement should also be changed. Finally, in FIDE, stronger chess countries should have more votes; it shouldn't be a straight democracy where a country with almost no players and no strong ones should have the same influence as a Russia, Germany or United States.

And with that, he left, but Foidos watchers can now find him in the Russian commentary room.

10:15 Update:

The moves since last time (for those new today, Anand has White):

11.Rc1 Bb7 12.a3 Bxc3 13.Bxc3 Qd5 14.Qxd5 Nxd5 15.Bd2 Clock times: Anand 1:14(!), Kramnik 1:34 and counting.

Anand's 11.Rc1 transformed the game into a more traditional 4.Qc2 Nimzo, in that the queens came off and the question now is whether Black can do something with his activity. If he can't, then White can use his bishops for the long run.

Kramnik has just moved, 15...Nf6, and this is both a good move and simultaneously a threat: 16...Nxd4. Karpov, who is now in the booth with Seirawan in the English language commentary booth in the Foidos broadcast, recommends 16.g5 Ne4 17.Bf4 Rac8 18.Bg2 with an edge. Anand chose 16.Rg1 instead, and both moves seem to give White a very small edge. And now it's time to post, so I can listen to Karpov!

9:45

Good morning and welcome once again! This game will complete the first half of the match, and it begins with Anand enjoying a 2-point lead, 3.5-1.5. For Anand, a draw is pretty good and a win just about ends the match; for Kramnik, it's still a little too early to do something stupid in a desperate bid for victory, but if he can safely introduce some aggression into the game he certainly should. Let's see the moves (Anand has White):

1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Qc2 d5 5.cxd5 Qxd5 6.Nf3 Qf5 7.Qb3 Nc6 8.Bd2 O-O 9.h3N b6 10.g4 Qa5

Some initial comments:

(1) Anand's choice of the 4.Qc2 Nimzo-Indian at least seems more conventional, and it's a good choice. White's losing chances are quite low in most of the main lines, so Anand can play for a win "for free".

(2) The line chosen by Kramnik, the Romanishin Variation (4...d5 - 6...Qf5) is considered pretty solid for Black, and according to theory White's most principled move is to trade queens. Anand's choice of 7.Qb3 is feistier, giving both sides more chances to stir up mischief.

(3) Anand's 9.h3 is a very interesting novelty. The point is obvious but no less strong for that: White wants to play g4. Black's queen will have problems finding a safe home. And it's not clear that 9...b6 was the best reaction.

(4) One fantastic line: Seirawan thought that 10...Qg6 was impossible due to 11.Nh4 Nxd4 12.Qc4, winning a piece. But White's position is funny somehow, and Black has tactical resources: 12...Qc2 13.Qxb4 (Seirawan stopped here, but) 13...c5! 14.Qa3 Ne4! 15.Nxe4 Qxe4 16.f3 Nc2+ 17.Kd1 Qc4 18.b3 Ne3+! 19.Bxc3 Qc3 20.Qc1 Rd8+ 21.Bd2 Rxd2+ 22.Qxd2 Qxa1+ 23.Kc2 Qxa2+ 24.Kc1 Qa1+ with equality. This doesn't mean that 10...Qg6 is good, because if White doesn't hurry with Nh4 the Black queen remains vulnerable and out of play on g6. Unfortunately for Kramnik, the queen may not be thrilled with life on a5, either, so it's nervous time for Kramnik fans.

Time so far: Anand is at 1:29 and counting, Kramnik at 1:45. For once, Kramnik is ahead on the clock in the opening, but it might be because Anand is looking for the kill.

Earlier today, Anand won his second game to take a big 3.5-1.5 lead in his world championship match against Kramnik. It was another great game in the Meran Semi-Slav, but it came to a screeching halt when Kramnik fell for a very neat trap - but one within his pay grade to figure out and sidestep.

Kramnik is now in biiiiig trouble, and if he loses tomorrow he can just about book a plane for Moscow (or Paris) for this weekend. If only for the sake of excitement, let's hope Kramnik overcomes his chess version of "the yips".

Meanwhile, here's today's game, with my notes. (On a replayable board, with deeper commentary than in the live blog post.)

12:25 Post-Press Conference Update:

Most of the questions covered points already made in the analysis. Kramnik's initial summary of the game noted that 15...Rg8 was a novelty that allowed 16.Bf4. He called Black's 16...Bd6 17.Bg3 f5 plan "tricky", and felt that after 18.Rfc1 he was better, though he never saw a way to get a big advantage. Anand felt his plan of ...f5, ...f4, ...Be7 with the idea of getting his king on e7 was a good one, though of course it had some risk. He didn't see how White could attack (too seriously) without the dark-squared bishop, and if he central pawn mass can get going he's in good shape. Most of the further analytical point were already addressed in the earlier comments, so for now (before the "official" game presentation post) I'll mention Kramnik's comment about why 18.Rfc1 rather than 18.Rfd1, in response to a question from GM Norwood: its point is to restrict Black's queen, to not allow something like ...Qc5-d5, when Black's attacking chances become extremely dangerous.

12:00 Update:

Oh noooooooooooooooooooo! Kramnik avoided it two moves ago, but now fell for it! He played 29.Nd4??, and they blitzed out 29...Qxd4 30.Rd1 Nf6 31.Rxd4 Nxg4 32.Rd7+ Kf6 33.Rxb7 Rc1+ 34.Bf1. Here Anand reached out to play 34...Ne3, then retracted his hand to quadruple check to make sure it works (we probably all know the feeling!). Only then did Kramnik realize what was happening, and he slumped back in his chair. Anand made the move and got up; Kramnik stopped for a minute, no doubt in great misery and embarrassment, and finally played 35.fxe3. Anand returned, played 35...fxe3, and that was it: Kramnik resigned. He can stop ...e2 only by playing 36.Rc7, but after 36...Rxc7 there's nothing to play for.

So Anand now leads 3.5-1.5, and Kramnik is in deep, deep trouble in the match. The score is heavily against him, he's getting outprepared, and although he is making many excellent moves, he's also blundering a great deal.

Check back in a few minutes for the post-press conference update.

11:45 Update:

After 26.Nf3, Anand played 26...Qf6, setting a nasty trap.



After 27.Nxd4? White loses: 27...Qxd4 28.Rd1 Nf6 29.Rxd4 Nxg4 30.Rd7+ Kf6 31.Rxb7



31...Ne3!! and it's lights out: 32.fxe3 fxe3 and White must throw away the rook with 33.Rc7 to stop the crushing ...e2. Kramnik didn't fall for it and played 27.Re1, but it cost him 6 minutes on the clock, and he's down to 19 minutes for his next 13 moves.

11:35 Update:

The last moves are 22...Rac8 23.Rxc8 Rxc8 24.Ra1 Qc5 25.Qg4; the times are 25 minutes for Kramnik, 59 minutse and counting for Anand. Black's 22nd did small favors for both sides. White benefited from the trade of rooks and by Black's rook(s) quitting the g-file, as his king is significantly safer. On the other hand, Black has gained some time and has the c-file, not only for the rook but also to allow the queen access to c5. This centralizing move helps not only the queen but prepares ...Ne5 in some cases as well. White in turn is searching for whatever play he can get with 25.Qg4. White cannot afford to allow Black's central pawn mass to be consolidated, as he'll then be positionally lost or nearly so.

The poisition is approximately equal, and there are some perpetual lines in the offing here. One is 25...Qc2 26.Qxf4 d3 27.Nf5+ exf5 28.Re1+ Kf8 29.Qh6+ Kg8 30.Qg5+ Kf8 and so on.

I was about to post, but now I see that Anand has played 25...Qe5, and Kramnik has zipped out 26.Nf3. This looks pretty okay for White, and at least there aren't any perpets in sight.

11:05 Update:

For those of you analyzing at home, the position after 21...Ke7 is a rich one. White can consider many candidate moves, including 22.Bxd7 (22...Kxd7 23.b4 and 23.Qh5 both deserve a look), 22.Qh5, 22.Kf1, 22.b4 (but 22...Rxg2+ 23.Nxg2 Rg8 looks pretty good for Black) and 22.Ra3, as played by Kramnik. Another good move by the Russian, but he's down to 32 minutes.

11:00 Update:

It's hard to comment on the game in any useful way, as almost everything now has to do with tactics. Both sides have attacking ideas, and need to employ them without getting their own king mated. White has pressure on the bishops' scissoring diagonals, the rook is strong on the c-file, where it keeps Black from castling and could someday penetrate, and White's queen has various possible avenues of egress (e.g. h7 via d3). Black's attacking punch comes from the pressure exerted by the Bb7 and the Rg8, as we saw repeatedly in game 3. Beyond these generalities, useful as they are, one must calculate like crazy to work out the details.

After Kramnik's 18.Rfc1, Anand replied with 18...f4 pretty quickly, indicating that he's almost surely still in his preparation, and after 19.Bh4 played 19...Be7. This surprised the comentators (GMs Seirawan and Rogers), not because Black's dark-squared bishop isn't bad (it is) and not because giving Black's king a little more wiggle room and the chance to get off the back rank isn't to his benefit (again, it is). Rather, it's that general exchanges may favor White, whose king at least has a safe home.

After 19...Be7, two tries that look plausible but get nowhere are 20.Bxe7 and 20.Qd3. After 20.Bxe7 Kxe7 21.Bxd7 Kxd7 22.Ne5+ Ke7 23.g3 it's Black's attack that's faster, not White's, after 23...fxg3 24.hxg3 Rg5. As for 20.Qd3, hitting d4 and h7, Black has excellent resources in this case as well: 20...Rd8 21.Qxh7 Rxg2! 22.Kxg2 Qxb5 and Black seems to be better, maybe even close to winning. 23.Qg8+ Bf8 24.Bxd8 loses immediately to 24...Qh5, while the better 23.Qh8+ Nf8 24.Bxe7 Kxe7 25.Rc7+ Nd7 26.Qh4+ f6 27.Qh7 Kd6 is still a good deal better for Black.

So Kramnik chose a better move, 20.a4, avoiding the problem of the loose bishop and clearing a3 for a rook lift. Anand finally started to think here (Kramnik was already down to 52 minutes), and after fifteen minutes thought chose 20...Bxh4, and after 21.Nxh4 played 21...Ke7. (21...Qd6 is better, but 22.Ra3 and 22.a5 might give White the beginnings of an advantage.)

I'm not sure what's going on here, but I think that for now White has avoided the biggest dangers of Anand's preparation. The times are 44 minutes and counting for Kramnik (to complete his 40th move), 1:23 for Anand, so that's likely to be a factor as the game goes on. Kramnik has played very well so far, but it's a shame that he has to keep figuring everything out over the board while Anand can sleep each game until near move 20.

Current position:



10:10 Update:

Still no new moves, but I should mention some other possibilities. 18.Rfd1 looks logical, though "logic" isn't enough to overcome concrete factors, and after 18...f4 19.Bh4 Ra5 20.Bxd7+ (20.a4 Rxb5! 21.axb5 Ne5) Kxd7 Black looks very good, with all his pieces in the game and heading for White's king.


18.Nxd4 looks much more human and I'm sure Kramnik is considering it very carefully. There are at least two possibilities here: 18...Rxg3 and 18...f4. On the first, 19.hxg3 Qxd4 20.Rfd1 Qb6 21.Rd2 Ba6 22.Bxa6 Qxa6 23.Qxa6 Rxa6 24.Rad1 Ke7 25.b4 Nf6 26.b5 Rb6 27.a4 Ne4 28.Rc2 with approximate equality. On 18...f4, 19.Nxe6 fxe6 20.Qxe6+ Kf8 gives White enough attack for a draw, but I don't know about more.

A move has just happened: White has played 18.Rfc1, a rather abstract-looking move that no one but Rybka recommended. More on this next time.

10:00 Update:

Kramnik is already in deep thought, and if he loses this game (and maybe even if he doesn't) he's going to give his seconds some serious heck for not preparing this. So what now? 18.Ne5 makes good sense positionally, but as Seirawan's viewers on Foidos (very entertaining!) have hastened to tell him, 18...d3! makes White's life very difficult. After 19.Bxd7+ Ke7, White has to step very carefully. If 20.Nxd3, then 20...Bxg3 21.hxg3 Rxg3 22.Nf4 Rag8 and White looks like toast. 20.Qxd3 is probably better, and after 20...Bxe5 21.Qb5 may just keep body and soul together.

9:30: First post

It's speed chess so far - in the first five minutes, the players repeated the first 15 moves from game 3. Or rather, the first 14 and a half moves: 1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 e6 5.e3 Nbd7 6.Bd3 dxc4 7.Bxc4 b5 8.Bd3 a6 9.e4 c5 10.e5 cxd4 11.Nxb5 axb5 12.exf6 gxf6 13.O-O Qb6 14.Qe2 Bb7 15.Bxb5. Last time Anand played 15...Bd6, but he varied before allowing Kramnik to show whatever improvement he had in mind; this time, he played 15...Rg8.



As mentioned when covering game 3, the "old" move 15...Bd6 prevented Bf4, when the bishop performs double duty. It can be active there, and it can also drop back to g3 to neutralize Black's pressure along the g-file. So after a minute's thought, Kramnik indeed chose 16.Bf4, after a minute's thought in reply, Anand played 16...Bd6, and now Kramnik has started to think. His body language hasn't exactly exuded confidence, that Anand's choice was something he and his team had investigated over the past three days. Anand, on the other hand, looks reasonably relaxed, though the more I look at him the more antsy he seems to be.

Kramnik took about 14 minutes here, perhaps considering variations starting from 17.Bxd6 Qxd6 18.Rfd1, but finally chose the obvious, safe (or at least safe-looking) 17.Bg3. Perhaps at some point Black will try to prise open the g-file with ...f6-f5-f4 or h5-h4, but for now White has bought his king a little time for other projects.

Sure enough, Anand has played 17...f5, and now it's time to upload this post.

The ideas and analyses from the live coverage post are expanded on here, and you'll find at least one interesting correction of something Kramnik claimed in the post-game press conference. Curious? Have a look!

12:00 Update: Press conference edition:

The players agreed that Black was always fairly comfortable and never had any real problems. They seemed to agree that each player had one moment when they could have played slightly more accurately, but that it didn't amount to much in either case. Black should perhaps have played 18...d4 rather than 18...Nc5, while White should have preferred 21.Rac1 to 21.Rd4. One interesting line mentioned by Kramnik came near the end, when considering 28...Nxd4 instead of 28...Rxd4. The hope is to achieve some play on the kingside, but Kramnik pointed out that after 29.Kh1 h4 White has 30.gxh4, with the essential point that the otherwise devastating 30...Nf3, threatening both 31...Qxh2# and 31...Rxd3 winning a rook, is refuted by 31.Qxg4 - check - and then 32.Qxf3.

11:30 Update: DRAW AGREED:

No surprise here. The game continued 28...Rxd4 29.Rxd4 Rxd4 and the players agreed to a draw, due to variations like the one demonstrated in the previous update. A very correct game, at first glance, though of course not as impressive as yesterday's very complicated affair. Anand now leads 2.5-1.5; game 5 will take place on Monday. (Stay tuned for one more update, after the press conference.)

11:15 Update:

I just had some company over, so I haven't been able to look at the board since the last update. (On the other hand, I was showing my friend the game, and predicted a position very similar to what has shown up on the board, so nothing too weird has happened.) Here are the moves:

21.Rd4 a6 22.Qb4 h5 23.Bh3 Rad8 24.g3 g5 25.Rad1 g4 26.Bg2 Ne6 27.R4d3 d4 28.exd4

Quick comment on the current position: Black has finally achieved his thematic break, and now unless White has something very concrete, Black is absolutely fine. Here's a sample line that ends quite satisfactorily for Black: 28...Rxd4 29.Rxd4 Rxd4 30.Rxd4 Nxd4 31.Qxb6 Qe1+ 32.Bf1 Nf3+ 33.Kg2 Qe4 34.Qxa6 and Black has perpetual check.

10:30 Update:

Here are the intervening moves since last time:

14.Nd4 Ne6 15.Nxf5 Qxf5 16.O-O Rfd8 17.Bg4 Qe5 18.Qb3 Nc5 19.Qb5 b6 20.Rfd1 Rd6

Blockading the d-pawn was the first order of business, and the knight is the classic blockader. That explains White's 14th, and it explains Black's rejoinder as well. Black wants to swap off the isolated d-pawn, if possible, and if not to give White his own isolani in turn.

After 15.Nxf5 Qxf5 16.O-O Rfd8, Black is ready to play ...d4, and if White wants to find any basis for continuing the game he'll have to prevent it or gain some other advantage in the meantime. 17.Bf3 doesn't achieve anything, as 17...d4 18.exd4 Nxd4 19.Bxb7 Rab8 regains the pawn with full activity. (Note that this last move shows why Black plays 16...Rfd8 instead of 16...Rad8.) Therefore White looks for a different avenue of play with 17.Bg4 Qe5 18.Qb3, hassling Black's queenside pawns.

After 18...Nc5 19.Qb3 b6 20.Rfd1, White has still managed to prevent ...d4 (20...d4? 21.b4), so Kramnik played 20...Rd6. This seems to make ...d4 playable (21...d4 22.b4 a6 and the b6 pawn isn't hanging), prepares to double rooks on the d-file and, in some circumstances, gets ready to swing the rook over to the kingside in search of possible counterplay there.

Here's the current position:



9:45

After the excitement of yesterday's game, the players have found their way to a more sedate position this time around. Here are the moves so far:

1.d4 Nf6 2.c4 e6 3.Nf3 d5 4.Nc3 Be7 5.Bf4 O-O 6.e3 Nbd7 7.a3 c5 8.cxd5 Nxd5 9.Nxd5 exd5 10.dxc5 Nxc5 11.Be5 Bf5 12.Be2 Bf6 13.Bxf6 Qxf6

Some brief comments.

(1) No Nimzo-Indian today! What that means is unclear, but it's at least worth noting.

(2) 5.Bf4 isn't as popular as 5.Bg5, but it's also important and common. Oversimplified, the advantage is that the bishop is better on f4 than on g5, and the disadvantage is that Black easily achieves ...c5. Black winds up with an isolated pawn, but in a situation where it's generally a pretty minor disadvantage.

(3) 6...c5 is both more popular and more sharp than Kramnik's choice, when the main line continues 7.dxc5 Bxc5 8.a3 Nc6 9.Qc2 Qa5 10.O-O-O. This has occurred in hundreds if not thousands of games, including a 2002 Anand-Kramnik game! That was an "Advanced Chess" game (i.e. man + machine), and wound up a lively draw. Kramnik's choice is more solid, but has the drawback that the knight is more passive on d7 and leads to an IQP (isolated queen pawn) position where he can only draw.

(4) Kramnik's 12th move was not yet a novelty, but is rare - as far as I can tell, it has occurred just twice before, in a couple of 2007 games by GM Goloschapov.

Pretty heavy notes, too, including one quite important line offered in the comments to my live coverage post. Have a look at Anand's big win, right here.

Post-game conference update:

Kramnik clearly hadn't taken 33.Kb3 seriously. As best as I could tell, he thought 33...Rc1 34.a5 e5 won, but after 35.Ra4 (forced) Qc5 36.Bc4 (again forced), there is no win in sight. Anand saw this and felt that the position was "getting out of control", while Kramnik was dismissive about the whole thing, calling it (if it worked) a "miracle draw". He felt that his last best chance was 32.Rd3, which does seem to give him a defensible position.

1:00 Update: Anand wins.

A great game for Anand, but a horrible finish for Kramnik. After defending brilliantly for many moves, he made an incomprehensible blunder to lose - and with sufficient time on his clock to boot. After 29...Rg1+ (as given in the 12:30 update), play continued 30.Kd2 Qd4+ 31.Kc2 Bg4 (31...Bf5+ was discussed in the 12:35 update) 32.f3 Bf5+ and now White can and must use the idea mentioned in the 12:35 update: 33.Kb3 Rc1 34.a5 R/Bc2 35.Qxc2 B/Rxc2 36.Kxc2 Qc5+ 37.Kb1 Qxb5 38.a6 and Black cannot win.

Instead, Kramnik played 33.Bd3?? which loses to both the elegant 33...Bxd3+ 34.Rxd3 Qc4+ 35.Kd2 Qc1# or the longer but also winning approach chosen by Anand in the game, 33...Bh3 when White lacks an adequate defense to ...Rg2. Kramnik tried to run the a-pawn, but it doesn't work this time: 34.a5 Rg2 35.a6 Rxe2+ 36.Bxe2 Bf5+ 37.Kb3 Qe3+ 38.Ka2 Qxe2 39.a7 Qc4+ 40.Ka1 Qf1+ 41.Ka2 Bb1+ and White resigned. Ugh.

12:35 Update!

Aha, I've figured out what Kramnik might have in mind. After 31...Bf5+ in the line below, White plays 32.Kb3, and after 32...Rc1 33.a5 Bc2+ 34.Qxc2 Rxc2 35.Kxc2 Qc5+ 36.Kb1 Qxb5, Black has won lots of material but has nothing better than a draw in response to 37.a6. The a-pawn is too strong.

12:30 Update:

Since last time, the game has continued 27.a4 Bg2+ 28.Ke1 Bh3 29.Ra3 Rg1+, and now Kramnik is in trouble. I think 29.Rd1 was forced, when the game would have remained unclear. Now Black is at least clearly better after 30.Kd2 Qd4+ 31.Kc2 Bf5+, which will cost White at least the exchange. Uh oh!

12:00 Update:

Line (4) from the 11:30 update occurred: 22...Rg7! 23.Rxg7 Kxg7 24.gxf4 Rd8! 25.Qe2?! Kh6, and now White is in danger. (25.Qb3 was better, when the position remains unclear.) A possible disaster: 26.Bd3(?) Rg8+ 27.Kf1 Bg2+ 28.Ke1 Bh3 29.Rb1 Qb4+ 30.Kd1 Bg4 31.f3 Bf5 32.Bxf5 exf5 and Black's queen and rook are killing.

Instead, 26.Kf1 looks right, and after 26...Rg8 27.f5! with the idea of swapping queens by 28.Qe3+. Black can prevent this with 27...Bg2+ 28.Ke1 Bh3 (idea: ...Rg1+), and White's king must streak across with 29.Kd2 Qd4+ 30.Kc2. If Black has nothing better than 30...Rg2 31.Qd2+ Qxd2+ 32.Kxd2 Rxf2+ 33.Kc3 Bxf5, then White can probably breathe a sigh of relief, as 34.a4 keeps the game balanced. But this may not be easy for Kramnik to find, as time pressure approaches, and I wouldn't be surprised if Black has other, more dangerous tries. (Note: 26.Kf1 Rg8 has been played. Current time: 22 minutes and counting for Kramnik to reach move 40, 28 for Anand.)

11:30 Update:

A few moves, but not too many. Anand had a choice between 19...Rg6 (discussed in the last update) and 19...h5, which he chose. Kramnik's moves came pretty much immediately: 20.Nxe6 fxe6 21.Rxd7 Kf8 22.Qd3!, and now Anand is again deep in thought (13 minutes and counting). He has several possibilities here, of which I'll mention two bad ones and two good ones:

(1) 22...Bh6? 23.Qh7 Bg7 24.Qxh5 with a forced mate.

(2) 22...f5? 23.Qc3 e5 24.Qb4+ and again a forced mate is in the offing.

(3) 22...Bxg3 23.hxg3 h4 24.Rd6 (24.Qd6+ is also playable, leading to a drawn rook ending) 24...Qc5 25.b4 Qe5 26.Rd8+ Rxd8 27.Qxd8+ Kg7 28.Qe7+ Kh6 29.Qf8+ Rg7 30.Qh8+ Rh7 31.Qf8+ with perpet.

(4) 22...Rg7! (Quite possibly best, returning the piece but gaining serious counterplay.) 23.Rxg7 Kxg7 24.gxf4 Rd8! 25.Qe2 Kh6 with the idea of ...Rg8+. White has to be very careful here.

11:00 Update:

No new moves! Anand has been thinking for the past half an hour; obviously 19.Rxd4 rather than 19.Nxd4 was what he had prepared for - assuming he had seen 18.Bf4 in advance. It's easy for White to get in danger here, but with best play he seems to have compensation. Here's the current position, followed by some sample lines:



19...Rg6 (19...h5 is also possible) and now:

(a) 20.Nxe6 fxe6 21.Rxd7 Kf8 and here White must be careful:

(a1) 22.Rxh7? Bxg3! 23.hxg3 Rxg3+ 24.Kf1 Rg5 and the threats to the Bb5 and to play ...Bg2+ Ke1 Re5 give Black a decisive advantage.

(a2) 22.Qh5? Bxg3! 23.hxg3 Rxg3+ 24.Kf1 Bg2+ 25.Ke2 (25.Ke1 Qa5+ wins) 25...Bf3+ 26.Qxf3 Qxb5+ is lights out.

(a3) 22.Bd3 Rg5 23.Rxb7 Qxb7 24.Be4 Qb8 25.Bxa8 Qxa8 26.Qxe6 with an unclear position.

(b) 20.a4! and now it's Black who needs to exercise care:

(b1) 20...O-O-O? 21.a5 Qa7? 22.Qc4+ followed by 23.Nc6 with a winning position.

(b2) 20...Bd5 21.Nxe6 Qxe6 22.Rxd5 Qxe2 23.Bxe2 with a very unclear ending.

10:30 Update:

Very little has happened since the initial update. Kramnik played 17.g3 after 12 minutes, Anand instantly replied 17...Rg4, and then Kramnik went into the tank for 26 minutes before playing the ingenious and brilliant rejoinder 18.Bf4!!. Anand seemed a little surprised by this - in an unpleasant way - but quickly played 18...Bxf4. Now a surprise: instead of the obvious and logical 19.Rxd4, Kramnik played 19.Nxd4. (This might be a mistake.)

We'll see what happens here, but as evidence of Anand's deep concept, here's a possible line from Illescas (with a little help from Seirawan) after 18.Nd2, Black plays 18...Ke7! 19.Bxd7 Rag8!! with huge compensation. After 20.Bb5 d3 21.Qxd3 Bxg3 22.hxg3 Rxg3+ 23.Kf1 Rxd3 24.Bxd3 Qd4 Black is in good shape thanks to his attacking chances and the h-pawn's eagerness to scamper down the board.

10:00

We're up and running, and it's again Anand who has done the better job of preparing. Here's the game so far: 1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 (Of course no Exchange Variation this time.) 4...e6 5.e3 (Avoiding, at least for now, the sharpest lines with 5.Bg5.) 5...Nbd7 6.Bd3 dxc4 7.Bxc4 b5 8.Bd3 a6 9.e4 c5 10.e5 (Very old fashioned; 10.d5 is the main line here.) 10...cxd4 11.Nxb5 axb5 12.exf6 gxf6 13.O-O Qb6 14.Qe2 (All known theory, but now comes a bombshell:) 14...Bb7.



This isn't quite a novelty, but it is for professional chess. Black doesn't even bother with the b-pawn, but gets on with his development as quickly as possible. The usual moves were 14...b4 and 14...Ba6. After 10 minutes, Kramnik played 15.Bxb5, and Anand almost immediately played 15...Bd6, a good move that prevents, among other things, the White maneuver Bf4-g3, activating the bishop and shoring up the White king against attacks on the g-file. It might seem like White can play 16.Nxd4 here, when 16...Qxd4 17.Rd1 gives him tremendous possibilities. However, Black has the much better 16...Rg8!, and after 17.Nf3 Ke7 18.Bxd7 Qc5!!, with the idea of ...Qh5, is a powerful suggestion by Illescas that looks 0-1-ish.

So, after 23 minutes' thought, Kramnik chose 16.Rd1, Anand played 16...Rg8, and that's where we are now.

The game was covered live in this post, but here's the game with an expanded set of notes.

(Every half hour or so I will update the moves and commentary for this, game 2. Anand has White.)

Post-script: Tonight's ChessBase show will recap the first two games of the match. Be sure to join me in the Broadcast Room in Playchess.com at 9 p.m. ET (look for Anand-Kramnik recap in the Games list) - I'll see you there!

1:00 Update:

DRAW?!?!?!?! White played 31.Bc2, which quite possibly a mistake. The idea of bringing the bishop back into play is logical, but for the moment it disrupts his coordination - maybe 31.Rb2 deserves consideration. Kramnik played the obvious 31...Ne6 (31...Rh6!?), threatening moves like ...c5 and ...Nd4+, but Anand found a very nice reply: 32.Kg3!. This not only prevents 32...Nd4 from being a check that allows Black to reach opposite-colored bishops, but also gives him the pleasant reply 33.Bd1, bouncing the bishop to g4. 32...c5 looked natural, but Kramnik instead chose 32...Rd4 and offered a draw, and incredibly Anand accepted! White is still up a pawn, and while Black has some compensation, there doesn't seem to be any objective reason for White to give up on the game. One relevant factor that might explain it is that he was down to 2:30 or so when he accepted, but it's still a surprise.

The Final Position after Kramnik's 32...Rd4:




12:30 Update:

Most of the line I suggested in the previous update occurred, but after 24...Ne5 25.Nf3 Nh5+ 26.Kf2 Kramnik preferred 26...Nxf3 27.Kxf3 e5, rightly keeping lines closed on the queenside. Black's 27th gives Black's knight the f4 square, which it can use as a pivot point on the way to e6 (and then d4) or d3. Fundamentally though, Black is just trying to survive: Anand's extra pawn and bishop pair have Kramnik in a world of trouble. White continued with 28.Rc1, protecting the c-pawn, and after 28...Nf4 activated the Ra1 with 29.Ra2.

Now Kramnik has a fundamental choice to make: 29...Ne6 or 29...Nd3? The former tries to keep more activity, the latter tries to reach chances of salvation in an opposite-colored bishop ending. I doubt either try should hold, but Kramnik has chosen 29...Nd3, and now Anand has thwarted Kramnik's hopes with 30.Rc3. Taking on b4 would be hopeless, fixing White's queenside pawns, so Kramnik retreated with 30...Nf4. Maybe that's not as horrible for Black as it looks, since ...Rd1 is maybe half a threat, but it's nothing to feel good about either.

12:00 Update:

Only one move from each side: 22.Bb1 h5. Anand's move avoided all the mischief that could arise after something like 22.Rhd1 h5 23.h3 Rxd3 24.Rxd3 Nxe4+. After 22.Bb1, Black's knights look a little stupid compared to how they'd be after 21...Nge5, when one possibility is 22.Be2 Bxc4 23.Bxc4 Nxc4 24.Rhc1 Nde5 25.Nxe5 Nxe5 26.Rc5 with a very, very slight advantage for White.

The play after 22.Bb1 h5 is fairly forcing - and is starting to show up: 23.h3 h4+ 24.Nxh4 and now probably something like 24...Ne5 25.Nf3 Nh5+ 26.Kf2 Nxc4 27.Ba2 will occur, when White's bishop pair and Black's four(!) isolated pawns seem to give Anand very good chances. 21...Ndf6 looks like a positional blunder - maybe he just missed 22.Bb1.

11:30 Update:

The players have traded queens, but the game is heating up! Here's what has happened since last time: 18...O-O-O 19.Qxe3 Nxe3 20.Kf2 Ng4+ 21.Kg3 Ndf6



White's bishop on d3 and pawns on c4 and e4 are vulnerable, and additionally Black might have ideas like ...h5 followed by ...Rxd3, ...Nxe4+ and going for an attack on the king. There's a lots to analyze here, and I'll hopefully have a lot more to say about it in the next update.

11:00 Update:

Things are continuing to proceed slowly. Kramnik chose 16...Ng4, and the next move and a half were obvious and expected: 17.Bb4 Qe3+ 18.Qe2, and that's where we are at the moment. Here's a non-computer checked line where things could go: 18...c5 19.Bd2 Qxe2+ 20.Bxe2 Nge5 21.Rc1, which looks slightly better for White. Black would like to continue with 21...Rb8, but it has a direct tactical problem: 22.Nxe5 Nxe5 23.Bf4. So Black might consider 21...Nxf3+ first, and after 22.gxf3 Rb8. It looks pretty dangerous though - Black has the b-file, but White has the bishops and a better structure. Stay tuned!

10:30 Update:

Until about two minutes ago, there was almost nothing to update. Black played 14...Ba6, very logically trying to swap off one of White's bishops. Maybe White could have tried something frisky like 15.Ng5 in reply, but after a fairly long think he played the obvious 15.c4, shutting out Black's bishop and avoiding the trade. Black moved the attacked queen with 15...Qc5, threatening ...Qe3+ and to capture on e4 in response to 16.Bb4, so White played 16.Bd3, and that's where we stand at the moment. Castling now looks reasonable for Black, and another idea for Black might be ...Ng4-e5. Something like the following looks possible: 16...O-O 17.Bb4 Qe3+ 18.Qe2 Qxe2+ 19.Kxe2 c5 and ...Nb6.

10:00 Update:

As is usual, once the players are out of their prep, things slow down. The game continued 12.c6, a typical idea whereby White returns the doomed extra pawn to ruin Black's structure. Black took, 12...bxc6, and then 13.Nf3 Qa5 14.Bd2 brings us up to the minute. This position has occurred once before, in a correspondence game Gilman-Lepsenyi that was won by Black, but for the moment it's White who should have a very slight upper hand thanks to the long-term chances given by the bishop pair. In return, Black has the slightly better development (for the moment) and nice control over squares like e5. It should be very interesting!

9:30: Initial thoughts:

Wow! Anand has gone heavily against type by opening with 1.d4! He has been an almost exclusive 1.e4 player in his career, so this is certainly a surprise for Kramnik. In Kramnik's 2004 world championship match, he was similarly shocked when Peter Leko trotted out 1.d4 in game 5, and has admitted that he spent no time at all preparing for that move. I'm sure this time will be different, but whether he spent enough time on it remains to be seen.

The game continued 1...Nf6 2.c4 e6 3.Nc3! Bb4 4.f3 - a very aggressive line that has the further advantage of being somewhat out of fashion. This is a great choice for Anand, and if Kramnik survives today, it will be a tribute to his preparation and/or skill. More moves: 4...d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5 Nxd5 8.dxc5

All well-known so far. White's pawn structure may look hideous, but he has an extra pawn and controls lots of dark squares thanks to the queenside pawns. And the nice thing for us as fans is that whatever happens here, a dull draw is the least likely possibility. The game continued 8...f5 (a main line, but less commonly seen than 8...Qa5) 9.Qc2 Nd7 (9...O-O is usual, as I'm sure Kramnik knows. He is probably making a calculated decision to look for good, playable moves that aren't the main lines, in the hopes of achieving a playable if slightly worse position he can patiently defend, rather than go into some main line Anand has prepared to the teeth) 10.e4 (after a fair amount of thought) 10...fxe4 11.fxe4 N5f6 and this is as good a place as any to stop and post. Here's the current position:

The game was covered live(ish) in this post, but here's the game with an expanded set of notes.

It's not as if I want to send people away, but it is the world championship, after all, and for an event of this magnitude, more is generally better. So here are some other places you can find coverage:

The official site (One interesting article I commend to you is Artur Jussupow's match preview. [Unfortunately, a direct link is impossible - for some reason, all sub-links, no matter the topic or language, are hidden.])

The Week in Chess (with a new format) (Games briefly annotated by IM Malcolm Pein.)

ChessPro (Game 1 is annotated here, in Russian.)

Chessdom (Live annotations by GM Dmitrov.)

Crestbook (GM Sergei Shipov's site; game 1 is here.)

Europe Echecs (French language; game 1 video coverage here.)

ChessVibes (Game 1 pictures, comments by IM Merijn van Delft and at some point tonight a video, here.)

(This post, covering game 1 of the Anand-Kramnik match, will be updated as the game goes along. Kramnik has White in this game.)

12:02 Update:

The game has been drawn. The final moves were 30.Rc7 Rxc7 31.Bxc7 Bc2 32.Bxa5 Bxb3 1/2-1/2. Here's the final position:



12:00 Update:

We're speeding along to a draw here. Instead of 24.a3 or some other way of dealing with the a-pawn, White played 24.e5, taking care of one weak pawn while weakening Black's kingside structure. After 24...fxe5 25.Bxe5 (I expected 25.Rxe5, but after 25...Rxa2 26.Rxe6 Bc2 Black has nothing to complain about) 25...Rxa2 material equality had been safely reestablished. We've seen a few more moves since then (26.Ra1 Rxa1 27.Rxa1 Rd5 28.Rc1 Rd7 29.Rc5 Ra7), but the draw will probably be offered and accepted any second now.

11:30 Update:

From the diagram position below, Kramnik sprang into action with 21.e4. This is a natural move, but as it leaves White with hanging pawns after 21...dxe4 22.fxe4 Rd8 23.Rad1 Rc2, this needed to be calculated accurately. Those moves all occurred rapidly, and now Kramnik has been thinking for about 20 minutes, suggesting he might have overlooked something.

An idea I have, though I don't know if it's any good (for one thing, I'm not going to use a computer during the games), starts with 24.a3. If Black chases the queenside pawns with 24...Rc3, then 25.Bd2 Rxb3 26.Bxa5 followed by Bb4 and Bc5 leaves White in good stead. If Black tries 24...e5 instead, then after 25.dxe5 Rxd1 26.Rxd1 Bxe4 (26...fxe5 is better, but White still seems to have chances after 27.Bxe5 Bxe4 28.g3) 27.e6 is a fly in the ointment. Black loses after both 27...Rxg2+ 28.Kf1 g5 29.Rd8+ Kg7 30.e7 Bc6 31.e8Q Bxe8 32.Kxg2 Bc6+ 33.Kg3 gxf4+ 34.Kxf4 and 27...Kf8 28.Rd7 Rxg2+ 29.Kf1 Rb2 30.Bd6+ Ke8 31.Rxg7 with mate coming on g8.

Going back, though, Black might be ok after 24...Rb2. One short line, and then it's time to post: 25.Bc7 (25.Bd2 Rxd4 shows why 24...Rb2 is better than 24...Rc3) Rc8 26.Bxa5 Rcc2 27.Kh1 Rxg2 28.Bc7 e5 with at least a draw for Black.

11:00 Update:

As is common in professional chess, and should be but isn't common for most of us in the amateur ranks, the players have slown down considerably now that they're out of their book/prep. Since we left off at 10:30, three half moves have occurred: 19...Bg6 20.b3 f6, which brings us to the following position:


Kramnik-Anand, position after Black's 20th move

10:30 Update:

Kramnik played 17.O-O and after a moderate think, Anand produced the first new move of the game, 17...a5. This is a good, typical move, aimed at restricting White's queenside. After a return think of moderate dimension, the game continued 18.f3 Bf5 19.Rfe1, preparing e4. I like this on several levels, not least of which is that White is looking for other avenues of activity. With e4, he'll either get an open e-file or a potential passed d-pawn (after ...dxe4 fxe4 and an eventual d5).

It may be a technical game, but at least it's not looking like a trivial draw just yet.

9:50 Update:

Maybe we have our first answer as to what Kramnik is up to. On move 14 he played the slightly unusual 14.Qb4, and after 14...Rxc6 took on b7 (15.Qxb7). This has only been played once before in this exact position, but didn't lead to anything: 15...Qc8 16.Qxc8 Rfxc8 17.O-O Be2 18.Rfe1 Bd3 19.b3 with a draw in Molde (2273)-Marti Pericot (2342), Email 2005. Not a high-level game, but it is representative.

9:55 The game has continued in just that way: 15...Qc8 16.Qxc8 Rfxc8. We'll see if Kramnik has worked out a way to keep the pawn and maintain some chances, or if Anand's activity will net him an easy draw.

9:30 Update:

A few more moves have happened: 10.Bxc6 Bxc3+ 11.Qxc3 Rc8 12.Ne5 Ng4 13.Nxg4 Bxg4 (All book.) Both players are slowing down a bit, though it might also be website delays. (I notice that Playchess.com has several more moves than ICC.) It's sad that the official site may not have properly prepared for the traffic they're getting, though on the other hand it's nice that they're getting so much traffic!

Initial reactions:

8:55: Kramnik has White for game 1. I know this is reputed to be an advantage, but I think there's something positive about starting with Black. You get to take your opponent's best shot right up front, and once you've done it you're playing with a shot of confidence and house money for game 2. Further, at least in the old days, players would often waste game 1 in such matches just getting their head into the game. Why not let the opponent burn his White while you get acclimated?

9:05: The game hasn't started yet - I wonder if they're going to do the 15-minute delay thing they do in Dortmund nowadays, or if this is just because all of the speechifying and such typically goes on at the start of an event.

9:10: They're off! The game has started 1.d4 d5 2.c4 c6 3.Nc3, and I'm already scratching my head a little. Just about everyone automatically plays 3.Nf3 instead, so I'm trying to recall in what lines this makes a difference. But not for long: the players blitz out 3...Nf6 4.cxd5 (Massive booing!) 4...cxd5 5.Bf4 Nc6 6.e3 Bf5 7.Nf3 e6 8.Qb3 Bb4 9.Bb5 O-O, and then it slows down.

This is really surprising. Kramnik's a great technical player and all, but the Exchange Slav is widely believed to give White absolutely nothing. I'm sure that Kramnik can play this as well if not better than anyone else in the world, but it's surprising to see him use what at least seems to be a fundamentally unprincipled approach with White.

I've solicited readers' guesses (and will accept them up until the first game starts (9 a.m. ET this (Tuesday) morning; 3 p.m. local time in Bonn, Germany); now it's my turn.

Let's consider five possibilities:

1a. Anand wins by 3 or more games.
1b. Kramnik wins by 3 or more games.
2a. Anand wins by 1 or 2 games.
2b. Kramnik wins by 1 or 2 games.
3. The match ends in a 6-6 tie and goes to tiebreak games.

Let's also consider the players' strengths and weaknesses.

A. Peak strength
B. Average strength
C. 2007 results in general
D. Most recent results
E. Head-to-head, classical chess
F. Head-to-head, rapid chess
G. Physical condition
H. Psychological strength

On (A), I think Kramnik comes out slightly ahead. His peak in 2000 and 2001 seems unmatched by Anand, and I don't think there was ever a time when Anand could have done to Kasparov what Kramnik did. On the other hand, that was a long time ago, and one might argue with some propriety that Kramnik played especially well against Kasparov while Anand did especially poorly. Still, I'm inclined to go with Kramnik on this one.

On (B), TrueChess.com seems to give the edge to Kramnik over most extended time periods; on the other hand, ChessMetrics gives the edge to Anand, and over a period that includes Kramnik's best years. So here too it's awfully close, but given Kramnik's up and down results since 2004 I'll give the nod to Anand.

(C) is clearly in Anand's favor. They started the year with identical ratings, and Kramnik has fallen further. Additionally, Anand has enjoyed success in rapid events not matched by Kramnik.

(D) on the other hand slightly favors Kramnik. He had a fair +1 score in his most recent event, the Tal Memorial, but that's a lot better than Anand's lousy and winless -2 score at the Grand Slam Final. Plus for Kramnik.

(E) favors Kramnik. In 51 classical chess games, Kramnik leads 6-4 with 41 draws.

(F) on the other hand, favors Anand decisively. In 46 rapid games, 34 games have been drawn, but Anand leads by a lopsided 10 wins to 2.

(G) is tough to assess. Kramnik is six years younger, but neither is really young and Kramnik has had serious health problems. Both have taken special care for their physical fitness leading up to the match, but neither seems exactly ascetic. Kramnik claimed in a recent interview that he was feeling very good, even better than he did when facing Kasparov in 2000. If that's true, then I'd give a serious edge to Kramnik. If not, then it's a pick 'em.

(H) is probably Kramnik's biggest asset. Kramnik has demonstrated his psychological strength against adversity on the board and off, while Anand's track record in this realm has been spotty at best. (I wondered if he might have overcome this, but his performance in the Grand Slam Final and the expression on his face when recalling past failures in his new ChessBase disks suggested otherwise. Watching him talk about bad events like Dortmund 2001 (-4 in 10 games) was to see the personification of a black cloud. Chances are there will come a moment in this match when Anand will have to bounce back, and hopefully the youngsters around him will help in that respect.

OK, with all this in mind it's time to evaluate the possibilities.

(1a) seems to me highly unlikely. I doubt either player will dominate the other, and given Kramnik's mental toughness I don't think such a collapse is possible. (1b) seems very slightly more likely, given the discussion in (H), but it still seems highly improbable.

(2a) is reasonably likely, especially if the erratic Kramnik of 2004-5 and this year's Dortmund tournament shows up. Anand is a good front-runner, and if he gets out to a lead and is in good form, his ability to dry up a game out of the opening will make him extremely hard to beat. Likewise, (2b) is also very likely, and if Kramnik is in something near his best form it's the likeliest scenario - especially because Anand hasn't been very good at catching up.

As for (3), that's most likely MATCH OVER for Kramnik. Kramnik will have a chance thanks to his better nerves, but Anand is so dominant at faster time controls that Anand has de facto draw odds for the match.

I think the first player to be up after an even number of games will probably win the match, and I'm going to guess that the winner will be...

As mentioned earlier today (yesterday by the time I finish this post), Anand's four seconds for the match, starting in about nine hours, are Peter Heine Nielsen, Rustam Kasimdzhanov, Radoslaw Wojtasek and Surya Shekhar Ganguly. In this post, I'll take a look at their repertoires, with an eye to seeing how in particular they might help Anand battle Kramnik.

(1) Peter Heine Nielsen (FIDE Rating: 2662):

With White, Nielsen plays all four major moves, but prefers 1.d4 above the rest. After 1...Nf6 2.c4 e6, Nielsen plays all three main moves (3.Nf3, 3.Nc3 and 3.g3) in fairly equal proportion. Against the Semi-Slav, he goes down the main lines and plays the Anti-Moscow Gambit with 6.Bh4.

When he plays 1.e4, he plays conventional main lines. Interestingly, he doesn't seem to have had much OTB experience against the Petroff, but whatever might be lacking in his White games is made up when he has Black - see below. His experience vs. the Sveshnikov is pretty slim (one Bg5 Qa5+ Bd2 Qd8 draw); in the few relevant games, he generally chose 3.Bb5 or 3.Nc3 followed by Bb5. As for the Berlin, he hasn't faced it in 17 years.

With Black, Nielsen meets 1.e4 with both 1...c5 and 1...e5. Based on his recent track records, it's unlikely that Kramnik will play 1.e4 or meet that move with 1...c5, so let's concentrate on Nielsen's 1...e5 repertoire. He plays 2...Nc6 most often, but he's a regular exponent of the Petroff as well, choosing the main line with 6...Nc6. In the Ruy, he plays both 3...a6 (heading for both the Closed Main Line and the Marshall) and 3...Nf6, the Berlin. His record with the Berlin is pretty good, though not great, and he hasn't played it since 2005. Still, it's likely that his years of playing it will put Anand in good stead if Kramnik uses it.

Against 1.d4, Nielsen plays a bit of almost everything: King's Indian, Nimzo-Indian, QGD (possibly his main opening), Slav and Semi-Slav.

(2) Rustam Kasimdzhanov (FIDE Rating: 2672):

With White, Kasimdzhanov plays both 1.e4 and 1.d4, with a slight preference for the former. In his 1.e4 games, his record against the Berlin isn't very good - he's -1 in 8 games - and in five of the eight games he avoided the Berlin Endgame. Against the Petroff, he does go for the main lines, though without particular success - 50% against his fellow GMs. Against the Sicilian with 2...Nc6, Kasimdzhanov generally avoids the Sveshnikov and plays 3.Bb5, with good results.

With 1.d4, he is willing to play both the QID and the Nimzo-Indian. As for his record and experience against the Slav and Semi-Slav, he has done extremely well against the former (+7 =4 against strong opposition) but had mediocre results against the latter. He generally played 5.e3, only taking up 5.Bg5 in 2007, so in comparison with Anand, Kramnik and even Heine Nielsen he's a Johnny-come-lately to that opening. (But see below.)

With Black, Kasimdzhanov meets 1.e4 with 1...e5 and 1...c5 with equal frequency. He has never played the Berlin, preferring the Closed Main Lines and the Marshall. Against 1.d4, he, like Heine Nielsen, has a diverse repertoire: the KID, Nimzo/Queen's Indian, QGD and nowadays, the Semi-Slav. As with 5.Bg5 against it, this entered his repertoire just last year, but with Black his results have been excellent - three wins (including one against his fellow second Wojtaszek) and four draws (including one against Kramnik).

(3) Radoslaw Wojtaszek (FIDE Rating: 2599)

Wojtaszek is a "lefty" in chess, with a strong preference for 1.d4 (and 1.Nf3 comes in second). At the Nimzo/Queen's junction, he primarily plays 3.Nf3. Against the Semi-Slav, he has only played 5.Bg5, so far without much success (+0 -3 =3 against players > 2450).

With Black against 1.e4, he's a Sicilianista, with an occasional Caro-Kann and still more occasional French. 1...e5 almost doesn't exist for him - only 8 games in Mega, only one of which occurred after 2001. (On the other hand, it was a win with Black over Volokitin in the Gajewski Gambit of the Ruy Lopez.) Against 1.d4, he always plays 1...Nf6, and after 2.c4 e6 3.Nf3 goes for both the Modern Benoni and the Queen's Indian. After 3.Nc3 he generally goes for the Nimzo-Indian, but even here he sometimes braves the Modern Benoni.

(4) Surya Shekhar Ganguly (FIDE Rating: 2603)

Ganguly, like Rauzer, Fischer and Weaver Adams, believes that 1.e4 is "best by test". He used to play the Scotch from time to time, but now seems a Ruy specialist. He has a terrific record against the Berlin, mostly attained against sub-2500 players and using the sideline 4.O-O Nxe4 5.d4 Nd6 6.Ba4. He has faced the Sveshnikov Sicilian many times, and here too his score is very good but based largely on beating up on (relative) weakies. There he used to prefer 9.Bxf6, but since 2004 his main move has been 9.Nd5.

With Black, Ganguly has a very strong preference for 1...e5 against 1.e4, and his main line is the Moeller against the Ruy (3...a6 4.Ba4 Nf6 5.O-O b5 6.Bb3 Bc5). Against 1.d4 he plays both 1...Nf6 and 1...d5. When using the former, he goes for the Gruenfeld and Benko Gambit; when the latter, the QGA (which Anand used to play - with very poor results - against Kramnik in the early 2000s).

Superficially, then, Heine Nielsen and Kasimdzhanov make obvious sense not only by strength but their repertoires as well. We'll see as the match progresses if Wojtaszek and Ganguly fit in the same obvious way, but even if they don't I'm sure Anand will put them to good use. They are young and will have a lot of energy for the misery Anand is bound to put them through in the match. And Ganguly, as a fellow Indian with Anand, may help him feel more at home (though for Anand, it's plain that in the main he experiences the rain in Spain).

I'll update this post some time later today, but for now I'll just list the seconds:

Peter Heine Nielsen
Rustam Kasimdzhanov
Radoslav Wojtaszek
Surya Shekhar Ganguly

Nielsen is no surprise, as they've been working together for quite some time now. The rest were unexpected to me, and I'm sure we've all noticed one name missing - that of Magnus Carlsen. Kasimdzhanov is very strong (near 2700 and has previously surpassed that mark), as befits a former FIDE champion, but the other two are strong but not elite GMs. Wojtaszek is a young (21) Polish GM with a FIDE rating of 2599, while Ganguly is Indian, 25 years old and rated 2603.

Does anyone know much about Wojtaszek and Ganguly? I'll do spelunking into their opening preferences later on today, but any other info readers in the know can supply (other than that Ganguly and Anand are both Indian) to illuminate their connection with Anand could be interesting.

UPDATE: I elaborate on their repertoires as they seem likely to relate to those of the principals here.

Here it is:

Game 1: Tuesday, October 14
Game 2: Wednesday, October 15
Game 3: Friday, October 17
Game 4: Saturday, October 18
Game 5: Monday, October 20
Game 6: Tuesday, October 21
Game 7: Thursday, October 23
Game 8: Friday, October 24
Game 9: Sunday, October 26
Game 10: Monday, October 27
Game 11: Wednesday, October 29
Game 12: Friday, October 31
Tiebreak: Sunday, November 2

All rounds, including the tiebreak, start at 3 p.m. local time (= 9 a.m. ET).

Only 30 hours to go!

In just three more days, World Champion Viswanathan Anand and ex-champ Vladimir Kramnik start their 12 game championship match in Bonn, Germany. Here's your chance for supreme bragging rights: pick the winner and the score!

You can read the remaining details for yourself, but here's the most interesting bit: Vladimir Kramnik's seconds are [drum roll]:

It's a very interesting one, too, and it includes an interesting psychological/physiological insight we can all use in our own games. Topics include Kramnik, preparation, computers, emotions, Kasparov and Fischer, and the question of whether Carlsen is one of his seconds. And now, so you can get to it already, the link is here.