Some really fun rapid games this round, paired with some typical blindfold blunders. The heavyweight Kramnik-Anand battle fizzled a bit at the start when Vishy, apparently under the assumption that the white rook was on f2 and not f1, hung half his pieces before realizing his mistake and resigning. In the rapid game Anand sacrificed a rook for a blistering attack but Kramnik defended well and even had the better half of the endgame draw. (31.g6!! Bxg2 32.Bxf7+ was winning for White, sez Fritz. So maybe Black had to give up the queen with 29..d5 and pray for rain. Or 28..Bc8! 29.fxg6 (29.Rg2 Kh8) 29..Bxh3 30.gxf7+ Kh8 31.fxe8Q Rxe8) Kramnik, who has always shown a special talent for blindfold, is now 4/4 in that section and leads the overall with 6.5/8, a half point ahead of Aronian. Morozevich has also had tremendous blindfold success over the past few years but he's off his game so far.
Ivanchuk-Radjabov was the real thriller of round 4. It looked like Radjabov, in another King's Indian, had found a clever clearance sac that would at least force a draw with 36..Bd7. If 37.Nxd7 Rh8 forces a perpetual check. But Ivanchuk found a great counter with 37.Rxf6! Rh8 38.Rxf7+! Kxf7 39.Bc4+ Ke8 40.Qe2 and the bishops are much better than the rook. 37..Kxf6 38.Nxd7+ Ke7 would have been a better choice for Black, although White is still doing okay after 39.Nxe5 Ngxe5 40.Qxd4. Svidler beat Vallejo with a speculative queen sac. He also won their blindfold game when Vallejo hung his queen. I'm not sure where he thought the black king was based on the previous few moves. [Ah, the official site now reports the thought the previous king move had been a rook move.] Aronian beat Carlsen in the blindfold in a theoretically drawn endgame. 63..Nc8+ 64.Kc7 Ne7! 65.Bxh6 Ke6 is a miracle draw.
From the official site:
Napkingate!
Anand missed a win with 31.g6!, check it out, some nice lines. Apparently Kramnik could not have done anything about it anyway once he allowed the attack, so it doesn't mean he defended less than perfectly after he did.
Yah, nice line. Fritz also points out that Black can give the rook back with 28..Bc8!!? and that looks about even. Still, hard for a human to consider walking into gxf7 or seeing that white is winning after g6, allowing the rook on g2 to just be captured while hitting the white queen at the same time. Yow.
Regarding, Anand-Kramnik. 29...d5 seems to lose, as well as other attempts to sac back the queen for a rook, such as 29...Kh8. An approximate line after 29...d5 is 30. gxh6 Qxg2 31. Kxg2 Kh7 (Kh8 Bg5 Be7 Bxe7 Qh4 Rg8 Kf2 Rd7 Qf6+ Kh7 e5) 32. Qh5 Rd7 33. f6 Red8 34. e5 +- In a nutshell, with uncoordinated pieces and no counterplay, black is doomed in this type of position. On the other hand, 28...Bc8! seems to hold.
Still fantastic play by both Kramnik and Anand. It seems like he used about a minute to decide on the rook sac, and it turned out to be sound! And Kramnik's play after the sac is all Fritz first line, except 29... Rxe4 (which at that point was probably the best chance). That in a rapid game and under ferocious attack. Simply unreal!
Yeah, 28..Bc8!?! might work, at least not to lose. I had assumed that 28..Qg7 was forced. Great stuff.
I wish Kramnik and Anand would play like the rapid game in a real classical game when rating points are at stake.
Kramnik seems to be some combination of Korchnoi and Smyslov. I for one don't understand why the man gets such a bad rap.
Kramnik and Anand HAVE played some killer games against each other. But I hope Kramnik and Anand have a dozen or so opportunities (in a WCC match) to make dirtbag's wish come true.
Can a GM go to the bathroom anymore without Geurt Gijssen around?
Anybody who didn't get excited watching Kramnik defend against Anand live today doesn't have a chess pulse. I of course had to work. Bah.
Killer Kramnik Anand game.
http://www.newinchess.com/NICBase/Default.aspx?PageID=400
Sorry my link didn't show up. It was a 23 move draw from some Linares game.
Does 28 ..Bc8 hold? I am not sure I think white comes out on top after: 29 Rg2 Kh8 30 gxh6 Bxf5 31 exf5 eg 31 ..Qh7 32 Bg5 Be7 33 Bxe7 Rxe7
34 Qh4 Qxf5 35 Qxe7 Rg8 36 Qxf7 is winning for white or 31 ..Qf6 32 Bg5 Re1+ 33 Kf2 Qe5 34 Qg4 Be7 35 d4 Qe4 36 Qx4 Rxe4 37 f6 again winning
On the other hand I think 29 ..d5!? does give black chances to draw 30 gxh6 Qxg2+ 31 Qxg2 Kh7 as black can prevent e5 and get activity down the g file for his rooks. Its hard for white to pick up the knight without allowing the rooks out one sample line 32 Qh3 Bd6 33 Qh5 rg8+ 34 Kf1 Rdf8 35 d4 f6 ( 35 Bg5 f6) and black is ok.
So the sac by Anand was sound but the logical result was a surely a draw as blacks pieces are all activated and well placed for defence. It looks like the logical 23 ...d5 (instead of Nd4) leads to the same result but is spectacular : 24 e5 Bc5+ 25 d4 Nxd4! 26 Be3! (exf6 loses)Nxf5! (Qb6 leaves whites attack to strong after 27 b4 Nxf5 28 Bxc5) and black gets enough for the Queen eg 27 exf6 Nxe3 28 fxg7 Nxc2+ and the game ends in perpetual after black follows up Nxa1 with Re2+ Great stuff. A fantastic rapid game by both players worthy of classical time controls
I must correct the analysis in the Anand Kramnik rapid game of my previous post in the line 23 ..d5 24 e5 Bc5+ 25 d4 Nd4 26 Be3 Nxf5 simply 27 Bxc5 seems to give white good chances. Perhaps 24 ..Qg6 followed by Qh7 is ok for black in this line
Andy,
The way to hold after 28...Bc8 is 29. Rg2 Kh8 30. gxh6 Bxf5 31. exf5 Qf6 32. Bg5 Re1+! 33. Kf2 Qe5. Now taking a rook, in fact, loses: 34. Bxd8? Rd1 35. Qh4 Qe1+ 36. Kf3 Rxd3+ 37. Kg4 Qe4+ 38. Kh5 Qxg2 39. Qf6 Bg7 -+.
Still white can continue atacking and force a sharp endgame 34. Qg4! Be7 35. d4 Qe4 36. Qxe4 Rxe4 37. f6 Bf8 38. Bxf7. This endgame is better for white, but black seems to hold.
I think this game will silence the 'Kramnik is boring' critics, huge contribution of Anand too, of course
Is the "scourge" of conservative play due in part to our recent obsession with rating points?
Back in "the day" it was all about the world championship. Fischer could boast that he had a higher ELO rating than Spassky but everyone (including Fischer himself) knew that beating up on zonal opponents, interzonal opponents, Taimanov, Larsen, and Petrosian didn't matter unless he beat Spassky in a match.
In the old days, as long as you could get yourself invited to your local zonal tournament your rating meant nothing. So in any non-WCC-related game you could trot out your wild ideas without hurting your professional standing much if you lost. Didn't chess in those days have more of a Melody Amber spirit about it?
These days the old qualifying system is gone and WCC event invitations are often issued to the highest rated. Not to mention legions of posters obsessing over the insignificant three or four rating points separating #1 from #2. Is it any wonder that nowadays many of our top GMs play conservatively, hoarding their precious rating points?
Ahh, the eternal "in the days of old even lemons were sweeter". I don't think rating points are an issue simply because rating points or not, everybody wants the best possible result. It's the same story now and 50 years ago. Do you think Leko would play more wildly if there are no rating points at stake? No chance. Just look at his Amber games.
IMO everybody in Amber plays their usual game, with an exception of Kramnik. I believe Kramnik plays differently in preparation to Mexico. I wouldn't be surprised if he continues with this more agressive style in Dortmund. So far Amber is 10x more exciting than Linares mainly due to the fact that it's rapid, hence players are taking more liberties (like Aronian with Vallejo, for example).
The only reason Kramnik is playing chess normally is because no rating points are at stake.
This doesnt 'silence' critics. As soon as he plays in the next rated tourney he will revert back to his ultra-safe approach which is good for him.. but extremely boring for chessfans.
"I think this game will silence the 'Kramnik is boring' critics."
Do you now? What a pity the betting exchanges are not offering this particular bet. I would be pleased to accommodate allcomers at more or less any odds they cared to name that these critics will not be silenced.
It does rather surprise me that they should forget where the pieces are quite so frequently in the blindfold - Anand's ...Rf2, for example. I find it hard to imagine the GMs I've known doing it, and of course they're not Anand. I once did a very little blindfold simul myself, and while I'm sure I didn't play very well forgetting where the pieces were didn't seem to be much of an issue.
I think the way of making moves can't help - having to keep your eyes open and move the mouse and so on must be distracting. It must be easier to be literally blindfold and speak the moves. And of course playing against the clock adds a pressure.
I find it is easier to play blindfold when I can see the board. That way you can immediately see what squares a piece is attacking, without having to think about it a little.
"In the old days, as long as you could get yourself invited to your local zonal tournament your rating meant nothing. So in any non-WCC-related game you could trot out your wild ideas without hurting your professional standing much if you lost."
If that was the case, then the WCC games, interzonals, etc. of those days would be considerably more conservative than non-cycle tournaments. I haven't seen any mentions of that.
Undoubtedly, increased conservatism (or I would prefer calling it increased passivity, since conservative gameplay can be aggressive) of gameplay is partially due to GMs not wanting to lose standing. But I doubt a full elaborate cycle would matter much. Unless you are a world champion, you come out of the cycle a loser and except for the final two candidates, few will remember who made it to what stage--in that way candidates cycle is like a regular tournament, try to recall off the top of your heard which round-robins Adams did well in over the past three years. You don't remember, because he didn't win any of them.
The chances of a mid-level GM playing more aggressively in this tournament are low, because he will not beat an equivalent of Larsen or Petrosian in a candidates cycle and so all he can hope for is to lose as little as possible in a tournament like Linares.
The major reason for this change is different. As chess grows older, experimental wild-style combinations become easier to disprove, offensive approach easier to neutralize, attack easier to refute. We can find many a blunder in old day games away from the board and today's GMs would find them while in the game. It's the same idea as to why mixed martial arts today doesn't have as many knockouts, why boxing is much more conservative, why soccer no longer features five forwards.