Greengard's ChessNinja.com

1st, 2nd, and 3rd Blood in Dortmund

| Permalink | 28 comments

well, the Chess.FM "de Firmian Curse" was quickly supplanted by the Benjamin Bump. Four draws on the first day were followed by three decisive games in the second round of Dortmund. Let it be recorded that for around 90 minutes on Sunday, Jan Gustafsson was the clear leader of Dortmund. He crushed his countryman Naiditsch in just 23 moves, only seven of which were original. They played a very dangerous Queen's Gambit line that Kramnik tried against Anand in Dortmund five years ago. Gustafsson played the novelty 16.f4, giving up the c-pawn to activate his rooks against the itinerant black king. Black just looks lost. 17..Qe3+ 18.Kh1 Qxe4 loses spectacularly to a knight-rook sacrifice. 19.Nf5+! Qxf5 20.Rxd7+! Kxd7 21.Qxf7+ with a forced mate. Giving up the queen with 19..exf5 20.Re1 just loses slowly. In the game Naiditsch cracked a move later with 18..Qe3+?? when 18..Rhd8 still requires accurate play. An instructive moment, playing a bad check and taking the queen away from control of several key squares. Gustafsson finished nicely to score the first win of the tournament. And as promised, behold the shiny new Gustafsson tag!

Kramnik ripped through van Wely as just about everyone predicted. The excellent 18.c5! forced open the long diagonal for the b2 bishop. Kramnik didn't rush things, as usual. He methodically moved in on the black king until there was no defense against massive material losses. The computer says Black can survive after 22..f5, as horrible as that looks after 23.Rad1 with total domination and continued threats. 23..f6, giving back two pawns, was probably the last desperate hope. A very smooth attacking effort by Kramnik to join Gustafsson in the lead. He was joined a bit later by Peter Leko, who beat Ivanchuk in a very messy game. Most of the mess was caused by Ivanchuk's odd knight tour of death. He moved his knight from d7 to c5 to e4 to g5, where it was duly trapped and lost. 19..Nxe5!? looks like a more successful try. Ivanchuk made a fight of it a piece down, but eventually Leko liquidated.

Nepomniachtchi-Mamedyarov was an excellent Scrabble score. It was also an entertaining battle that eventually boiled down into a draw. Nepo had a solid plus most of the way but rushed things in Mamedyarov's serious time trouble. Very nice defense by Black to reach the draw. Earlier, the cute 28.Qe4, cutting off the black knight, was strong. 28..Qxe6? 27.Nf4.

As last year, the organizers announced they would delay the broadcast of the moves by 15 minutes as an anti-cheating precaution. They did it for round one, but we got a surprise on Sunday when the broadcast started promptly at nine. Maybe they wanted to speed things up a bit so everyone could go watch the German-Spain final match of the Eurocopa? If so, it was a bad deal for the German fans, as Spain dominated and won. We'll see if they're back to 9:15 EDT in round three.

Round 3: Naiditsch-Kramnik, Mamedyarov-Gustafsson, Ivanchuk-Nepomniachtchi, van Wely-Leko.


Fetch a coffin and some nails! let's put the petroff where it belongs!! every time someone beats it it puts a smile on baby jesus face! hooray hooray!

Wow. Naiditsch just struck a blow for Anand by introducing a novelty in the Petroff that took Kramnik by surprise. That was also pretty instructive by the German to wrap up the win.

omigawd, what a game by Naiditsch! Could somebody versed in the petroff please point out the novelty?

Was it 16. Bf4! ??

Naiditsch also deserve a shiny tag!

Who would have thought? a case of the mouse that roared.

Or 19 Qd2! apparently, from banter on chessgames.com

16.Bf4 was thought to just pan out into a draw as in Kasimdzhanov-Yusupov 2001: http://www.chessgames.com/perl/chessgame?gid=1163206

(1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. d4 d5 6. Bd3 Nc6 7. O-O Be7 8. Re1 Bg4 9. c4 Nf6 10. Nc3 Bxf3 11. Qxf3 Nxd4 12. Qd1 Ne6 13. cxd5 Nxd5 14. Bb5+ c6 15. Nxd5 cxb5 16. Bf4 Nxf4 17. Rxe7+ Kf8 18. Re5 Qd6 19. Rf5 Rd8 20. Ne3 Qxd1+ 21. Rxd1 Rxd1+ 22. Nxd1 Ne6 23. Rxb5 b6 24. Nc3 Ke7 1/2-1/2)

That's how a lot of us were almost sure it would turn out at that point. Then came this very nice 19.Qd2 novelty and turned things upside down... Kramnik should probably have captured on e5, objectively, but it is very dangerous and I don't know what happens really. 19..Qxe5 20.Qb4+ Ke8 21.Qxb5+ Kd8 22.Rd1 looks horribly dangerous, and you can understand that Kramnik did not want to enter things like this against a presumably very well-prepared opponent. What he did was no better, though - White gets a very strong initiative.

I've been waiting for something like this to happen. When you play the exact same opening all the time, it is very easy to prepare against you, and especially such a tactical opening as the Petroff in today's computer age... no matter how hard you work yourself, you're just not able to cover everything in every line. This makes it pretty impressive that Kramnik hasn't lost with it for so long, but it had to be due some time. I also think it's less and less likely that he will actually use it in the match.

Focus should be on Naiditsch rather than Kramnik right now, of course... an excellent achievement, and pretty satisfying revenge for the Gustafsson debacle.

Naiditsch gets to play Kramnik relatively often as the usual favorite son in Dortmund. You'd expect he might eventually find something given that Kramnik has become rather predictable in his openings on the tournament circuit.

Give him all credit for the followup, as well, since Kramnik defended so well up until 40. Nd8 (40...Nd8 doesn't so much lose, but 41 ...Ne6 sure does, and if not to play 41...Ne6, 40...Nd8 has no purpose.)

What was Ivanchuk thinking with 6.Bc4 if he's just going to sack the e pawn and take a draw?

the commedy: Kramnik gets toasted and the Kramnik's fan acirce goes on about the impressive Kramnik.

Whatever, slomarko. I mentioned that as an aside.

btw how did you think of the "such a tactical opening as the Petroff" part? i mean lets be fair there is hardly any more sedate opening than the Petroff (save maybe the Berlin).

Because it is - at least there are so many such lines that it certainly deserves this general characterisation. Today's game was one example among many. That's why it is so well suited for computer preparation (as compared to the Berlin indeed, for example). Of course White can avoid them if he pleases, but still.

completely unrelated, but does anyone else notice that the dirt´s top blog sometimes vanishes? Like now I obviously see this blog entry, but 1 hour ago i didn't see it, and he blog from 28/6 was first. Even though I refresh the page, delete cookies and change browsers, browsing the dirt and getting the top blog is very random. Sometimes it´s there and sometimes it´s not. This happens VERY frequently, and only on this site. Am I the only one, perhaps?

I dont know about computer preparation. At least my comp does not recommend Qd2 at any depth I have set it to. The lines may have been checked with a computer ... but this was human insight, not computer.


It happens to me all the time. Mig hasn't gotten all the bugs worked out on this web page yet, despite his concerted efforts to do so. I get around the problem by looking for the "Recent Comments" tag on the left hand side of the web page and the most recent blog will be there. Click and you're in.

Well, 19 Qd2, what to say.

The thing is, it may not actually be any good. I haven't seen any serious analysis of 19...Qxe5 yet, so I don't know, and taking the rook looks dangerous, which is surely why Kramnik avoided it...but Black's up a Rook!

Isn't it at least possible that the reason 19 Qd4 is the computer's preferred move is that 19 Qd4 is better?

"I dont know about computer preparation. At least my comp does not recommend Qd2 at any depth I have set it to. The lines may have been checked with a computer ... but this was human insight, not computer."

If you check the lines with a computer it is certainly computer preparation. It is obvious that nobody is simply sitting down watching TV while he lets Fritz or Rybka does all the work. So, yes, I am obviously talking about computer-assisted preparation. It does not rule out human insight, on the contrary human insight is almost always essential.

Of course I haven't heard him talk about it, so it could be that 19.Qd2 was an OTB conception.

gmc, it's conceivable that 19.Qd2 does not _objectively_ give White an advantage.. but I don't know. The computers so far seem to say it's about equal in what seems like the most critical line, despite White being a rook down, but it seems very scary for Black with his king that exposed etc. Exhaustive analysis is certainly to follow. Anyway, the alternative to 19.Qd2 is simply to go down the road of Kasim-Yusupov and agree to draw a few moves later.

very poor defending by Kramnik I thought; after the novelty, he just fell apart. Yes there are tactical themes, but this is hardly Tal vs Keller, Zurich 1959 http://www.chessgames.com/perl/chessgame?gid=1139481. As for Petroff being a tactical opening, I guess its all relative. I mean the Sicilian is also a positional opening I guess.

Btw Acirce, many thanks for the previously played lines and the theory.

seems that Kramnik reacts badly to novelties which radicaly change the nature of the position creating many complication (Topalov's Nf7 and now this 19.Qd2). surely Anand is taking notice.

Yesterday wasn't great (though Kramnik had a horrible choice, knowing his opponent would play computer-analysed moves whatever he did), but I don't agree about Topalov's Nf7. Kramnik played perfectly there for quite a while and even when he went "astray" it was into a position that gave him decent practical chances.

Overall I'd say he's among the best players at working things out over the board even when suprised - remember Topalov's team complaining he must be cheating because he played Fritz lines against their novelties.

"Kramnik played perfectly there for quite a while", I don't know - he made five reasonable moves (including 12..Kxf7...) but then 17..Qxd4 was already perhaps the decisive mistake. Even before that there are many other reasonable alternatives, so it's very hard to say whether Kramnik played "perfectly" - can Karjakin's 16..h5 be cracked, for example?

Not to be either too naive or too cynical, but in a way this isn't really a real catastrophe for Kramnik, is it? If there's a hole in one of his favoured defences, surely better that he knows about it now rather than while playing Anand. Naiditsch perhaps did him a favour.

Who knows, maybe Naiditsch has actually "scooped" something Anand was working on. Back to the drawing board not for Kramnik, but Anand! And maybe -- but this is surely too cynical -- Kramnik deliberately sent a "wrong" message to Anand...

If that last one is true, then I hereby award him the Chesshire Cat Psychological Genius Trophy.

I don't know. There are always "holes" in anyone's favorite defence, given that opening theory isn't 100% perfect or complete. But I don't even know if Naiditsch uncovered a real "hole," or just a continuation that Kramnik happened to be unfamiliar with. Given computer analysis it's likely the latter.

Anyway, I doubt this game has much significance. If it does, then it's along the lines of what Jean said--exposing Kramnik as a player who doesn't react optimally to aggressive theoretical novelties.

Acirce is right. The Petroff is a rich and sharp opening that lends itself to home analysis. Anyone who doesn't believe this should check out the old classic book by Karpov on this opening. There was the famous game Browne-Bisguier and a more recent classic(circa 1995) with Anand against Kramnik, which were both won entirely at home. Anyways, I have never understood why it has a bad a reputation.

Twitter Updates

    Follow me on Twitter



    About this Entry

    This page contains a single entry by Mig published on June 30, 2008 6:16 PM.

    Drawish in Dortmund was the previous entry in this blog.

    Petroff Loses! Petroff Loses! is the next entry in this blog.

    Find recent content on the main index or look in the archives to find all content.