Dortmund Rd 7: Kramnik misses his chance

If Vladimir Kramnik doesn’t emerge as the winner of the Dortmund supertournament for an 11th time he’ll know where to lay the blame. In Round 7 all that remained was to put the ball in the back of the net after a classic Catalan squeeze against Peter Leko, but momentary tactical blindness condemned him to a draw in no less than 134 moves. Elsewhere, apart from a brief flurry of activity in Bartel – Fridman, the games looked drawish from start to finish. So the standings are unchanged and with two rounds to go Kramnik, Karjakin and Naiditsch are the men to beat. See all the games with Houdini analysis here at WhyChess.

We noted in our report on the previous round that the leaders were all set to play those half a point behind in Round 7. Although that made it potentially crucial for the outcome of the tournament, the corollary was that the players were closely matched on all boards, and so it proved.

Ponomariov showed no ill effects from his loss the day before | all photos: Georgios Souleidis, more here

Meier – Gustafsson was a four rooks ending by move 22, only left theory on move 26, and could have been agreed a draw long before it was on move 53. Naiditsch – Ponomariov saw queens leave the board on move 17 and a draw agreed in an opposite-coloured bishop ending on move 42.

Fabiano Caruana was apparently ready for a sharp tactical encounter in the Sicilian, but Sergey Karjakin wasn’t interested:

Instead of 9.Qxg7 he chose 9.Qg3, and despite dropping a pawn at one stage he easily drew another opposite-coloured bishop ending. He commented on Twitter (here I’ve translated his Russian – Sergey also tweets in English, but the grammar is a bit erratic and he finds it hard to include as many details in 140 characters!):

Today I decided not to get involved in the sharp opening systems after 9.Qg7 and simply to play chess. I was better, then worse, and the result was a draw.        

Before we get to the chess feast that was Kramnik – Leko, Bartel – Fridman makes a good appetizer, though purists may choke on their food.

It seemed on paper to be a good chance for Mateusz Bartel to build on his win against Ruslan Ponomariov and score a respectable result in Dortmund, but again some uninspiring play in the opening held him back. At least this time he stayed solid, and the game seemed to be stuck in a lifeless position where only Fridman could possibly have claims on a full point. Then the rock-solid German slipped with 30…Qb4?!

Here with 31.d4! Bartel seized his chance, and after 31…cxd4?! he was suddenly able to break through to the black king: 32.Qc8+ Nf8 33.Qe8 Ra7 34.Rc8 Kg7. However, 35.Nxb5?! turned the tables (35.Nc2! seems to lead to an ending where White has an extra pawn, if not great winning chances). And then Bartel almost threw it all away with 38…Qc6??

Black’s king is surprisingly safe, while the same can’t be said about White’s. After …f4 and …Ng6 the fat lady would soon sing, but 38…Qd1+! first was essential. Instead Fridman carelessly picked up a pawn with 38…Qxd3+??, allowing Bartel to bring the attack to a grinding halt by exchanging queens on c3. All in all, however, it was perhaps a fair outcome to the game.

The highlight of the round by a country mile was an echo of World Championship matches past, Kramnik - Leko. After single-handedly making the Catalan one of the most popular openings of the last decade Kramnik has played it relatively rarely himself of late, but against Leko it worked to perfection. By around move 16 it was already simply a question of whether his opponent could hold on, and when the Hungarian went astray in the run-up to the time control it seemed the answer was no.

Leko has just played 36…Bf6?, and his idea must have been to give up the pawn, exchange a pair of rooks and the minor pieces and hold the rook ending a pawn down. There was just one slight flaw… After 37.Rbxb6! Black can’t reply 37…Rxb6 due to 38.Ra8+! and it’s a back rank mate. Not the toughest tactic to spot, though perhaps Kramnik missed that the bishop leaving the back rank made it possible on this move, but not the last. Kramnik still had plenty of time at this point, but he trusted Leko and went about his strategic plans with 37.Nd3?!.

However, as chess fans we can relish the fact that the game didn’t end abruptly. Instead we got to see Kramnik weave his magic, as he painstakingly combined different threats and sought out a structure that would allow him to break through. As so often, it was difficult to pinpoint exactly where his opponent went wrong, though things went downhill fast after 66…Ra5:

Kramnik exchanged rooks and was able to win the a-pawn before attacking the e6-pawn with his rook and knight. He conducted the following stage to perfection and it seemed as though he knew exactly how he was going to finish off the game:

The black “army” is utterly incapable of protecting the black king, and here Kramnik had a crystal clear win: 80.Rd7! There’s no sensible defence against the simple threat of Kb6 and Rd8 mate (the reason the rook had to be on d7 is that if you play 80.Kb6 immediately the king escapes via c8 and d7). Kramnik must also have been trying to work out if the rook ending after 80…Nxf4 is winning – yes, seems to be the answer! – but in the end he couldn’t find a clear path and played 80.Kb6?! Watching live it was hard to know if this was just playing for a repetition before going in for the kill, but no – after the excruciating 80…Kc8 81.Kc6 Kb8 82.Rb7+ Ka8 83.Rb5? (83.Rb3! still seems to win) the moment had gone.

From that point on both players were almost playing on increments and although Kramnik eventually fought on in a rook + knight vs. rook ending until move 134 it always seemed as though Leko would hold the draw.

What to make of it all? Mig Greengard had the most amusing comment on Twitter:

Kramnik missing a win in a technical endgame is like finding out Santa isn’t real.

In general he thought Kramnik's endgame technique has been slipping of late, although while watching the game it was hard not to recall Kramnik’s interview at WhyChess last year (it also occurred to Dennis Monokroussos):  

Vlad Tkachiev: I had the impression that you’ve deteriorated a little in that regard in recent years. I can recall a few won positions that you couldn’t…

Vladimir Kramnik: No, I’ve always played won endgames poorly and couldn’t even tell you why myself. Perhaps I relax too soon. It’s when the evaluation isn’t yet clear, += or =+, that I play well and turn those endings into won ones, which I then sometimes make a mess of, just as I did in my younger years.

In any case, the relative standings in Dortmund are unaltered with only two rounds to go:

1-3. Vladimir Kramnik, Sergey Karjakin, Arkadij Naidtisch: 4.5/7
4-6. Fabiano Caruana, Peter Leko, Ruslan Ponomariov: 4 
7. Georg Meier: 3.5
8. Daniel Fridman: 2.5
9. Mateusz Bartel: 2
10. Jan Gustafsson: 1.5

In Saturday's Round 8 the leaders have the black pieces. Kramnik will want to avoid another ignominious defeat like the one he suffered at the hands of Caruana at the recent Tal Memorial:

  • Gustafsson - Ponomariov
  • Fridman - Naiditsch
  • Leko - Bartel
  • Caruana - Kramnik
  • Meier - Karjakin

As always you can follow the games live with Houdini analysis here at WhyChess.


Kramnik's win

"after the excruciating 80…Kc8 81.Kc6 Kb8 82.Rb7+ Ka8 83.Rb5? (83.Rb3! still seems to win) the moment had gone"

No, not quite. The win was still there after Rb5 according to the Nalimov tablebase.
83.Rb5 Rg4
84.Nf4 Rf4
85.d5 and Black loses in 29

Thanks for the correction!

Thanks for the correction! Yep, it seems there were more chances - I think 94.d6! was also given as a tablebase win?