Super GM Tournament in Sofia Six of the world's top players clash in the M-Tel Masters May 11 to 22, 2005 in the Grand Hotel Sofia, Bulgaria

The Mtel Masters Super Tournament was a category 20 event with an Elo average of 2744. The time controls were classic (up to seven hours per game) and the tournament was a double round robin (every player plays every other player twice). There was a special rule in place at this Super Tournament: draw offers are not allowed, i.e. draws by mutual agreement between the players are forbidden, only technical draws may be given by the arbiter.

Round ten report

The situation before this critical final round was that Topalov was half a point ahead of Anand, but had the tougher game: black against Kramnik. Anand had white against Judit Polgar, against whom he has a huge plus score in past games. So it had the makings of a very exciting day, with a possibility that Anand would catch Topalov and take or share first with him.

However, it turns out that neither Topalov nor Anand would play the decisive role in this round – it was Vladimir Kramnik who decided the outcome of the event.

Kramnik,Vladimir - Topalov,Veselin [B80]
Mtel-Masters, 22.05.2005

1.e4 c5. Since his return to 1.e4 recent years Kramnik has had more than a few disasters against sharp Sicilians. These include a 20-move loss to Topalov in a Najdorf at Corus this year. 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6.

Kramnik vs Topalov at move five. In the background Ponomariov vs Adams.

6.Be3 e6 7.f3 b5 8.Qd2 b4.

Topalov's early 8...b4 comes as a surprise to Kramnik, since it hasn't been played in top-level competition.

What is this, he plays 8...b4 against me? Is this some kind of a trick?

I've got him thinking, I've got him worried. This is going to be a sharp game...

Okay, it's clear I retract the knight to e2. Let us see what he has in store for me...

9.Nce2 e5 10.Nb3 Nc6 11.c4 Be7 12.Ng3 g6 13.Bd3 Nd7 14.Rd1 0-0 15.Qf2 a5 16.0-0 a4 17.Nc1 Nc5 18.Bb1 Qc7 19.Nce2 Be6 20.Bh6 Rfe8 21.Nf5.

21...Bxc4. The knight "sacrifice" of course cannot be accepted: 21...gxf5?? 22.Qg3+ Bg5 23.Qxg5+ Kh8 24.Qg7#.

22.Ne3 Ba6 23.f4 exf4 24.Bxf4 Ne5 25.Nd5 Qa7 26.Bxe5 dxe5.

27.Kh1. Kramnik does not take the poisoned pawn on f7 with check, with good reason: after 27.Qxf7+ Kh8 28.Nxe7 Rxe7 29.Qf6+ Kg8 30.Qf2 (the threat was 31...Nxe4+ and NxQ) 30...Bc4 Black is better. Kramnik has played it safe by castling kingside. He has got a visually attractive attacking position, with the open f-file and pressure. Meanwhile, Topalov has grabbed a pawn and is playing accurate defence. Now Kramnik gets his pawn back, and the position was still very sharp and dynamically equal.

27...Bg5 28.Nxb4 Bc4 29.Rfe1 Qb7. At this stage it looked like the game might end in a repetition with 30.Qxc5 Rac8 31.Qa5 Rca8 and a draw. But: 30.Nd5

This is when things started to get really weird. 30...Qxb2? Topalov leaves his knight hanging on c5! 30...Rac8 was a reasonable continuation. 31.Nc7? On the Playchess server we thought the moves might be coming in wrong. Both players were fairly low on time, but not that low, you would think. Kramnik had over five minutes, Topalov over 10. It is possible that both players assumed that if White takes with 31.Qxc5 Black gets the piece back with Bxe2. But that loses to 32.Qf2, which pins and wins the bishop (as well as the game). Both players remained blissfully unaware of this double blunder until spectators brought it up at the end of the press conference.

31...Ne6 32.Nxe8 Rxe8. Now Kramnik had an exchange for a pawn, but a somewhat shaky position. Topalov had good compensation and it looked like a long, interesting struggle was coming. 33.Rf1 Rf8 34.Nc1??

WHAT?? That simply gives away a piece. 34...Qxb1 and since White is a piece down with absolutely no compensation: 0-1.

There were long discussions on the Playchess server about how Kramnik could have played this terrible blunder. One can only assume he thought the Black could not take the bishop, because after 35.Nb3 the black queen was trapped. Somehow he didn't realize that the e4-pawn wasn't protected. It had been defended since move 18 by the bishop on b1. Rich Weaver of Philadelphia suggests that Kramnik may have imagined that the Nb3 trap worked because the same bishop he sac'ed on b1 was somehow still protecting the e4 pawn!! Sounds wierd, but what other explanation is there? Perhaps we should take his statement in yesterday's press conference seriously. Kramnik said that he had still not fully recovered from the illness that struck him half-way through his world championship match against Peter Leko in Brissago in early October last year.

Vishy Anand got a small advantage over Judit Polgar and worked on it for 45 moves. But the tough lady from Hungary did not go astray and the game ended in the standard draw offer in Sofia: repetition of moves.

Ruslan Ponomariov tried hard to win against Michael Adams, after the game had traded down at move 27 to queen, two rooks and symmetrical pawns on both sides. But Mickey is just as resilient as Judit and the game ended in a repetition at move 53.

Final standings

So in the end Topalov won the strongest tournament of the year a full point clear of the second-placed Anand. His performance rating was 2847 – Pedro Forquesato of Brazil calculated his performance in the second half of the tournament, with four wins and one draw, to be 3106 (we put it at 3122).

Anand Nair of New York sent us the following rating changes that can be expected on the FIDE list:

This means that if a rating list was published today it would have the following players in the top slots (without the retired Garry Kasparov, and without a possible small gain by Anand in the German Bundesliga):

The currently strongest active male and female players

Schedule and results

• Download all games in PGN format
• All reports so far

Links

• Official M-Tel Masters web site
• Live coverage ; also on Playchess.com
• Mobiltel