GM Mihail Marin: 'A day of curious moments'

9/16/2007 – Morozevich fell behind against Svidler, who allowed his opponent to come back and win. Aronian tried to sacrifice a pawn against Grischuk, but found himself a pawn up and under attack. Anand got nothing out of the opening against Kramnik and had to fight for a draw. Leko vs Gelfand went for 100 moves. Here are comments by GM Mihail Marin and postgame video interviews by Vijay Kumar.

ChessBase 14 Download ChessBase 14 Download

Everyone uses ChessBase, from the World Champion to the amateur next door. Start your personal success story with ChessBase 14 and enjoy your chess even more!

Along with the ChessBase 14 program you can access the Live Database of 8 million games, and receive three months of free ChesssBase Account Premium membership and all of our online apps! Have a look today!


World Championship 2007 Mexico

The World Championship 2007 will take place from September 12 to 30 in the Sheraton Centro Histórico Hotel in Mexico City. Eight players are qualified – the tournament will be a double round robin. The prize fund is US $1.3 million.

Round three commentary by GM Mihail Marin

This was a day with several curious moments. Playing with White against Svidler, Morozevich quickly fell behind in development and his position looked highly suspicious. Svidler might have been surprized by the ease with which he had obtained an advantage or just confused by the wide choice of tempting continuations; in any case, he first allowed to his opponent to equalize and then take over the initiative. Shortly before the time control, he resigned.

Anand's opening experiment against Kramnik did not yield him anything concrete. His attempt to maintain the initiative led to a rook endgame with a pawn down, which, however, he defended quite confidently. The game ended in a... stalemate!

In Grischuk-Aronian it was not entirely clear who tricked whom in the opening. Aronian intended to sacrifice a pawn, but eventually found himself with a pawn up and... defending. After 27 moves Grischuk had obtained a decisive attack, which he converted into an immediate draw by repetition.

In the longest game of the evening, Gelfand struggled for a long while with Black against Leko. The situation changed abruptly when the Hungarian GM retreated with his queen on the back rank shortly before the 40th move. Play simplified to a queen ending with winning chances for Black, but Leko defended well and squeezed a draw on the 100th move.

P.S. My special thanks to Knut Johnsen, who highlighted a mistake in my comments to Svidler-Leko, from the previous round: "The move 15.Qe2 is not new! I have five games with this variation in my database: Apicella-Boudre 0-1, 1987; Schrlank-Perlega 1/2-1/2, 1998; Rousar-Smutny 1/2-1/2, 2001; Nolan-Watson 0-1, 2002; and Nemkova-Bobras 1-0 from 2007. After move 16...Qh5 we have reached the same position as in the game Kapengut-Malaniuk from 1985, but this game was on move 16 instead of 17!"

Morozevich,A (2758) - Svidler,P (2735) [C45]
WCh Mexico City MEX (3), 15.09.2007 [Mihail Marin]

1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Bc5 5.Be3 Qf6 6.c3 Nge7 7.Bc4 0-0 8.0-0 Ne5

9.Bb3!? Morozevich remains faithful to his style. Instead of the approved 9.Be2, he plays a move that has been only rarely seen in games between strong players. 9...d6 10.f3. But this is "almost" a novelty. In most of the games, White preferred to develop his queen's knight, ignoring the threat ...Ng4. 10...Be6 11.Kh1 Bc4 12.Rf2 d5

White's opening experiment seems to have ended into a complete fiasco. He is behind in development and his advantage of space is just about to vanish. 13.Bc2 dxe4. Black had a wide choice of tempting moves. The most logical seems to be 13...Rad8 , but then White could consider playing f4, immediately or after driving the bishop away with b3.; Maybe 13...Qh4 is best, preventing f4 in view of ...Ng4 and planning to place his rooks on the open files. It would not have been easy for White to complete his development. 14.Nd2! White hurries to bring his pieces into play. The point behind his last move is that 14...exf3? loses a piece to 15.Nxc4 Nxc4 16.Qd3, atacking c4 and h7. 14...Bd3?! After this move White is back in the game. Once he cleared the d5-square with his previous move, Black should have used it with 14...Nd5 for instance 15.Nxe4 Qb6 and White is still under pressure. 15.Nxe4 Bxe4 16.fxe4 Qg6 17.Rf4

White has a normal position now; chances are about equal. Svidler must have ben still under the impression of his missed advantage, because in the next phase of the game he effectuates a series of pseudo-active moves, completely losing coordination. 17...Nc4 18.Bg1 Qh6 19.Rf3 Qd2 20.Qb1 Bb6 21.Bb3!

Suddenly, Black is in trouble. If the knight moves, Be3 wins the queen. 21...Bxd4 22.cxd4. But now, White's pair of bishops and his mobile centre offers him a huge advantage. 22...Na5 23.Bc2 Rad8 24.Rc3 Nac6 25.d5 Nb4 26.Bb3 Na6 27.Be3 Qe2 28.Bc4 Qg4 29.h3 Qh4 30.Bxa6 bxa6 31.Rxc7 f5 32.Bc5 Rfe8 33.d6 Ng6 34.exf5 Nf4 35.Qc2 Re2 36.Qb3+ Kh8 37.Rg1 1-0. [Click to replay]

Anand,V (2792) - Kramnik,V (2769) [C42]
WCh Mexico City MEX (3), 15.09.2007 [Mihail Marin]

1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.d4 d5 6.Bd3 Nc6 7.0-0 Be7 8.c4 Nb4 9.Be2 0-0 10.Nc3 Bf5 11.a3 Nxc3 12.bxc3 Nc6 13.Re1 Re8 14.cxd5 Qxd5 15.Bf4 Rac8

16.Qa4. A rare continuation, aiming to take Kramnik "out of book". 16...Bd7 17.Qc2. This seems to be a new move. 17.Rab1 was played a couple fo times. 17...Qf5 18.Qxf5 Bxf5 19.Bb5 Bd7 20.d5 Ne5 21.Bxd7 Nxd7

Anand's experiment did not work out too well. Kramnik played relatively quickly and obtained an entirely satisfactory position. White's centre could become more of a weakness if Black is allowed to install his minor pieces on c5 and d6. 22.Bxc7!? White tries to maintain his initiative, but further simplifications will turn the position slightly dangerous for him. 22...Rxc7 23.d6 Rxc3 24.dxe7 f6

With queens on board, the e7-pawn would most probably ensure White's win, but in the endgame it is more of a source of worries. 25.Rad1 Rc7 26.Nd4 Ne5 27.f4 Nc6 28.Nxc6 bxc6 29.Rd6 c5 30.Ree6 c4 31.Rc6 Rexe7 32.Rxc4 Rxc4 33.Rxe7 Ra4 34.Rb7

White will lose a pawn, but the drawish tendency of this type of ending is well-known. In the "theoretical" position, White's f-pawn stands on its initial square. The way it is, it could easily become a weakness, but also ensure White a stable advantage of space on the kingside. 34...h6. 34...h5!? looks more active. 35.f5 Rxa3 36.Kf2 h5 37.g3 a5. 37...h4!? would have offered more chances to keep White under pressure, by keeping the king's access to the f5-pawn open. 38.Ra7 a4 39.h4!

White is out of any danger now. The kingside is practically frozen and the huge advantage of space prevents the standard plan: push the pawn to a3, in order to leave the a2-square available for the transfer of His Majesty. If Black tries to do so, he would lose the g7-pawn, after which White's counterplay (for instance g4) is much more dangerous than in the standard position (pawns on f2, g3, h4 and symmetrical for Black). 39...Ra2+ 40.Kf3 a3 41.Ke3 Ra1 42.Kf2 Kf8 43.Kg2 a2. This move somewhat restricts the activity of White's pieces (Kf2? loses to ...Rh1 and things like that), but there is no place to hide for the black king either. 44.Kh2 Ke8 45.Kg2 Kd8 46.Kh2 Kc8 47.Kg2 Kb8 48.Ra3 Kb7 49.Ra4 Kb6 50.Ra8 Kc5 51.Ra7 Kd5 52.Ra4 Ke5 53.Ra5+ Ke4 54.Kh2. Actually, this move is not forced. decades ago, Kholmov has proven that even without the f5-pawn White can hold a draw. The only winning chance for Black is to create a passed pawn on the f-file, but this is impossible in case of accurate defence from White. 54...Kf3 55.Ra3+ Kf2 56.Ra4 Kf1

57.Kh1. But this is an important move. In case of a neutral move such as 57.Ra5? , Black can simplify to a won pawn ending with 57...Re1 58.Rxa2 Re2+ 59.Rxe2 Kxe2 , for instance 60.Kg2 Ke3 61.g4 Kf4! 62.gxh5 Kg4 and it is all over. 57...Ke1 58.Kg2. But now, there was no need to get back on the second rank. 58...Kd1 [58...Rd1 is very much the same as in the game. 59.Ra7 Rc1 60.Rxa2 Rc2+ 61.Rxc2 Kxc2 62.Kf3. This is the difference. The white king is active enough now. Curiously, he will have to use this acitivity for getting... stalemated!! 62...Kd3 63.g4 hxg4+ 64.Kxg4 Ke4

Did Anand miscalculate anything? 65.Kh5! Not really! 65...Kxf5 and, entirely in accordance with the Sofia rule, players had to split the point. 1/2-1/2. [Click to replay]

Grischuk,A (2726) - Aronian,L (2750) [C88]
WCh Mexico City MEX (3), 15.09.2007 [Mihail Marin]

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 0-0 8.h3 Bb7 9.d3

Nothing really new or thrilling so far... 9...d5!? But this looks quite unexpected! Aronian wants to play in the spirit of the Marshall Attack irrespective of White's cautious opening choice. At this level, the move has never been played before. 10.exd5 Nxd5 11.c3!? Grischuk played this move after a long thought. He probably did not feel ready to check Aronian's analysis. The whole variation had been successfully employed with Black earlier this year by Ivan Sokolov and Peter Heine Nielsen. The main point is that after 11.Nxe5 Black does not capture on e5 (as he does in the genuine Marshall Attack), but takes advantage of the fact that the d4-square is not under white control with 11...Nd4! The idea is not new in fact, previously it was seen in an under 16 girl competition (1996)... After the text move, it was Aronian's turn to sink into deep thought. It appears that play has just transposed to a different variation, where the moves c3 and d3 are effectuated in reversed order... 11...Bf6. In the most recent example at high level wher this position occured (with reversed move order, though), Black prefered 11...Qd6 12.Nbd2 Rad8 13.Ne4 Qd7 , with an acceptable position for Black, although he later went down abruptly, Kamsky-Bacrot, Elista (m) 2007. 12.Nbd2 Nf4. 12...g6 is more cautious. 13.Ne4 Nxd3 14.Nxf6+ gxf6 15.Re4 Nxc1 16.Rg4+ Kh8 17.Qxc1

White's compensation for the sacrificed pawn is beyond any doubt. The chronic weakness of the enemy kingside offers him excellent atacking chances. Imagine Black's frustration: he intended to sacrifice a pawn himself, in order to take over the initiative and all he got is the opposite situation! 17...Ne7 18.Bc2 The start of a massive migration of White's pieces towards the enemy kingside. 18...Rg8 [Avoiding the trap 18...Bxf3?? 19.Qh6 Ng6 20.Rh4!! with mate to follow.] 19.Qh6 Ng6 20.Nh4 Qf8 21.Qh5 Qe8 22.Re1 Rd8 23.Bf5 Bc8 24.Bxc8 Rxc8 25.Nf5 Qd7 26.Ree4 Rcd8 27.Kh2 Qd1

White has completed his regroupment, but, unexpectedely, Grischuk decides to force a draw. 28.Nh6?! Instead, he could have obtained an overwhelming advantage with 28.Qh6! , threatening Rh4, and if ...Nxh4, then Qxf6+ followed by mate on g7. 28...Qd2 The only try. 29.f4! Rd6 (Defending f6. If 29...exf4 30.Rh4 , Black is forced to sacrifice the queen with 30...Qxg2+ 31.Kxg2 Nxh4+ 32.Kf2 Nxf5 , but after 33.Qxf6+ Ng7 34.Rd4 he is too passive to count on saving the game.) 30.fxe5 Qxh6 (30...fxe5?! shortens Black's suffering because of 31.Rg5! , with the terrible threat Qxh7+ followed by Rh5+ and mate on h6.) 31.Nxh6 Rb6 32.exf6!? The most ambitious continuation, although the g8-rook was not poisoned either. 32...Rf8 33.Ref4! White calmly defends his important f6-pawn and Black is helpless in view of the threat h4-h5. The point is that 33...Nxf4 allows mate in 2 with 34.Rg8+! Rxg8 35.Nxf7#. 28...Rg7 29.Nf5 Rgg8 30.Nh6?? Rg7 31.Nf5 1/2-1/2. [Click to replay]

Leko,P (2751) - Gelfand,B (2733) [C42]
WCh Mexico City MEX (3), 15.09.2007 [Mihail Marin]

1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.Nc3 Nxc3 6.dxc3 Be7 7.Be3 Nd7 8.Qd2 Ne5 9.0-0-0 0-0

10.h3. A very rare move. We are familiar with 10.h4 from the game Anand-Gelfand, played just two days earlier. 10...Nxf3 11.gxf3 Bf5 12.h4 Re8 13.h5 Bf6

Optically speaking, White's position does not look too appealing. He has several structural deffects, while Black has a normal development. However, one of the first (if not the very first) criteria of evaluating a position is the situation of the kings. White's pressure along the g-file could become unpleasant for the black king, meaning that Black has to overcome certain difficulties before entering a long-dreamed better ending. 14.Rg1 Qe7 15.Kb1 d5 16.Bd3 Bxd3 17.Qxd3 Rad8 18.Rg4 Kh8 19.a3 b6 20.Rdg1 h6 21.Qd2 Kh7 22.Bd4 Rg8 23.Re1 Qd6

24.Bxf6 gxf6. Forced, because 24...Qxf6 loses the f7-pawn to 25.Rf4 , for instance 25...Qg5 26.Qd3+ Kh8 27.Rf5 Qh4 28.Rxf7 with advantage to White. . After the text move, we can conclude that White has converted hislong-term pressure into something of more stable nature, by chronically weakening Black's kingside. Besides, the pressure persists. 25.Qd3+ Kh8 26.Qa6 Rxg4 27.fxg4. Ironically, it is White who has the better structure on the kingside now. However, the black king starts feeling slightly more secure. 27...Qc6 28.Re7 Qd6 29.Re3 Qc6 30.Qd3 Kg7 31.Re7 Kf8 32.Re1 Kg7 33.f4 Qd6 34.Qf3 c5 Facing the permanent threat of g5, Black decides to react in the centre. 35.Qf2 d4 36.Rd1 Qe6 37.Re1 Qd5 38.Rd1 Qe6 39.Qg1

From here, the queen sustains the planned break g5 and keeps the d4-pawn under pressure. However, placing such an important piece far from the centre hardly offers chances for an advantage. 39...Qe4. Gelfand's major pieces are familiar with this square from his game against Anand. I assume that this time he would not refrain from grabbing the f4-pawn, if only given a chance. 40.g5 fxg5 41.fxg5 d3 42.cxd3 Rxd3 43.Rxd3 Qxd3+ 44.Ka1 Qf5

Suddenly, the situation has changed very much to Black's favour. Both his remaining pieces (yes, the king IS a piece in the endgame) are more active than White's. Soon, Black will get a dangerous kingside passed pawn. 45.g6 fxg6 46.hxg6 h5 47.Qg3 Qf1+ 48.Ka2 Qc4+ 49.Ka1 Qf1+ 50.Ka2 Qc4+ 51.Ka1 Qg4 52.Qc7+ Kxg6 53.Qxa7 Qd1+ 54.Ka2 Qd5+ 55.Ka1 Qd1+ 56.Ka2 Qd6 57.Qa8 Qe6+ 58.Ka1 Qe1+ 59.Ka2 Qe6+ 60.Ka1 h4 61.Qg2+ Kh5 62.Qf3+ Kg5 63.Qg2+ Kh5 64.Qf3+ Kg5 65.Qg2+ Qg4 66.Qd2+ Kf5 67.Qd5+ Kf4 68.Qd6+ Kf3 69.Qf6+ Kg2 70.Qc6+ Qf3 71.Qg6+ Kf1 72.Qxb6

In queen endings, the number of pawns is less relevant than the degree of danger they are posing. Obviously, the h3-pawn is a very serious candidate to promotion, but the part played by White's queenside pawns should not be underestimated. Without them, Black would probably find a form to parry the checks with a counter-check, or by pining the enemy queen. 72...h3 73.Qd6 Qf2 74.Qd3+ Kg2 75.Qd5+ Kg1 76.Ka2 Qf4 77.Qxc5+

For instance, in the absence of the b2-pawn, ...Qf2+ would win now. 77...Kf1 78.Qb5+ Ke1 79.Qd5 h2 80.c4 Qh4 81.Qe5+ Kd1 82.Qd5+ Ke2 83.Qe5+ Kd1 84.Qd5+ Kc1 85.Qh1+ Kd2 86.Qd5+ Ke3 87.Qe5+ Kf3 88.Qf5+ Qf4 89.Qd3+ Kf2 90.Qd5 Kg1 91.Qg8+ Kf2 92.Qd5 Qf3 93.Qd2+ Kg1 94.Qg5+ Kf1 95.Qc1+ Kf2 96.Qd2+ Kg3 97.Qg5+ Kh3 98.Qh6+ Kg3 99.Qg5+ Qg4 100.Qd5 Qf3 and, having reached the 100th move, Gelfand decided to call it a day. 1/2-1/2. [Click to replay]

JavaScript replay

Note that on our JavaScript replay page, to which we link at the end of each game, you can follow the moves by clicking on them – even in variations and analysis. The board will follow. You can scroll the moves and commentary using the scroll bar or your mouse wheel. The four games are listed on the left and can be loaded by clicking them. There is also a PGN download, which will get you the games with full commentary, to replay and analyse in Fritz or a compatible program, including the free ChessBase Light (download link given below).

All results of the round

Round 3: Saturday, Sept. 15th 2007, 14:00h
Alexander Morozevich 
 Peter Svidler
Viswanathan Anand 
 Vladimir Kramnik
Alexander Grischuk 
 Levon Aronian
Peter Leko 
 Boris Gelfand

Current standings


Discussion and Feedback Join the public discussion or submit your feedback to the editors


Rules for reader comments


Not registered yet? Register