MTel Round 3 analysis by GM Mihail Marin

by ChessBase
5/14/2006 – Lowest seed Gata Kamsky shocked chess fans all over the world by beating Vishy Anand, 132 points his superior. And he did it with the black pieces. If you want to know what transpired in that and the other two games you are welcome to replay the highly instructive commentary by Mihail Marin.

ChessBase 17 - Mega package - Edition 2024 ChessBase 17 - Mega package - Edition 2024

It is the program of choice for anyone who loves the game and wants to know more about it. Start your personal success story with ChessBase and enjoy the game even more.


Round three analysis

Round 3: Saturday, May 13, 2006

Ruslan Ponomariov 
 Veselin Topalov
Etienne Bacrot 
 Peter Svidler
Vishy Anand 
 Gata Kamsky

All games so far in PGN


Commentary by GM Mihail Marin

The only decisive game of the day, Anand-Kamsky, caused a change of lead in the tournament. After having won his first two games with Black, Anand somewhat unexpectedely lost his first white game, possibly underestimating the hidden dangers of an apparently simple position. Ponomariov-Topalov was a fine fighting game, where neither of the players could turn the balance to his favour. Bacrot obtained a promissing position against Svidler’s original treatment of the Gruenfeld, but gave up his central domination too easily and had to content himself with a draw.

The following annotated games can be replayed on a special JavaScript board in a new window. Note that you can scroll the notation (without scrolling the board) and click on it to replay the game. PGN download also available.

Bacrot,E (2708) - Svidler,P (2743) [D85]
Mtel Masters Sofia BUL (3), 13.05.2006 [Mihail Marin]

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Nf3 c5 8.Rb1 0-0 9.Be2 b6 10.0-0 Bb7 11.Qd3 Qc7!? Theory is relatively well established after 11...Ba6 , when White is supposed to maintain an edge. With his last move, Black declares his intentions to enter a less explored territory. 12.d5 Nd7

13.Qc2!? This move was introduced in practice recently by Bacrot. White retreats his queen on a less exposed square, waiting for Black to declare his intentions. After the more straightforward 13.c4 e6 14.Qc2 Rfe8 15.Bg5 Nf6 16.Bd3 h6 17.Bd2 exd5 18.exd5 Re7 19.a4 Rae8 Black had sufficient counterplay in Navara-Jansa, Karlovy Vary 2005. 13...Rad8 Deviating from Bacrot's previous game, which continued with 13...Ne5 14.Nxe5 Qxe5 15.c4 e6 16.Bb2 Qc7 17.Bxg7 Kxg7 18.a4 with a stable advantage, which White failed to convert into a win in Bacrot-Navara, Gothenburg 2005. We can see that Navara adopted 11...Qc7, too possibly under the impression of the fact that he could not prove an advantage with White. 14.Bg5 Bf6 15.Bh6 Bg7 16.Bxg7 Kxg7 17.a4 e6

18.dxe6?! [Possibly missing Black's strong repyl. 18.c4 , consolidating the centre, would have left White in control of the position.] 18...Nf6! [Of course, 18...fxe6 would have left Black with a terrible structure.] 19.Ng5 h6 20.Nxf7 Bxe4 21.Qb3 Rxf7 22.exf7 Bxb1 23.Rxb1 Qxf7 24.Qxf7+ Kxf7 25.Bc4+ Kf8 26.f3 Rd2 27.a5 bxa5 28.Ra1 Rc2 29.Bd3 Rd2 30.Bc4 Rc2 31.Bd3 Rd2 32.Bc4 1/2-1/2

Anand,V (2803) - Kamsky,G (2671) [C88]
Mtel Masters Sofia BUL (3), 13.05.2006 [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. Actually, I am quite puzzled about the fact that nobody enters the Marshall Attack in top level tournaments any more. To my understanding, this represents some sort of stagnation of the researching process. The Anti-Marshall systems are not bad, of course, and offer practical chances for both sides, but, if objectively speaking White has nothing better than keep his pawn on d3, one natural question is whether he shouldn't just deviate at an earlier stage of the game, with, say, 2.f4. Hopefully, the slight decline in White's result in this tournament will give a new impulse to the investigation of the main line starting with 8.c3 8...Bb7 9.d3 d6 10.a3 Qd7 11.Nbd2. 11.Nc3 is by far more popular in this position. Anand's desire to try something different is explainable by the fact that Kamsky had very recent experience with it, just two days earlier, in his game against Ponomariov. 11...Nd8 Kamsky remains faithfull to the regroupment that he employed in his previous game. The main alternative consists of 11...Rae8 eventually followed by ...Bd8 in order to consolidate the d4-pawn. 12.c3 A new move. After 12.Nf1 c5 13.Ng3 Ne6 14.Nf5 Bd8 15.Be3 Bb6 Black managed to put the d4-square under strict observation in Zhigalko-Staniszewski, Warsaw 2005. However, the immediate occupation of the centre carried out by Anand makes the moves a3 and d3 look like mere losses of time. 12...Ne6 13.d4

13...Rad8! A very instructive moment. Anand might have hoped that Black would be forced to give up the centre because of the pressure against the e5-pawn. Instead of that, Kamsky simply brings another piece into play, underlining the fact that he is 3 whole tempi ahead in the development, which makes the opening of the position undesirable for White, even if this would yield him an extra-pawn. 14.d5 There is no clear refutation of 14.dxe5 dxe5 15.Nxe5 but after 15...Qd6 it is quite obvious that Black has excellent compensation for the pawn. For instance 16.Ng4 (Sligthly trickier than 16.Nef3 when Black could win his pawn back with 16...Nxe4) 16...Nc5 (Now, 16...Nxe4?! 17.Nxe4 Qxd1 could be met by 18.Nef6+! when after 18...Bxf6 there is a reserve-knight available to provoke the deterioration of Black's queen side with 19.Nxf6+) 17.Bc2 (17.e5 is undangerous because of 17...Qc6) and now, even such a "neutral" developing move as 17...Rfe8 leaves White with problems releasing the pressure. By blocking the centre, White intends to make Black's advance in development lose part of its significance.

14...Nf4 In spite of winning a tempo, 14...Nc5?! is hardly adequate in view of 15.Bc2 when Black has problems finding a favourable way of ensuring stability of his knight, for instance 15...a5 16.b4 Na4 17.Bxa4 bxa4 18.c4 when the b7-bishop would be strongly restricted by the white central pawns. 15.Nf1 Ng6 The knight has reached this square after jumping around none less than 5 times, slightly more than this usually happens in this type of structure. The retreat was necessary in view of the positional threat Bxf4, when Black would lose stability in the centre. 16.Ng3 White has managed to reduce his delay in development to bearable proportions. Black has to react energetically before his opponent manages to consolidate his spatial advantage. 16...c6! In the following phase of the game, the fight will be focused around the d5-square. The outcome of this fight will be to a certain extent determining for the further course of the game. 17.Bg5 cxd5 18.Bxf6 Quite logical. White eliminates one of the main defenders of the d5-square. However, I would allow myself to make a general remark here. Along the game, we shall see several similar situations: whenever two enemy pieces will be in contact with eachother, it will be mostly White who will release the tension, sometimes helping Black to slightly improve his position. Event though in certain cases such exchanges are compensated by other positional factors (as is the case with Bxf6), my personal feeling is that the psychologic tendency favoured Black. 18...Bxf6 19.Bxd5 Nf4 The knight jumps back to the square that was made available by the previous exchange of minor pieces. 20.Bxb7 Qxb7 21.Qc2 g6 22.Rad1 d5 Black has managed to unblock this pawn with relative ease. However, this is not by far not the end of the story: after the exchange of the central pawns, the gravity centre will move to e4. Here we can appreciate the merits of the exchange on f6, because the black knight controlled the e4-square, too. 23.exd5 Rxd5 24.Rxd5 Qxd5 25.Rd1 White has lost ground in the centre, but he manages to take over the control of the only open file. 25...Qe6 The queen is well placed here, keeping both wings under control. 26.Qe4

The same can be said about Her White Majesty. Besides creating some threats against the queen side, she blocks the e5-pawn, putting it under some pressure at the same time, thus causing Black problems with his regroupment. Usually, the knight is considered to be the best blocking piece, but after 26.Ne4 Be7 we can hardly speak about a blockade, in view of the threat ...f5 (eventually after the preliminary ...h6). After 26.Qe4 the position is very interesting from strategic point of view. Both sides face problems in the desirable process of putting into motion their pawn majorities. However, there is a significant difference regarding the reasons for this situation in each case. The immobility of the white queen side pawns has a relatively stable character: Black's structure on that wing is simply better. On the contrary, Black's king side majority has no organic deffects. In order to prevent this dangerous majority from advancing, White employs piece pressure alone, which is entirely sufficient for the time being but has a basically temporary character in general. This means that Black will manage to keep his opponent's activity under control, in the long run he could count on an advantage. As for White, he should find ways of maintaining his slight initiative, or to convert it into something of more stable nature.

26...Rb8! [From the point of view of the previous comment, quite a logical move. Black prevents the infiltration of the enemy queen to b7 and threatens to unblock his pawns by means of ...h6, ...Bg7 and ...f5. In fact, the control of the d-file and the h1-a8 diagonal does not yield White anything concrete now, because all the important squares of these lines of communication are under strict Black observation. Here is an illustrative line of what could happen if Black neglects this aspect: 26...Re8 27.Qb7 Be7 28.Rd7 f5 29.Kh2 White's pressure along the seventh rank is quite annoying, counter-balancing Black's supremacy in the centre. The generally desirable 29...e4 would leave Black with some problems of coordination after 30.Nd4] 27.Ne2 By exchanging his passive knight for the active black knight, White makes some central light squares available for his major pieces. 27...Nxe2+ 28.Qxe2 Re8!? Now that the enemy queen has been temporarily removed from e4, Kamsky decides that the rook does not need to guard the b7-square any more and transfers it to a central file. The possible usefullness of such a move can be seen from the following line: 28...e4?! 29.Nd2 Rd8 30.Re1 and Black has to sacrifice a pawn with 30...e3 when his compensation would remain uncertain.; The attempt to make use of the rook on b8 with 28...a5 29.Qe4 Qc4 leads tounclear consequences after 30.Qe3 when the threat Qa7 is quite unpleasant.; However, Black could have initiated the plan of regroupment based on 28...h6 followed by ...Bg7 and f5 immediately, leaving for later the decision about where the rook really belongs. 29.Qe4 The queen gladly returns to this (still) stable square, taking advantage of Black's slight indecision. 29...h6

30.g4?! This move marks the start of White's gradual decline. Anand probably felt that he was about to lose stability in the centre and tried to remedy this problems with the help of the pawns, but he seem to have chosen the wrong pawn! His last move not only fails to block the black mass of pawns, but also weakens the king'sposition. According to Mikhail Golubev in Chess Today, Kamsky recommended 30.h4 during the press conference, with the obvious intention of weakening Black's light squares with h5. Anand suggested 30...Bg7 31.h5 f5 32.Qd5 g5 . Together with GM Constantin Ionescu, we worked out such a continuation: 33.Qxe6+ Rxe6 34.g4!? f4 (34...fxg4 35.Nh2 followed by the installment of the knight on g4 should be avoided.) 35.Re1 (White cannot transfer the knight to e4 yet with 35.Nd2 because of 35...Rd6 when after 36.Kf1 e4 37.Ke2 Re6 Black is certainly not worse.) 35...e4! (This move implies a pawn sacrifice, but otherwise Black would be clearly worse after Nd2-e4.) 36.Nd2 e3 37.fxe3 fxe3 38.Nf1 Bf8! (The only way to activate the bishop) 39.Nxe3 (The attempt to restrict the bishop with 39.b4 only weakens the c3-square and can be met by 39...Bg7!) 39...Bc5 40.Kf2 Re4 . Black's activity combined with the weakness of the g4-pawn maintains the balance even. 41.Kf3 Rf4+ 42.Kg3 Kf7!? (This neutral move puts White in some sort of zugzwang. neither 42...Bd6 43.Rd1!; nor 42...Re4 43.Nc2 are satisfactory for Black.) 43.Re2 (What else? In view of the fork on f2, the only possible knight move would be the passive 43.Nd1 allowing 43...Bd6! with strong initiative for Black.; while a waiting move such as 43.b3 could be answered with the same method by 43...Bb6) 43...Re4 44.Kf3 Rf4+ with a draw by repetition.; The other reasonable way of using the temporary absence of threats from Black's part would consists of advancing the other marginal pawn with 30.a4 aiming to weaken the black queen side structure and improve the own. In this case, too, White would be out of danger.

30...Bg5! This is much more ambitious than 30...Bg7 . Because of the possibility of ...Bc1, White cannot offer the exchange of queens on d5 without prior capturing the bishop. 31.Nxg5 From the situations mentioned in the comment on White's 18th move, this is the first major concession. Black's king side structure becomes more threatening now. On the other hand, it is hard to recommend anything else for White. 31...hxg5 32.Qd5 But this looks like aiming for a draw too openly. In fact, it will be precisely in the absence of queens when Black will be able to advance his strong mass of pawns without fearing to weaken his king's position. From this point of view, strengthening the control of the d-file with 32.Rd5 looks better, when 32...f5?! would be at least dangerous for Black after 33.gxf5 gxf5 34.Qd3. 32...Kg7 33.Qxe6 Rxe6 During the last two moves, Black has improved the position of his both remaining pieces. 34.Rd7 After the recent exchanges, the static elements of the position have become more significant, inclining the balance to Black's favour already. From the former piece activity, White has only retained the control of the only open file. 34...Kf6 35.Kf1 Rc6 36.Ke2 Ke6 37.Rd8 e4

38.f3? As mentioned in the comments to the game Kamsky-Bacrot from the previous round, rook endings are rarely suitable to definitive evaluations. Therefore I shall refrain from naming this move the decisive mistake; I would just mention that now White has huge problems defending his position. This was the last moment when Anand could have improved his queen side situation with 38.a4 for instance 38...bxa4 39.Rd4 Ke5 (39...a3 does not work because after 40.Rxe4+ Kd5 Black has the intermediate 41.Rd4+ leading to an obvious draw after 41...Ke5 42.bxa3 Rxc3 43.Ra4) 40.Rxa4 f5 41.f3 Kf4 42.fxe4 fxe4 and now Black's supremacy in the centre can be met by active queen side counterplay starting with 43.c4. 38...exf3+ 39.Kxf3 Rd6!

Black can ensure the free wandering around of his king because White cannot really exchange rooks: this would remove the last remaining element of activity from his position, when the static balance would be hopeless for him. 40.Re8+ I shall support the rather abstract previous comment with some concrete lines that I analyzed with Constantin Ionescu: 40.Rxd6+? Kxd6 41.Ke4 Ke6 (Of course, not 41...Kc5? because of 42.Ke5 when White is suddenly winning.) 42.b4!? (In view of Black's threat of creating a passed pawn with ...f5, White has to look for some queen side counterplay.) 42...f5+ 43.Kd4 f4 44.c4 (White cannot defend his fortress passively because after 44.Ke4 Kd6 45.Kd4 Black forces the opening of play with 45...a5 . Indeed, continuing the neutral policy with 46.Ke4 leads to disaster after 46...a4 47.Kd4 f3 followed by the march of the black king to b3.) 44...bxc4 45.Kxc4 Ke5! (Just in time to push the white king back.) 46.Kd3 Kd5 47.a4 Ke5 48.Ke2 Kd4 49.Kf3 Kc4 50.b5 axb5 51.axb5 Kxb5 and now the pawn break 52.h4 does not save White after 52...gxh4 53.Kxf4 (or if 53.g5 then 53...h3!) 53...g5+! 40...Kd5 41.b3

A depressing image. The numerous weaknesses along the third rank will soon be very difficult to defend. However, there was no other way of preventing the enemy king from penetrating through c4-b3. For instance, 41.Rc8 would be met by 41...Rc6. 41...Rf6+ 42.Kg2 [Sad necessity. 42.Ke3 allows the exchange of rooks with 42...Re6+] 42...Rc6 43.Re3 f5 44.gxf5 As in so many other moments of this game, White needs to releasethe tension in order to get some breathing space, something he would definitely lack after 44.Kf3 f4 45.Rd3+ Ke5 46.Ke2 Rd6 47.Rf3 Ke4. 44...gxf5 45.Rg3 Ke4!

46.a4 This move comes much too late and in the worst possible form. True, after 46.Rxg5 Rxc3 White cannot defend his third rank with 47.Rg3? because of 47...Rxg3+ 48.Kxg3 Ke3 49.Kg2 f4! (49...Ke2 would lead nowhere because of 50.Kg3!) 50.Kf1 Kf3 and Black manages to win the h-pawn without losing his own. However, 47.Rg6 would have faced Black with some technical problems in view of the reduced quantity of material remaining on board, although the f-pawn would probably decide the outcome of the battle. 46...bxa4 47.bxa4 Kf4 48.Rf3+ Ke5 49.Re3+ Kf6 50.Rd3 f4 Black slowly increases his spatial advantage, while White is reduced to passivity. 51.Kf3 Ke5 52.Kg4 Rd6 53.Rxd6 This is making things faster and is a possible consequence of a mistake of calculation, but one can hardly blame Anand for avoiding such a line as 53.Rf3 Rd1 54.Kxg5 Rg1+ 55.Kh5 Ke4 with an easy win for Black. 53...Kxd6 54.h4 gxh4 55.Kxh4 Kd5 56.Kh3 Ke4 57.Kg2

57...Ke3! Quite possibly, Anand counted on 57...Kd3?? when after 58.Kf3 the white king would be just in time to reach the saving c1-square.; After the accurate 57...Ke3 Black wins a decisive tempo, for instance 58.Kf1 (forced, in view of the threat ...Ke2) 58...Kd3 (The sadistic 58...a5 is also possible, but not really necessary.) 59.Kf2 Kxc3 60.Kf3 Kb4 61.Kxf4 Kxa4 62.Ke3 Kb3 63.Kd2 Kb2 and the a-pawn is unstoppable. After Wijk aan Zee, this is Anand's second defeat by hands of Kamsky in a relatively short period of time. The reader might remember that in the mid-nineties these two outstanding players had to cross swords in two of Candidate's Matches from the parallel cycles of the FIDE and PCA World Championships. Each of them won one of the matches to become challenger of the respective version, meaning that they both remained with a mixture of pleasant and annoying memories. Botwinnik once wrote that he considered the repeated confrontations along the years with certain players as making part of a "life-long" match, with the same psychologic implications as a normal match. Judging from the recent results, we can state that, at least for the moment, Kamsky managed to deal with this delicate psychological issue better than his opponent. Which cannot mean that there will be no further returns of the balance... 0-1.

Ponomariov,R (2738) - Topalov,V (2804) [D45]
Mtel Masters Sofia BUL (3), 13.05.2006 [Mihail Marin]

1.d4 Nf6 2.c4 e6 3.Nf3 d5 4.Nc3 c6 5.e3 a6 Once considered insufficient on the basis of the games of the matches Euwe-Alekhine, this move is on top of the fashion nowadays. 6.c5 b6 7.cxb6 Nbd7 8.Na4 Nxb6 9.Bd2 Nxa4 10.Qxa4

10...Bd7 One of the first games where this line was played continued with 10...Qb6!? Euwe-Alekhine, Netherlands 1935] 11.Ne5 [The most active continuation. Black managed to equalise gradually in Nielsen-Krivoshey, Germany 2003 after 11.Qc2 Ne4 12.Bd3 Nxd2 13.Nxd2 Bd6 14.Nb3 a5 15.0-0 h6 16.Rac1 0-0 17.Nc5 Bc8 18.Be2 Qc7 19.g3 Bxc5 20.Qxc5 Ba6 21.Rfe1 Bxe2 22.Rxe2 Ra6 23.Rc3 Rb8 24.Rec2 a4 when the weaknesses of the b2- and c6-pawns compensate for eachother. 11...Ne4 A new move. Previously, 11...c5 had been tried. After 12.Nxd7 Qxd7 13.Qxd7+ Kxd7 14.dxc5 Bxc5 15.b4 Bd6 16.Bd3 Rhc8 17.Ke2 Ne4 18.Bxe4 dxe4 19.Rhc1 Rab8 20.a3 Be5 Black's superior activity unexpectedly prevailed over White's queen side majority in another historical game, Spielmann-Van den Bosch, Amsterdam 1936. 12.Ba5 c5 13.Nxd7 Qxd7 14.Qxd7+ Kxd7 15.f3 Nf6 16.Rc1 c4 17.b3 The most natural way of activating the f1-bishop. 17...Rb8!? 18.bxc4 Bb4+ 19.Bxb4 Rxb4 20.c5 Rb2

Black's activity offers him adequate compensation for the sacrificed pawn. 21.a4 Ra8 22.Be2 Kc7 23.Rf1 Rab8 24.Rf2 R8b3 25.Bxa6 Rxe3+ 26.Kf1 Rb4 27.Ra2 Reb3 28.Rd1

28...Ng8! A nice move, planning the transfer of the knight to c6, in order to put pressure on White's most vulnerable square, d4. 29.a5 Ne7 30.Be2 Nc6 31.a6 Kb8 32.Rad2 Ra3 33.Kf2 g5 [Quite ambitious. Black could have more or less forced a draw with 33...Raa4 34.Ke3 Ra3+ since 35.Rd3?! would leave the a6-pawn undefended.] 34.h4 Bringing new life into the psoition. 34...gxh4 35.Rh1 Rxd4 36.Rb2+ Kc7 37.Rhb1 Ne5 38.Rb8 Kc6 39.Rc8+ Kd7 40.Rg8 Ke7 41.Rb7+ Kf6 42.a7 Rda4 43.c6 Nxc6 44.Bb5 Ra2+ 45.Kg1 Rxa7 46.Bxc6 Rxb7 47.Bxb7

White has managed to win a piece, but Black's strong mass of pawns keep him out of danger. 47...h5 48.Kh2 Ra1 49.Bc6 Rc1 50.Bb5 Rb1 51.Be2 Re1 52.Ba6 Ra1 53.Bd3 Rd1 54.Bh7 d4 55.Rb8 Kg7 56.Rb1 Rd2 57.Be4 f5 58.Bb7 e5 59.Rb5 Kf6 60.f4 e4 61.Bc8 d3 62.Rxf5+ Kg7 63.Rg5+ Kf6 64.Rf5+ Kg7 65.Rg5+ Kf6 66.Rf5+ Kg7 1/2-1/2.

Replay games online, download PGN

Mihail Marin, 41, Romanian Grandmaster, three times national champion (1988, 1994, 1999), nine times member of the Olympic team, participant in two Interzonals (Szirak 1987 and Manila 1990). In 2005 Marin was the second of Judit Polgar at the FIDE world championship in San Luis. Highest rating: 2604. Author of the ChessBase opening CDs English 1.c4 e5 and The Catalan Opening and the books: Secrets of Chess Defense, Secrets of Attacking Chess and Learn from the Legends. Graduate from the Polytechnic Institute Bucharest (Specialty Electrotechnic) in 1989.

If you have enjoyed the commentary provided by GM Mihail Marin you should try the following training CDs by the same author. They are amongst the best in our ChessBase Shop. Get them now:

Reports about chess: tournaments, championships, portraits, interviews, World Championships, product launches and more.


Rules for reader comments


Not registered yet? Register