Melody Amber, Round 3

It's early yet - there are 16 games to go - but Anand is looking a bit like a runaway train. Today it was Vallejo who was tied to the tracks (ouch!), and now Anand is at 6-0. (And without receiving the sort of "help" offered by Shirov in round 1 or Topalov in round 2.)

Today's round was pretty clean - no big one-move gaffes changing the proper result of the game - but there was one mini-tragedy in the blindfold. Bareev had the advantage against Leko throughout the game, and could have drawn at will had he felt panicky. Instead, despite the 20 second per move increment, he lost on time.

Let's take a quick look at two of the games: Anand's blindfold win over Vallejo and the rapid game between Svidler and Morozevich.

Anand,Viswanathan (2786) - Vallejo Pons,Francisco (2686) [B90]
Amber Blindfold Monte Carlo MNC (3), 21.03.2005

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Be3 e6 7.g4 h6 8.Bg2 e5 9.Nf5 g6 10.Qe2 gxf5 11.exf5

11...Bd7 This seems to be a novelty. Rather than suffer permanent pressure on the light squares (especially d5), Black attempts to deaden White's initiative at the cost of the exchange. It's a reasonable transaction materially, since he's up a piece for a pawn, but it turns out that White's initiative persists just the same. [11...Rg8 is the most common move, but one that hasn't brought Black much success. 12.Bf3 Nc6 13.0-0-0 Be7 14.h4 Nd4 15.Bxd4 exd4 16.Rxd4 Kf8 17.Qe3 Rh8 18.g5 hxg5 19.hxg5 Rxh1+ 20.Bxh1 was how the game xx-yy continued, and you shouldn't be surprised to know that Black's miserable position collapsed in just a few more moves.] 12.Bxb7 Bc6 13.Bxa8 Bxa8 14.Rg1 Nbd7 15.0-0-0 Black doesn't have to worry about the d5 square, but now White gets to attack for free. The big question for Black is what exactly he's supposed to do with his king. Castling would be suicide, but White will pry the center open in due course as well. Does Black have any trumps by which to gain counterplay? 15...Be7 Black seems to be "threatening" d5 or O-O, but both are quite bad. [The ambitious 15...d5 looks threatening but actually speeds White's attack after 16.f4 Black's options are all bad: 16...d4 17.Bxd4 is terminal; 16...e4 drops the d-pawn to 17.g5, and on any normal move White plays 17.fxe5 Nxe5 followed by 18.Bd4 or 18.Bf4, blasting open the center.] 16.h4 Qa5 Protecting a6 and supporting the d5 square, so that 17.g5 hxg5 18.hxg5 Nd5 becomes possible. 17.Bd2 Nd5 [17...d5 works well in every variation but one: 18.g5 hxg5 19.hxg5 d4!? 20.gxf6 Nxf6 21.Rge1! dxc3 22.Bxc3 Qb5 23.Qxb5+ axb5 24.Rxe5+- and Black's position falls apart due to the threats of 25.Rde1 and 25.Rxe7+ Kxe7 26.Re1+.] 18.Nxd5 Qxd5 19.Qxa6 Bb7 20.Qa3 Preparing 21.Bb4, with serious pressure on the d6 pawn. [20.Qa7 , preventing the 20...Qc4+21...Be4 idea mentioned in the next note, might be an improvement.] 20...Qc6 [20...Qc4 might be a real improvement, with the idea that 21.Bb4 is met by 21...Be4, with at least a bit of counterplay.] 21.Bb4+- Now Black is just busted. 21...Nb6 [21...Nc5 22.f4 e4 23.Qc3 Rg8 24.g5 followed by g6 or f6, f5 and g6 will be more than Black can bear.] 22.g5 [22.f6 straight away, without any preparation, looks even stronger, according to my "little German friend" (that phrase, for those who haven't seen it before, is used to refer to chess engines sponsored by ChessBase - usually some version of Fritz, though my own preference is for Shredder 9): 22...Nc4 23.Qc3 Bxf6 24.b3 Nb6 25.Qxc6+ Bxc6 26.Rxd6 with an open buffet.] 22...hxg5 [22...Nd5 saves the d-pawn for the moment and lets Black kick on a bit, though his weaknesses are chronic and thus almost surely terminal.] 23.hxg5 Kd7 24.f6 Nc4 25.Qc3 Bf8 26.b3 Nb6 27.Qxe5

The remainder is unsuitable for small children and those with weak hearts. 27...Rh4 28.Bxd6 Qxd6 29.Rxd6+ Bxd6 30.Rd1 Nc8 31.g6 Rh1 32.Qf5+ 1-0

Svidler,Peter (2735) - Morozevich,Alexander (2741) [C11]
Amber Rapid Monte Carlo MNC (3), 21.03.2005

1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.e5 Nfd7 5.f4 c5 6.Nf3 Nc6 7.Be3 a6 8.Qd2 b5 9.Be2 Qb6 10.Nd1 b4 11.0-0 a5 12.c3 Be7 13.Bf2 0-0 14.f5 bxc3 15.bxc3 cxd4 16.cxd4 Ba6 17.Bxa6 Qxa6 18.Ne3 Bb4 19.Qd1 Rac8 20.Ng5 Nxd4 21.Qxd4 Bc3 22.Qh4 h6

Let's start here. White is up a piece for a pawn and his pieces are congregating in the neighborhood of the Black king. Black can at least do something about the first point - he's getting his material back - but what material he recoups is up to White. White decides the time has come to go Pulp Fiction on the Black king... 23.f6? [The patient 23.Nf3 leaves White with a big advantage after 23...Bxa1 (23...Nxe5 24.Nxe5 Bxe5 25.Rad1 is probably winning for White, too: Black has no real compensation for the piece, as his pawns aren't at all dangerous. In fact, it's White who maintains the initiative here - a quick Ng4 might produce some real attacking chances against the relatively lonely Black king.) 24.Rxa1 Qe2 25.Rf1 followed by f6, Ng4 and so on, with a crushing attack.] 23...hxg5 24.Qxg5 Nxf6 and White's decision was bad. Could Svidler have missed this move? Maybe he had some sort of hallucination in which Black had already exchanged the bishop for the rook on a1? Whatever the case, Black both stops the mate and winds up ahead in material. 25.exf6 Bxf6 26.Qh5 Bxa1 27.Rxa1 If White's minors were coordinated here, then Black would be in serious trouble: put the knight on g4, the bishop on d4, and then moves like Qg5, Bxg7, Nf6+ and so on are crushing. Black mustn't delay for too long, therefore - he needs to bring defenders to the kingside. 27...Qd3! 28.h3 Rb8! this ensures the trade of the White rook (except after 29.Qd1, which allows the trade of queens), after which White's attacking chances fall to approximately zero. It's a good rule of thumb in such positions that the side with the rook against the two pieces wants to trade the other side's rook. I can think of at least two reasons underwriting this principle. The first is similar to the idea of trading off the opponent's bishop when it's defending a weak color complex; to wit, you eliminate the only piece capable of defending certain sorts of attacks. In other words, since Black's advantage is in the vertical and horizontal department, the best way to maximize it is by getting rid of White's vertical/horizontal pieces (that includes the queen too, of course). The second reason is that the White rook, given the opportunity, would complement the minor pieces quite nicely; without the rook, however, their range of effective action is dramatically reduced. Careful reflection on the above will benefit your chess considerably, I think, but for those of you just after the bottom line, 28...Rb8 means that White's attack is over and Black should win. 29.Ng4 Rb1+ 30.Rxb1 Qxb1+ 31.Kh2 f6!

The point is to keep the knight off of e5 and to render Bd4 worthless (32.Bd4? e5). N.B. The Black queen covers g6, so 32.Nh6+ doesn't give White perpetual check. 32.Bc5 Rc8 33.Bd6 Qh7 Trading queens is hopeless, so Svidler pitches a piece and hopes for a miracle. 34.Nxf6+ gxf6 35.Qg4+ Kf7 36.Qa4 Qg8 37.h4 Rc4 38.Qd7+ Kg6 0-1

Round 3 summary:

Blindfold:

Morozevich-Svidler 1/2-1/2
Bareev-Leko 0-1
Gelfand-van Wely 1-0
Anand-Vallejo 1-0
Kramnik-Topalov 1-0
Shirov-Ivanchuk 0-1

Rapid:

Svidler-Morozevich 0-1
Leko-Bareev 1-0
van Wely-Gelfand 1/2-1/2
Vallejo-Anand 0-1
Topalov-Kramnik 1/2-1/2
Ivanchuk-Shirov 1/2-1/2

Standings:

Blindfold:

Anand, Kramnik 3
Svidler 2.5
Ivanchuk, Leko 2
Gelfand, Morozevich 1.5
Bareev, Vallejo 1
van Wely .5
Shirov, Topalov 0 (!!)

Rapid:

Anand 3
Ivanchuk, Leko, Morozevich 2
Bareev, Svidler 1.5
Gelfand, Kramnik, Shirov, Topalov, Vallejo, van Wely 1

Overall:

Anand 6
Ivanchuk, Kramnik, Leko, Svidler 4
Morozevich 3.5
Bareev, Gelfand 2.5
Vallejo 2
van Wely 1.5
Shirov, Topalov 1