(3838) Shirov,Alexei (2709) - Anand,Viswanathan (2817) [B12]
24th Leon Masters Leon ESP (5), 05.06.2011

1.e4 c6 2.d4 d5 3.e5 Bf5 4.h4 h5 5.Bg5 Qb6 6.Bd3 Bxd3 7.Qxd3 Qa6
Rarer than 7...e6, but it's quite sensible as it disallows the c2-c4 advance that White can play after 7...e6, e.g. 8.Nd2 Qa6 9.c4. [7...Qxb2 looks obvious, but it's extremely risky, as you'll see by taking a tour through this gallery of massacres: 8.e6 Qxa1 (8...fxe6 9.Nf3 Nd7 10.0-0 Qxa1 (10...e5 11.Qg6+ Kd8 12.Nbd2 e4 13.Rab1 Qc3 14.Rb3 Qxb3 15.Nxb3 exf3 16.Na5 Ngf6 17.Rb1 b6 18.Nxc6+ Kc7 19.Ne5 Nxe5 20.dxe5 Nd7 21.e6 Nc5 22.Qf5 Rd8 23.Rd1 d4 24.Qe5+ Kc6 25.Bf4 Rd6 26.Rxd4 Rxe6 27.Qc7+ Kb5 28.a4+ 1-0 Ivakhinova,I (2218)-Ostapenko,D (2320)/Irkutsk 2010) 11.Qb3 e5 (11...a5 12.Nbd2 a4 13.Qxb7 Rb8 14.Qxb8+ Nxb8 15.Rxa1 Nd7 16.Rb1 g6 17.Ne5 Nxe5 18.dxe5 Bg7 19.Nf3 Bh6 20.Rb8+ Kf7 21.Rc8 Bg7 22.Kf1 c5 23.Ra8 Nf6 24.Rxh8 Bxh8 25.exf6 exf6 26.Bc1 c4 27.Nd4 e5 28.Nb5 Ke6 29.Nc3 Bg7 30.Nxa4 d4 31.Ba3 Kd5 32.Ke2 1-0 Galkin,A (2608)-Jovic,S (2285)/playchess.com INT 2007) 12.Nc3 Qxf1+ 13.Kxf1 e4 14.Ne5 Nxe5 15.Qxb7 1-0 Firman,N (2494)-Makarov,V (2172)/Alushta 2006; 8...Nh6 9.Nf3 f6 10.Bd2 Na6 11.Ba5 f5 12.0-0 Nb4 13.Qd2 Nxa2 14.Nc3 Nb4 15.Qg5 b6 16.Bxb6 1-0 Kaufeld,J (2288)-Fuest,M (2230)/Germany 2007) 9.Qb3 Qxd4 (DM: 9...b5 10.Nf3 a5 11.0-0 a4 12.exf7+ Kxf7 13.Ne5+ Ke8 14.Qh3 Kd8 15.Nc3 Qb2 16.Nf7+ Kc7 17.Rb1+- Qxc2 (17...Qa3 18.Nxb5+ cxb5 19.Qxa3+- ; 17...Qxb1+ 18.Nxb1+- ) 18.Bf4+ Kb7 19.Nd8+ Kb6 20.Qc8+- ) 10.Qxb7 fxe6 11.Be3 Qa4 12.Qxa8 Qa5+ 13.Nd2 Qd8 14.Ngf3 Nf6 15.0-0 g6 16.Rb1 Nbd7 17.Qxc6 Bg7 18.Ng5 1-0 Perunovic,M (2580)-Lalith,B (2480)/Dubai 2009; The most common choice is 7...e6 8.Nd2 and only now 8...Qa6 . A previous Shirov game, played recently during his competitively and creatively successful event in Lublin, went 9.c4 Nd7 10.Ne2 Be7 11.Nf4 dxc4 12.Nxc4 Nb6 13.Rc1 Nxc4 14.Rxc4 g6 15.0-0 Bxg5 16.hxg5 Ne7 17.Qb3 Rb8 18.Nd3 Qb6 19.Qa3 a6 20.Nc5 Qc7 21.Ne4 Kf8 22.Nf6 Kg7 23.Re1 Nf5 24.Re4 Rbd8 25.Qc3 Qb6 26.Rf4 h4 27.a4 a5 28.Kh2 Qa7 29.Kh3 Qb6 30.Kh2 Qa7 31.b3 Qb6 32.Rc5 Ra8 33.Rc4 Rad8 34.Rc5 Ra8 35.Kg1 Qa7 36.Re4 Qb6 37.Rf4 Qa7 38.Kh2 1/2-1/2 Shirov,A (2709)-Roiz,M (2660)/Lublin 2011]

8.Qf3 e6 9.Ne2 c5 10.c3N
[10.0-0 Nc6 11.c3 Nge7 12.Nd2 cxd4 13.cxd4 Nf5 14.Nb3 Qb6 15.Rfd1 Be7 16.Bxe7 Ncxe7 17.g3 g6 18.Rac1 0-0 19.Rc3 1/2-1/2 Pavlovic,M (2483)-Trepat Herranz,J (2322)/La Massana 2010]

10...Nc6 11.Nd2 Nge7 12.Nb3
[12.0-0 would transpose to Pavlovic-Trepat Herranz, given in the note to White's 10th move.]

12...cxd4 13.cxd4 Nf5 14.0-0
And we've transposed back to that game after all.

14...Be7N 15.Bxe7 Ncxe7 16.g3 b6
It's pretty safe to say that White has nothing here. Even less than nothing - Black has the long-term advantages here, as barring major changes his d-pawn will always be the weakest pawn on the board.

17.Nf4 g6 18.Nh3=/+ 0-0 19.Qf4
Shirov will not play to "hold" the position, but will try to break through on the kingside before Black's long-term advantages (queenside control, White's light-squared weaknesses on c4, f5 and elsewhere, his better control of the c-file, and the weak d4 pawn) can bear fruit.

19...Qe2 20.Rfd1
[20.Qg5 Qxb2 21.Nf4 Kh7 22.g4 hxg4 23.h5 Rac8 24.Rac1 Rxc1 25.Rxc1 Kg8 26.Rc7 Qxa2 27.hxg6 Qxb3 28.gxf7+ Kh7 (28...Kxf7?? 29.Qg6# ; 28...Kh8?? 29.Qh5+ Kg7 30.Nxe6# ) 29.Qh5+ Nh6 30.Rxe7 Qd1+ 31.Kg2 (31.Kh2 g3+ 32.fxg3 Qxh5+ 33.Nxh5 Rxf7 34.Rxe6 Nf5= ) 31...Qf3+ 32.Kg1 Qd1+= ]

[20...Qxb2! 21.g4 hxg4 22.Qxg4 a5 23.h5 a4 24.hxg6 Nxg6 (24...axb3 25.gxf7+ Kxf7 26.Qh5+ Kg7 27.Ng5 Rh8 28.Nxe6+ Kg8 29.Qg5+ Kf7 30.Qf6+ Kg8 31.Qg5+= ) 25.Ng5 Kg7 Black has neutralized White's kingside threats, and can return to his queenside triumphs. White can avoid immediate losses with 26.Rdb1 Qa3 27.Nd2 , but Black is winning all the same after 27...Rh8 28.Ngf3 Rh6-+ That the plan of doubling on the kingside is quite dangerous can be seen from the following variation: 29.Rxb6 Rah8 30.Kg2 Nfh4+ 31.Nxh4 Rxh4 32.Qg3 Rh3!! 33.Qxa3 Nf4+ 34.Kf1 Rh1# ]

21.Rd2 Qg4 22.Qxg4 hxg4 23.Ng5
[23.Nf4 ]

23...a5=/+ 24.f3 Rc4!?
[24...gxf3 25.Kf2 a4 26.Nc1 Nc6 27.Nxf3 f6 28.Ne2 Nfxd4! 29.Nexd4 Nxd4 30.Nxd4 (30.Rxd4? fxe5-+ ) 30...fxe5+ 31.Kg2 exd4 32.Re1 Kf7-/+ ]

25.Kf2 Rfc8 26.fxg4 Nh6 27.Rad1 a4
[27...Nxg4+ 28.Kf3 Nh6 29.g4 looks like an argument for 24...gxf3, though here too Black remains slightly better. (Then again, there's almost no remotely plausible circumstance, given this structure and Black's iron grip on the c-file where he won't be at least a little better.)]

The knight looks ridiculous, but the important thing is that it's doing a real job by keeping c2 under control.

28...Nxg4+ 29.Kf3 Nh6
[29...Nf5!? 30.Kxg4 Ne3+ 31.Kf3 Nxd1 32.Rxd1 Rc1 would win, were it not for 33.Ke2= ]


Black's knight on h6 doesn't make for a much prettier sight than its counterpart on a1.

31.Nf3 b5 32.Ne1?
White doesn't like living with that knight on a1, but he should have maintained the rest of his position. Now Black obtains a winning position, thanks to the following tactical shot. [White is okay after a neutral waiting move like 32.Rg1 Black's pieces all look good, but they're operating at or near full potential and there's no win. White will soon play g4 and ask Black if he has any ideas about making progress.]

32...f6!!-+ 33.exf6 e5+! 34.Kxe5 Ne3
Good enough, but [34...Re8+! was even better: 35.Kf4 (35.Kxd5? Ng4 36.-- Nxf6# (36...Nge3# ) ) 35...Rc6 36.Nac2 Rxf6-+ puts the White king in a mating net, to say nothing of the discovered check possibilities.]

[35.Kf4 Nxd1 36.Rxd1 Rf8 will win in the long run - but at least White gets a long run.]

35...Re8+ 36.Kf4 Rc6
White's king is again in the mating net.

37.Kg5 Re4?!
Fastest if you're a computer, but a bit of needless and risky showiness that could have cost Anand half a point. [37...Nf7+ ; and 37...Kh7 both win very easily and routinely, without even a touch of risk. For example:; 37...Nf7+ 38.Kxg6 (38.Kf4 Rxf6# ) 38...Re4 Now this! 39.Re2 Rg4+ 40.Kh5 Rxf6 41.Rxe3 Rg7 followed by ...Rh6# or ...Rf5#.]

38.Kxh6 Rg4??
[38...Ng4+ was the best way: 39.Kg5 (39.Kxg6 Rxf6+ 40.Kg5 Kg7 41.h5 Nh6 followed by one of 42...Rg4#, 42...Rf5# an 42...Nf7#.) 39...Nxf6 40.Kh6 Rg4 41.h5 Rc7 42.hxg6 Stopping ...Rh7#, but it's still mate in two: 42...Rg7 43.Nac2 R7xg6# (43...R4xg6# ) ]

[39.f7+! Kxf7 40.Kh7! It's hard to believe, but not only is there no mate here, Black isn't even better anymore. 40...Rc7 41.Rf2+ Nf5 42.Nf3 Kf6+ 43.Kh8 Re4 44.Re1 is one possibility, leaving Black with nothing more than a perpetual check: 44...Rc8+ 45.Kh7 Rc7+ 46.Kh8= ]

39...Nf5+ 40.Rxf5 gxf5 41.Kh5 Rc7
...Rh7# is unstoppable, so White resigned. 0-1