So far, so good for the 2012 edition of Biel: the games have been full of excitement and youthful energy. That's not surprising in a tournament where 24-year-old Hikaru Nakamura is in the graybeard half of the table.
Speaking of Nakamura, he drew again today. He was pressing throughout against Anish Giri in a Catalan-turned-Bogo-Indian, but the youngster (youngerster?!) held and maintained a share of the lead.
The battle between Etienne Bacrot and Alexander Morozevich was a bit mysterious at one moment, at least to me as an online spectator. Bacrot was White in a Marshall Gambit Slav, and in a well-known theoretical position after 9...Qxg2 he thought for about 45 minutes - at least if the relay on ICC was correct. (There weren't any delays with the transmission of the other games, so that shouldn't be the explanation.) His response after that deep think (or brief nap) was the conventional one. Ironically, Morozevich's reply to 10.Qd2 was the very unusual 10...e5; 10...Nf6 is standard. (In Ruslan Scherbakov's book The Triangle System, he spends 12 and a half pages on 10...Nf6, and says only this about Morozevich's move: "10...e5!? followed by ...Bf5 might be playable though.")
Bacrot's natural reply 11.Bxe5 was already a new move, and the position grew incredible sharp. Seriously analysis of this game would take some time, but what is clear is that 21...Ba6 was a fatal error; Black needed to bravely play 21...bxc6 and hope that White had nothing better than 22.Qxa7 Rxd6 23.Qa8+ Kc7 24.Qa7+ Kc8 25.Qa8+ etc. He doesn't seem to. After 21...Ba6? Bacrot landed some nice blows: 22.Ng5! Nxg5(?) 23.Bd7+! Kxd7 24.Qe7+ Kc6 25.Qc7+ and Black preferred resignation over allowing 25...Kb5 26.Qc5+ Ka4 27.Qb4#.
Magnus Carlsen vs. Wang Hao was a 4.Qc2 Nimzo-Indian with a quick e4. They were in new territory pretty quickly - 8.d5 was a new move in what was already a rare position - and it was soon clear that the battle would be between Black's structural advantages and White's initiative and attacking chances. After 16...h6 the spectators were looking with bloodlust at ideas like 17.Bxh6, but that appears to be inconclusive: 17...gxh6 18.Rhg1+ (18.Qd2 will transpose) Kh8 19.Qd2 Nh7 20.Qxh6 Rf7 leaves White with sufficient compensation after 21.Qxd6 or 21.Re6, but not more than that.
So Carlsen kept squeezing, but maybe he could have played Bxh6 on move 21. Again, he preferred to keep up the pressure, and on move 22 he induced an error. Wang Hao should have played 22...Nh5, aiming to further activate one or both of his knights and maybe swapping off a White attacker or two. Instead, his 22...Nxd5? gave Carlsen what looks like a good opportunity, even if he chose not to play it: 23.Bxc5 bxc5 (23...Rxf5? 24.Bd6 or 24.Bd4 is crushing) 24.Be6 Nf4 25.Bxf7 Nd3+ 26.Kb1 Rb8+ 27.Ka1 Qxf7 28.Rxg7 Qxg7 29.Rxg7 Kxg7 30.Qa4 is not an ending Black should draw.
But Carlsen chose 23.Bd4, which while probably not as good certainly maintained a pleasant advantage. (23.Bxh6 was also possible.) A few moves later he gave up his rooks for Black's queen and g-pawn, and with his very strong bishops Black's position was hard to play; indeed, he was soon in something pretty close to zugzwang. His last chance to keep the ship sailing, at least for a little while, was with 31...Rh7. After 31...Rfe7? the loss was guaranteed and speedy, and Black resigned after a forcing sequence culminating with 35.f4 because after 35...Rf5 36.Bxf6 Rxf6 Black's king and rook are parted by 37.f5+ Ke5 38.f4+.
So Carlsen joins Giri in the lead, and pushes his unofficial rating to the verge of 2840. Here are tomorrow's pairings, with player scores given in parentheses. Note that the totals are based on 3-1-0 scoring: