The biggest story in today's action at Biel was a story of inaction. Alexander Morozevich, who started with two losses, withdrew pleading illness. (The nature of that illness hasn't been publicly specified; we wish him the best.) He didn't play in round 3, but his third round game will be made up on Sunday (the scheduled rest day) by his substitute, Viktor Bologan, who will play in Thursday's round 4 game with Hikaru Nakamura. Unfortunately for Bologan, he will inherit Morozevich's losses. Sometimes the substitute gets to start from scratch, but already three days behind it's a bit late for that.
Only two games were played today (Wednesday), but they were exciting, full of adventure and both finished with a decisive result. It was especially entertaining to see Nakamura dare the Polugaevsky Variation of the Najdorf against Wang Hao. The variation has been at death's door for a good 50 years, but intrepid souls (including of course the late, great Lev Polugaevsky himself) are always finding new resources to keep the ball rolling.
On move 17 Wang Hao tried a near-novelty, 17.Nfg5, which had been played only once by a 2000-level player in a loss to a GM. This probably wasn't any home prep - who expects the Polugaevsky Variation anymore? - and after 17...h6 he played a new move, 18.Qh3. In the earlier game White played 18.Nxe6?, a sac that left him down a piece for nothing. Another sac was playable though: 18.Nxf7, which seems to lead to a perpetual: 18...Kxf7 19.Ng5+ hxg5 20.fxg7+ Kxg7 21.Qxg5+ Kh7 22.Bd3+ Rxd3 23.Qh5+ Kg7 (23...Kg8? 24.Rxf8+ gives White a big advantage) 24.Qg5+ Kh7 etc. (In fact, Nxf7 was a decent option on moves 19, 20 and 22 as well.)
Wang Hao's move wasn't bad, but he didn't play with the kind of energy one needs against the Polugaevsky. The position remained very complicated, but Nakamura was making real progress and stood objectively better, though in a position where the cost of an inaccuracy would be much greater for him than for White. Sure enough, Nakamura made a serious slip with 26...Nd6. It's a natural move, centralizing a piece, moving it closer to his king and trying to trade off an attacker. Unfortunately for him, there was a tactical problem: 27.Rxd5! Before 26...Nd6 it would have been possible to respond with ...Rxd5; here, 27...Qxd5 was forced, and after 28.Bxe6! fxe6 29.f7+ all kinds of nasty tactics faced Black.
Both players were in real time trouble now, and after 29...Kd8?! (29...Ke7 was better, when White is better but maybe not yet winning after 30.Qxh6! Rdd8 31.f8Q+ Rdxf8 32.Qg7+) 30.Nxe6+ White was winning. A key point was that if Black later played 33...Qxd6, then 34.Rxd8+ Rxd8 35.Nxd8+ Kxd8 36.Qd3 produces a won king and pawn ending for White. So Nakamura tried 33...Kb8, but when the time scramble ended after move 40 White was two pawns up in a queen and rook ending. Wang Hao played too passively though, and Black still had some chances to resist with 46...Kb7. Instead, Nakamura erred one last time with 46...Qg3?, allowing White to transpose play into a trivially won rook ending with 47.Qh2, forcing immediate resignation.
In an interview after his win in round 2, Etienne Bacrot said of his round 1 defeat to Wang Hao on the black side of a Bayonet King's Indian that he had forgotten his prep. Today he had the chance to remember his preparation, as Anish Giri also played the Bayonet against him. It was Giri who deviated first, however, going for 12.f3 rather than 12.Bf3. This is in fact the main move, with nearly 500 games in ChessBase's online database, including van Wely-Bacrot from this year's French Team Championship.
They followed that game and quite a few others for a while. On move 19, some early games (e.g. Pachman-Taimanov, from the 1967(!) Capablanca Memorial) went towards an immediate perpetual with 19.Nxa8 Qxg3+ 20.Kh1 Qh3+ etc. 19.Rf2, as played by Giri here and van Wely many times, is the winning try. Here Black has two tries: 19...Nxe4, when White may have an edge after 20.fxe4 Rxf2 21.Kxf2 Rf8+ 22.Ke3 Qxg3+ 23.Kd2 Rf2 24.Ne8 (van Wely has gone 1-1 here: a win over Radjabov in 2008 and a loss to Stellwagen in 2009). "Mr." Houdini thinks it's 0.00 equal, but as I'm sure that neither van Wely nor Giri would allow a forced draw there are doubtlessly some mysteries to be explored. (Maybe there's something instead of 24.Ne8, or perhaps there's some new life to be found with 20.Rh2 - a move van Wely used - and lost with - to Jean Marc Degraeve back in 2000.)
Anyway, the other theoretical move, chosen by Bacrot both here and against van Wely earlier in the year, is 19...Rac8. That game was followed through move 25, when instead of van Wely's 25.c6 Giri chose 25.Be3. It looks like a good novelty. 25...Qxf5 was the right reply, but after 26.Rf1 it seems Black should play 26...Kh8, when White is only a very little better after 27.Bd3 e4! 28.fxe4 Qh3 29.Rg3 Qh4 30.Qf4 Qxf4 31.Bxf4 Rxc5. Bacrot's 26...Bf8 wasn't as good. Now 27.Bd3 e4 28.fxe4 cost White the exchange, but it was in fact a winning "sacrifice". White's bishops and central passers absolutely dominated, and the rest of the game was a piece of cake for the tournament's youngest player and (for now) sole leader.
The standings are slightly confused at the moment, but look like this (based on 3-1-0 scoring):
1. Giri 7 (3 games)
2. Wang Hao 6 (3)
3. Carlsen 4 (2)
4. Bacrot 3 (3)
5. Nakamura 2 (3)
6. Morozevich/Bologan 0 (2)
Round 4 Pairings:
- Bologan - Nakamura
- Giri - Wang Hao
- Bacrot - Carlsen