Shakmaty Bereolos - The Official Chess Site of Peter Bereolos

12/20/00 - Burris-Bereolos 2000 Knoxville City Championship

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Be2 0-0 6.Bg5 Na6 7.Qd2 e5 8.d5 Qe8 9.Bd3 Another choice is 9.Bd1, which leads to the game position after 9...Nc5 10.Bc2, but deprives Black of the option 9...Nh5. 9...Nc5 I did make the immediate jump 9...Nh5 against him in the qualifier. 10.Bc2 a5 In the qualifier game, I went for a more double edged setup with ...a6 but only after waiting until he had castled queenside. Here, 10...a6 is probably dubious since White can play 11.b4 then castle kingside having gained a couple of queenside tempi versus Kings Indian lines where Black plays ...Nbd7 and ...a6. 11.Nge2 Nh5 12.f3 In the game Seirawan-Piket Wijk aan Zee 1991, White tried to exploit Black's move order by 12.Nb5, but this move looks artificial to me taking the knight away from the center. Black played 12...Qd7 and went on to score a crushing victory. 12...f5 13.exf5 gxf5 14.0-0-0 After the game David claimed that GM Petursson assesses this position as +/=. That may be the case, but Black still has play. White's plan revolves around playing g4 to obtain the e4 square. Black will try to counter this by ...e4 or play on the dark squares. 14...b6 14...Bd7 would lead to a position known to theory, but I was already anticipating this bishop moving to f5. 15.h3 A consistent move, preparing g4 to stake his claim on the light squares. After the game, several spectators insisted that the immediate 15.g4 should be played, but I don't really see what White has for his pawn after 15...fxg4 16.fxg4 Bxg4 17.Rhg1 Bf5 since his knights will not participate in the attack. 15...f4

This looks like it just surrenders e4, but Black's pieces can still compete for that square and Black is also eyeing g3 and g2 as weaknesses in the White camp. 16.Ne4 During the game I thought 16.Rhe1 was better, but Black should still be fine after 16...Bf5. Another plan is 16.Bh4 intending Bf2 and Bxc5, which would remove one of Black's guardians of e4 and resolve White's worst piece the Bg5. Still, I think Black is doing OK even if White conquers e4 since he will still have play against g2. The only thing Black must take care to avoid is the unpleasant endgame with a White knight against a Black dark-squared bishop. 16...Nxe4 17.Bxe4 Bf5 18.Qc2 Qg6 19.Bxf5 White can't allow the exchange on e4 since then he wouldn't have any compensation for the g-file weaknesses. 19...Rxf5 20.Bh4 Bf6 21.Bxf6 I think he started to realize something had gone wrong here. Swapping off the bad bishop isn't what White really wants to do. I think it is better to retreat, but Black is still very comfortable after 21...Rg5 with play against g2. If White ever pushes g4, Black can take en passant exposing fresh weaknesses at f4 and f3 and the bishop will find a home on the h6-c1 diagonal. 21...Rxf6 22.Nc3 Qxc2+ 23.Kxc2 Ng3 24.Ne4 Leads to a very unpleasant, probably lost, ending for White, but 24.Rhe1 Nf5 25.Rd2 Rg6 will soon cost White his g-pawn. The only weakness in Black's camp is at c7, but the rooks can easily guard this until the king can walk over. In the meantime, White has no counterplay. 24...Nxe4 25.fxe4 Rg6 26.Rd2 Kf7 27.Rf1 Rg3 28.Rff2 28.Re2 is a bit better to avoid an immediate exchange of rooks. I think Black can still win in rather straightforward manner by bringing the King to the queenside and penetrating to d4 with ...b5. White is again rather helpless since his rooks will be tied to the g-pawn and his King must guard d3 or Black can play Rd3 Rgg3 and Rde3 exchanging a set of rooks and winning as in the game. 28...Re3 29.Rde2 Rxe2+ I saw the idea of a winning pawn ending here and there is little he can do to stop it. 30.Rxe2 Rg8 31.Kd2 Rg3 32.Ke1 Kf6 33.Kf2 Kg5 34.Rc2 [34.Kg1 f3 ; 34.Re1 Kh4 35.Re2 Re3] 34...Re3 35.Re2 Kh4

Threatening 36...Rxe2+ 37.Kxe2 Kg3 38.Kf1 f3 and the e-pawn soon falls. 36.Rxe3 fxe3+ 37.Kf3 a4 38.b3 a3 39.b4 e2 It was a bit cleaner to run him out of moves with 39...h6 40.c5 b5 41.c6 h5 40.Kxe2 Kg3 41.Kf1 b5 Removing the d-pawn's defender. 42.cxb5 Kf4 43.Ke2 Kxe4 White's potential outside passed pawn cannot compete against the two central passed pawns. 0:1

12/18/00 - Alexander-Bereolos 2000 Knoxville City Championship

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.d3 Nf6 5.0-0 d6 6.c3 0-0 7.Be3 The more usual moves 7.Bb3 or 7.Nbd2 are a bit better. 7...Bb6 8.Bxb6 I think Black has completely equalized after this exchange. 8...axb6 9.Qe2 Nh5 My other choice was to move the other knight to the edge 9...Na5 picking off his bishop since 10.Bb5 is met by 10...Nb3 10.d4 10.Nxe5? Nxe5 11.Qxh5 Bg4 is a typical trap in these types of positions 10...Nf4 11.Qe3 Qf6 12.Rd1 Bg4 13.Nbd2

13...Rfe8 I also considered 13...Ra5!? and 13...Ne7. The latter might be best, especially from a practical point of view. The position is still probably equal, but he had already consumed a massive amount of time (I already held an edge of over half an hour on the clock). In the game continuation he can play a few obvious moves quickly. After 13...Ne7, White's plan isn't as clear 14.Bb5 I guess another choice is 14.d5 trying to make the rook look misplaced on e8, but then White's bishop would be bad and he would again be in search of a constructive plan. Not 14.h3? Bxh3 15.gxh3 Qg6+ 16.Kf1 Qg2+ 17.Ke1 Qh1+ and 18...Ng2+ winning the queen 14...exd4 15.cxd4 d5 16.e5 Qh6 17.g3 Maybe 17.Kh1!? intending g3 and a trade of queens, but Black can avoid this with 17...g5!? 17...Nh3+ 18.Kg2 Qh5 19.Rdc1 Re6 20.Bxc6 bxc6 21.Nh4 Rh6 22.f3 Bd7 23.f4 g5 23... Nxf4+ 24.Qxf4 g5 is another choice, but I wanted to keep some pieces on the board 24.fxg5 Nxg5

25.Rf1? After the game we looked at 25.Rc3 and it appears White can still hold. He was desperately short of time here with less than two minutes left. 25...Bh3+ 26.Kg1 Bxf1 27.Kxf1 f6 28.Nf5? A final blunder that ends it quickly, but his position was lost anyway. 28...Qh3+ 0:1

12/17/00 - Bereolos-Barber 2000 Knoxville City Championship

1.d4 Nf6 2.c4 c5 3.d5 e6 4.Nc3 exd5 5.cxd5 d6 6.e4 g6 7.f4 Bg7 8.Nf3 0-0 9.Be2 Re8 10.e5 The sharpest, although I suspected Nick was well prepared. Still, you have to go for the main lines sometimes. 10...dxe5 11.fxe5 Ng4 12.Bg5 Qb6 13.0-0 Nxe5 14.d6 Qxb2 15.Nd5 Nxf3+ 16.Bxf3 Qd4+

The importance of this check (or Bd4+) to drive the king to h1 will become more apparent later. 17.Kh1 Qxa1 18.Qxa1 Bxa1 19.Rxa1 Nd7 20.Be7 Against John Langreck in the 1995 Indiana State Class Championship I had this position (but with Kg1) and went for Ne7+ Kg7 Nxc8 Raxc8 Bxb7. Although I won that game, I think Black has a slight advantage. I knew the idea of the bind with Be7 and Nc7, so I went for that here. That was the extent of my theoretical knowledge, but I think Nick's preparation may have gone deeper. 20...Rb8 21.Nc7 Rf8 22.Re1 b5 23.Bc6 b4 24.Bxf8 Kxf8 25.Re8+ Kg7 We are still within the realm of known theory. I now looked at 25.Bxd7 b3 and couldn't see a good way to maintain the extra piece. 26.axb3 Rxb3 and White's weak back rank (from Qd4+) lets Black play Bxd7 with a good ending. 26.Bb5 bxa2 27.Re1 a6 28.Bd3 Bf5 seems to win for Black.

26.Rxc8!? A theoretical novelty, but one that is unlikely to overturn the assessment of this variation. The book move is 26.h3 leading to a drawn ending after further complicated play. 26...Rxc8 27.Bxd7 Rd8 Black could do a little better by speeding up his queenside action with 27...c4 28.Bb5 Kf6 29.d7 Ke5 30.Kg1 Kd4 31.Kf2 c4 32.Ke2 Kc5 33.Ba4 a5 I had expected a draw after 33... b3 34.axb3 cxb3 35.Kd2 Kb4 36.Bc6 Kc5 34.Kd2 h6 I'm not really sure what the point of this move is, Black seems to be drifting a little. 35.Ne8 Kd5 Here I expected 35...h5 not letting the knight get out of enemy territory 36.Nf6+ Ke6 37.Ng4 h5 38.Ne3 b3 39.axb3 cxb3 40.Kc3 Better is 40.Nc4 so that if Black plays like the game, White will be able to win both queenside pawns before Black makes much headway on the kingside In that case White would have excellent winning chances. 40...Rxd7 41.Bxd7+ Kxd7

A difficult ending. I suspected White could win with extremely accurate play but I haven't been able to find it in analysis. Black's king is going to become active while White is chasing down queenside pawns, and Black only needs to trade off White's two remaining pawns. 42.Nc4 a4 43.Nb6+ Ke6 44.Nxa4 Ke5 45.Kd3 White can pick off the pawn with 45.Nc5 b2 46.Nd3+ Ke4 47.Nxb2 Ke3 48.Nd3 Ke2 but a win doesn't seem to be there 45...Kf4 46.Ke2 Kg4 47.Kf2 h4 48.h3+ [48.Nb2 h3 49.g3 f5 50.Ke2 g5 51.Ke3 f4+ 52.gxf4 gxf4+ 53.Kf2 Kg5 54.Kf3 Kf5 55.Nd3 Kg5 56.Ke4 Kg4 57.Nf2+ Kg5 58.Nd1 Kg4] and I don't see how I bump him off of defense of his pawn 48...Kf5 49.Ke3 g5 50.Kf3 g4+ 51.hxg4+ Kg5 52.Nb2 f5 53.gxf5 Kxf5 54.Nd3 Kg5 55.Nb2 I had missed that my intended 55.Ke4 is simply met by 55...Kg4 55...Kf5 56.Nc4 Kg5 57.Ke2 Kg4 58.Kf2 Kf4 59.Nb2 Kg4 60.Nd1 Kf4 61.Nb2 Kg4 62.Nd3 b2 63.Nxb2 h3 64.gxh3+ Kxh3 1/2:1/2

12/16/00 - Bereolos-Dickerson 2000 Knoxville City Championship

1.d4 d6 The other times I have had White against Leonard (only twice in ten games!) he went for the Slav. 2.c4 e5 3.Nc3 exd4 4.Qxd4 Nc6 5.Qd1 More usual is 5.Qd2, but the text move doesn't commit White to the fianchetto of the queen's bishop. 5...Be6 6.Nd5 A somewhat dubious leap. White should just protect his pawn with b3, e3, or e4. 6...Ne5 7.b3 c6 8.Nc3 I wasn't especially comfortable with the position after 8.Nf4 Bf5, so back to where he came from. At least I've provoked some weakness in his d-pawn. Still, there can be no doubt that Black has already equalized. 8...d5 9.cxd5 cxd5 10.e3 Bb4 11.Bb5+ Nc6 12.Ne2 Qa5 Active play, but maybe not the best spot for the queen. Black should develop his king's knight with equality. 13.Bd2 a6 This move isn't really useful since White would make the trade Bxc6 without being provoked. He should have continued developing with 13...Nge7 or 13...Nf6 14.Bxc6+ bxc6 15.Na4 After the forced exchange of bishops, Black is weak on the dark squares and White's knight is much superior to Black's bishop. 15...Nf6 16.f3 White will be forced to make this move at some point because of ...Ne4, but it may not be necessary immediately. It would be great if White could do without it since it creates a small weakness on e3 16...0-0 17.Bxb4 Qxb4+ 18.Qd2 Qh4+ The endgame without queens would be unpleasant for Black. White can occupy c5 and d4 then go to work on the pawns. In the game, White will follow a similar plan, but Black has some counterplay with queens on the board. 19.g3 Qh5 20.0-0 Nd7 21.Rac1 Rac8 22.Nf4 Qh6 23.Nc5 Nxc5 24.Rxc5 Bd7 25.Rfc1 Qd6 Allowing a tactical shot, which also may have been possible on the previous move. 25...Rfd8 was probably a bit better.

26.Qd4 A natural move, keeping with the dark square strategy, but White may have been able to steal a pawn. with 26.e4 Be6 27.exd5 cxd5 28.Rxc8 Rxc8 29.Rxc8+ Bxc8 30.Qxd5 Qxd5 31.Nxd5 Bb7 32.Nc7 a5 33.f4 26...Qe7 27.Ra5 Rfe8 28.Kf2 Ra8 29.Nd3 Qd8 30.Rac5 Re6 31.Rd1 Finally ready to get on with the idea e4, but Black is a bit better placed now. 31...Rh6 32.Kg1 Be8 33.Ne5 Qg5 34.e4 I thought I was just about winning here, but he spent a good chunk of his remaining time to find a way to keep material equality. 34...Qh5 35.Ng4 Rd6 36.exd5 Rad8 Black's point, taking advantage of the pin. I caught most of the way back up on time here, but didn't find a way to take advantage of the free move. 37.Re1 Rxd5 38.Rxd5 Qxd5 39.Qxd5 cxd5 40.Kf2 f6 41.Ke3 Bh5 I was more worried about 41...d4+ not letting me get my king to the great d4 square. 42.Kd4 Bxg4 This exchange can wait since White is not threatening to move the knight immediately. There is even some question that Black should swap at all since now both of White's pieces will be better than their Black counterparts. 43.fxg4 Kf7 44.Rc1 Rd6 45.Rc7+ Kg6 46.Rc5 Kg5?

47.h3? We were both down under 5 minutes at this point. I saw the simple 47.h3 was a clean pawn, then started looking at the pawn ending. 47.Rxd5+ Rxd5+ 48.Kxd5 Kxg4 49.b4 Kh3 50.a4 Kxh2 51.b5 axb5 52.axb5 Kxg3 53.b6 wins. So what else is there? 49... f5 50.a4 g5 51.b5 axb5 52.axb5 f4 and I stopped here looking only at counting when we both queen at the same time. With the clock starting to get low I went back for the safe move. If I had looked a bit further I would have seen 53.gxf4 gxf4 54.Ke4 f3 55.Ke3 and my pawn on h2 stops him from escorting his pawn. 47...Kg6 48.Rxd5 Re6 49.Ra5 Rd6+ 50.Ke3 Re6+ 51.Kf4 Rb6 52.Ra4 Kf7 53.Re4 a5 54.Ke3 Kg6 55.Kd3 Kg5 56.Kc3 g6 57.a3 h5 58.gxh5 gxh5 It's hard to criticize the remaining moves, played under extreme time pressure. 58...Kxh5 might be better and maybe he can generate counterplay with a passed f-pawn eventually, but it is still doubtful he can hold since White is coming very quickly on the queenside with his b-pawn. 59.b4 axb4+ 60.axb4 f5 61.Rd4 Rb5 This let's White's king advance with tempo. 62.Kc4 Re5 63.b5 Passed pawns must be pushed, but 63.h4+ to drive the Black king back also deserves attention. 63...Re3 64.Rd3 Re8 65.b6 h4 66.gxh4+ Kxh4 67.Rf3 Kg5 68.Kc5 Re1 69.Kc6 He lost on time here, but the position is an easy win e.g., [69... Rc1+ 70.Kb7 f4 71.Ka7 Kf5 72.b7 Ra1+ 73.Kb8 Rb1 74.Ra3 Ke4 75.Ka7 f3 76.b8=Q Rxb8 77.Kxb8 f2 78.Ra1 Kf3 79.h4 Kg4 80.Rf1] 1:0

New In Chess - Your Link to the Chess World

12/15/00 - Vaughan-Bereolos 2000 Knoxville City Championship

1.d4 Nf6 2.Nf3 g6 3.Bg5 Bg7 4.e3 Dickerson-Barber, also in the first round, saw 4.Nbd2, which gives up the chance for the knight to go to c3, but White may be able to build a sturdy center with e4 and c3. 4...0-0 5.Be2 d6 6.0-0 Nbd7 7.c4 e5 8.Nc3 h6 9.Bh4 c6 More common is the immediate 9...g5 with which Judit Polgar won a nice attacking game vs. Ruban in the 1993 PCA qualifier. 10.Rc1 Qe7 11.Bg3 voluntarily retreating the bishop does not feel right 11...Nh5 12.Qc2 Nxg3 13.hxg3 f5 14.Rfd1 e4 15.Nd2 Nf6 I was very pleased with my position here. My plan was to overprotect e4 with Nf6, Bd7 and Rae8 followed by a pawn attack with g5 and f4. It seems to me that White's queenside play will be very slow in comparison. 16.g4 I certainly didn't expect any kingside action with all his pieces bunched on the queenside. After the game Eric suggested 15...h5 by Black to stop this move. Black can then follow the plan suggested above without distraction. Still, Black must be better after the text. 16...Bd7 Taking on g4 also deserves serious consideration opening the f-file and increasing the scope of Bc8. I wanted to maintain the cramping effect of the pawn on e4 instead. 17.gxf5 gxf5 18.Bf1 White wants to take advantage of the weakened f4 square with Nc3-e2-f4. 18...Ng4 With the crude threat 19...Qh4 19.g3 Qg5 Now sacrifices on e3, f2, or d4 are in the air as well as the possibility of ...Qh5 20.Bg2

I spent some time looking at 20... Nxe3 21.fxe3 Qxe3+ 22.Kh2 Bxd4, but couldn't bring myself to pull the trigger. Black certainly has compensation in the form of 3 pawns, a loose White king, and the jumble of White's pieces on the queenside. On the other hand, it will be hard to get to the King quickly and Black's king is a little airy as well. I decided to go for a slower buildup without sacrificing 20...h5 21.Nf1 h4 22.Ne2 hxg3 23.Nfxg3? 23.Nexg3 had to be played, now the knight sacrifice is devastating. 23...Nxe3 24.fxe3 Declining with 24.Qd2 does not help after 24...f4 Qxe3+ 25.Kf1 [25.Kh2 Kf7 26.Rf1 Rh8+ 27.Bh3 Rxh3+ 28.Kxh3 f4+ 29.Kg2 f3+ 30.Kh2 Rh8+] 25...f4 26.Nxe4 f3 27.N2g3 Bg4 I wanted more than just the exchange which could be had by 27... Bxd4 28.Rxd4 fxg2+ 29.Kxg2 Qxd4 28.Rd3 28.Qd2 fxg2+ 29.Kxg2 Qxd2+ 30.Rxd2 Bh6 and Black gets the exchange after all 28...fxg2+ 29.Kxg2 Bf3+ 30.Kh2 Qh6+ 31.Kg1 Qg6

A nice geometric move lining up against all the White pieces on the g-file and h7-b1 diagonal. 32.Re1 Rae8 33.Rxf3 Rxf3 34.Nf6+ Rxf6 35.Qxg6 Rxe1+ 0:1 All the other games in this round were drawn, so I found myself as the clear leader after only one round.

12/14/00 - 2000 Knoxville City Championship

The latest edition of the Knoxville City Championship concluded last night, and I'm happy to report that I capped a successful year by three-peating as Champion. My 4th title overall moves me out of a tie with Tom Rowan and leaves me trailing only 6-time champion Charles Maddigan. Maybe this will propel Charles out of retirement.

The event was stronger than last year with the addition of former co-champion Nick Barber, who was taking the semester off from college, and US Postal Champion David Burris, who joined our club this year. After a year off, the championship field once again had an average rating over 2000. For the fifth straight year, I managed to draw a number in the bottom half, meaning 3 blacks and 2 whites. However, this year I got a small break that my whites were against the #2 and #3 seeds, Leonard Dickerson and Barber. The same 4 wins and 1 draw score that had one for me in the past worked here as I drew with runner-up Barber, who was nicked for a draw by Dickerson. So a perfect score remained out of my grasp (I don't know if anyone has ever had a perfect score in our championship).

For the first time in a couple of years, we managed to get all 15 games played. As has been my custom every year, I will annotate my own games for inclusion in the tournament bulletin (and probably they'll end up in the TCA magazine, too). I'll be posting them here, one at a time, starting tomorrow.

12/12/00 - Final 4

The FIDE championship is down to the final 4 and finally 4 game matches! We've got 3 superheavyweights (Anand, Adams, and Shirov) and 1 tourist (Grishchuk). Anand finally outlasted Khalifman to eliminate the defending champion. The decisive game was interesting from a theoretical point of view. In the Slav Defense, Khalifman played the well know sacrifice of a piece for 3 pawns. Common knowledge seemed to indicate that White should not play the queen exchange, Qxd5 allowing Black two connected passed pawns in the center. Anand indeed played this and, without doing anything extraordinary, held up the pawns and penetrated the queenside to win in relatively easy fashion. Adams got by Topalov without much problem. Grishcuk won a tiebreaker when Tkachiev found his king overexposed in a major piece ending. Shirov teetered on the brink by losing the first game of regulation, but roared back to win the second game and finished him off in the tiebreaker. The desicive game was a typically murky Shirov affair where he sacrificed a piece for unclear compensation. However, Bareev blinked and declined the sacrifice to exchange queens instead. The lack of queens didn't stop Shirov's initiative and Bareev soon gave up.

So, two of my picks made it to the final four (Adams and Shirov). I think I also did all right in picking Grishchuk's bracket as the one that was likely to produce a suprise. Adams vs. Anand should be a classic. They battled to the death two years ago in the finals going deep into the tiebreaks. Since then, Adams has slowly but surely moved up the world rankings to his current #4 spot. Anand has had some uncertain periods, but seems to have turned it on late in the year and has still maintained the #3 spot on the rating list. I think it will again go deep into the tiebreaks, but I think MICKEY ADAMS will continue his climb and eliminate Anand and disappoint the New Dehli crowd. On the other side ALEXEY SHIROV is a man on a mission, playing seek-and-destroy chess. Young Grischuk has had a great run, but it's time for the big boys now.

12/7/00 - Another One Bites the Dust

Another one of my picks for the final four was sent packing in the FIDE championship today. #2 seed Alexander Morozevich was eliminated by Vladislav Tkachiev. In other matches, #1 seed Vishy Anand continued to cruise along with the simple formula draw with Black, win with White. In the quarters, Anand will face defending champion Alexander Khalifman. I was suprised to see that prior to this round statistician Jeff Sonas had now installed Khalifman as the favorite to win the whole tournament! To me this indicated that he would also be favored in the head-to-head matchup with Anand. This didn't make sense to me for several reasons: 1. Anand has a higher rating in classical chess 2. Anand has a higher rating in rapid chess. 3. Anand has a good record vs. Khalifman (he's one of only 3 players to beat Khalifman this year, but one of the others was Peter Leko. Veselin Topalov is the 3rd.). Anand eliminated Khalifman in the first FIDE KO event in 1998. In fact, I cannot find a single game where Khalifman has ever beaten Anand (maybe someone with a better database can find one) 4. Anand is brutal vs. the French Defense, one of Khalifman's main defenses (2 of Anand's 3 wins against Khalifman this year were in the French Defense and chesslab has Anand at +9 =1 -1 against the French this year). Therefore, Anand should be a huge favorite in their head-to-head matchup. The only other way I could see for Khalifman to be most likely to win the tournament would be if there was some player in the rest of the field who gave Anand fits, but is a regular customer of Khalifman's (and still I wouldn't think this could overcome what I see as a huge Anand advantage head-to-head). I couldn't identify such a player (and doubt if one exists). So, I emailed Jeff Sonas to try to figure out what had gone so horribly wrong in his model 8-). He replied that his model makes Khalifman a 53 to 47 favorite in the match vs. Anand and promised details in his 4th round preview. I await with great interest! Speaking of Topalov, he won today to even his match with Bareev, so they'll be one of 5 tiebreakers tomorrow.

12/5/00 - Sweet Sixteen

Two more rounds have passed in the FIDE KO event, and we are down to sixteen players. Most of the top guns have survived, but #6 seed Vassily Ivanchuk was eliminated in the second round by Jaan Ehlvest. Although the top names have advanced there are still plenty of "tourists" hanging around. Gary Kasparov predicted that none of the players who played in the first round would survive to the quarterfinals, but he will be wrong since Ehlvest and young Russian GM Alexander Grishchuk will face off for one of the spots.

One of my picks for the final 4 was eliminated: #5 seed Peter Leko. However, many may not consider this an upset since he was bested by the defending champ, Alexander Khalifman in a real grudge match ( stats guru Jeff Sonas has Khalifman as the #2 favorite to win the event behind Anand, I haven't seen any bookmakers odds). If you recall in January, Leko dominated the Khalifman, in their match, which was the first appearance since El Khalif took the title in Vegas. Khalifman got his revenge the next month at Linares, beating Leko in a theoretical Grunfeld. Last month at the Olympiad, Leko won their individual encounter, but Khalfiman and his Russian teammates took gold. This match didn't disappoint. After two draws in the regulation games and 5 in the rapid tiebreakers, Khalifman finally broke through against Leko's Petroff Defense in the 8th game of the match.

Elsewhere, my 3 other picks for the final 4 (Shirov, Morozevich, and Adams) all made it through, but two of my longshots (van Wely and Serper) dropped out. Adams eliminated Yermolinsky, so Boris Gulko is the only one left to carry the US banner (Joel Benjamin lost in Round 2). Shirov seems to be playing the most interesting chess (as usual) seeming to be trying to sacrifice a rook at every opportunity from what I have seen so far. I thought he took this to an extreme at the end of Round 2 against Onischuk.

The players had just played 51...Ra1-b1+ 52. Kb5-c5 Ra1 53. Kb5 Rb1+ 54. Kc5 to reach this position. Shirov only needed a draw to advance, so the expected move would be 54...Ra1 to repeat the position. Nevertheless, the gamescore reads 54...Ke5!? 1/2-1/2 . I've waited a couple of days to see if this got corrected, but no change has been forthcoming. Black is obliged to sacrifice his rook for the White pawn after 55. a7. It is probably still a draw because of Black's well placed King, but certainly if this occured, I don't know why Onischuk would not continue playing. I'll keep my eye peeled for the solution to this mystery.

Top seed Vishy Anand has been cruising winning all of his matches in regulation so far. He'll face the real Cinderella #78 seed Bartlomiej Macieja. For all the griping about how the pairing system makes it hard on the top seed, Anand looks to have an easy bracket with #21 Khalifman and #66 Leitao on the other side of his draw. Compare this to the next bracket with #3 Adams, #7 Topalov, #11 Svidler, and #15 Dreev. I'd like to call attention to Svidler here. He bounced back against Peng in Round 3 after losing the first game to even the score and then win in the tiebreaks. This bodes well for him since I think being able to cope with the stress of facing elimination is crucial in this type of event (Remember in Las Vegas, Khalifman went to the brink in his first two matches with Barua and Kamsky). Not having to yet play a tiebreaker may handicap Anand later in the tournament when he'll have to adjust from playing classical games (In Vegas, Ivanchuk was cruising along, then self-destructed in the rapids vs. Nisipeanu)

Finally, the Alex watch: 5 remain (Khalifman, Shirov, Morozevich, Grischuk, and Dreev). They are distributed such that we could see an all-Alex final 4.