Round 8 saw a clash of veterans with Jan Timman of the Netherlands taking on the grand old man, Viktor Korchnoi. Timman played a bit too actively in a slightly worse ending and dropped a pawn. He then sought refuge in a pawn down rook ending after 42.fxe3
At first glance, it appears Black has a significant advantage. His extra pawn is already passed and White's King is cut off from the kingside. White's only plus seems to be that his rook is actively placed behind the passed pawn. 42...Rd7 better than 42...Rc5+ when the Black rook is poorly placed in front of the passer and the White King can get back in the game with 43.Kd2. Also, 42...c5 would be met by 43.e4 when 43...Rd4 44.Rxc5 Rxe4 is similar to the game. 43.e4 I think this is an important move denying Black access to the f5 and d5 squares. Timman opined after the game that he thought he had no further problems after this move. It would be interesting to see if White could do without this move via 43.Kc2 e4 to see if this affords Black any further advantage. I looked at it briefly, but still did not see how Black could take advantage of this. However, it may be that in these endings access to the e5 square for Black's king may also become important. 43...Kf6 Korchnoi anticipates swapping his c-pawn for White's e-pawn and does not want Rxc7 to come with check. 44.Kc2 Rd4 Of course Black would like to avoid pawn exchanges and also not let White's king into the game, but I don't see a way for him to activate his King without losing a kingside pawn. 44...Ke6 45.Rh8 g6 (the sacrifice 45...Kd6 46.Rxh5 doesn't seem to get anywhere since the e-pawn is under attack) 46.Rg8 Kf7 47.Rc8 now 47...Ke6 48.Rg8 is a repetition, 47...Rf7 48.Re8 cuts off the Black King and 47...Re7 48.Rf8+ Ke6 (48...Rf7 49.Re8) 49.Rg8 forces the King back. 45.Rxc7 Rxe4 Again, a typical rook ending with balanced pawns on one side and a passed pawn for one player. Black's problems in converting this ending come from several factors: 1) the passed pawn is an e-pawn. A b-pawn would be preferable since as a much more outside passed pawn it could better serve as a decoy. Here, both the White King and Rook can actively participate in stopping the passed pawn. 2) There are only 2 pawns left for each side on the kingside. An ending with 3 pawns each and a passed d-pawn would give Black much better winning chances. Here, if Black parts with the e-pawn to win one (or even both) of White's pawns, technical difficulties could still arise in coverting 2 vs. 1. 3) Black's rook is misplaced in front of his pawn. 46.Kd2 g6 Freeing the Black King form having to guard the pawn so he can now move forward. However, I like the idea of first improving the position of the rook with 46...Ra4 47.Ke3 Ra2 48.h3 Ra3+ 49.Kf2 47.h3 stopping an anticipated invasion by K-f5-g4. This move probably would have to be played sooner or later, but it creates some weakness in the White pawn chain. 47...g5 48.Rh7 Kg6 or 48...h4. This possibility will not disappear so Korchnoi applies the "Don't Rush" principle. 49.Re7 Kf5 50.Rf7+ Ke6 51.Rg7 Kf6 52.Rh7 h4 53.g4 The defending side in a pawn-down ending is always looking to trade pawns, but here 53.gxh4? Rxh4 would cost White another pawn and the game. It now looks like Korchnoi has made some progress since White has a fixed weakness on h3. However, this is balanced by the weakness of Black's g5 pawn. 53...Ra4 54.Rh6+ Kf7 55.Ke3 Ra3+ 56.Ke4 Rxh3 57. Ra6 A great defensive move by Timman. Not 57.Kxe5 Rf3 followed by R-f4xg4 -+. Timman also avoids 57.Kf5 Rf3+ 58.Kxg5 h3 which looks decisive. Timman's move reaches a similar position, but one where his rook defends from the side rather than behind. A great example of breaking the principle that rooks belong behind passed pawns since a rook can't be behind two passed pawns at once. 57...Rg3 58.Kf5 Rf3+ To cut off White's King from the e-pawn. 58...h3 59.Kxg5 and the rook can return to h6 to stop the h-pawn while the King crosses the f-file to take care of the e-pawn. 59.Kxg5 h3 60.Ra2 e4 61.Re2 e3 62.Kh4 Timman's idea is now clear. The Black rook is in a awkward position. White is ready for g5 and Kg4 forcing the Black rook to move and give up a pawn. 62...Kg6 63.g5 Kf5 taking away g4 from the White King. 64.g6! A diversionary pawn sacrifice to reclaim g4. It was not too late to go wrong with 64.Kh5? h2 65.Rxh2 e2 66.Rh1 e1Q 67.Rxe1 Rh3# 64...Kxg6 acquiesing to the draw. The last try was 64...Kf4. Then, not 65.g7? Rg3 66.Rxe3 Rxg7 threatening both Kxe3 and Rh7# but 65.Re1 and the game is drawn 65...Rg3 66.Rf1+ Rf3 67.Re1 = 65. Kg4 1/2-1/2
No rook endings appeared in Rounds 5 or 6, but the big showdown between Kasparov and Anand in Round 7 produced a somewhat instructive ending. The following position arose after Anand's 38...Rxc4
Kasparov is a pawn ahead, but it is a passed rook pawn with even pawns on the kingside, a typical type of pawn-up ending that is difficult to win. Kasparov could maintain his pawn advantage with 39.Rxe6 Rxa4 when an ending with 4 vs. 3 on the same side of the board occurs. More discussion about this type of ending follows Black's 40th move. Kasparov chose instead 40.a5 following the principle "passed pawns must be pushed". 40...Ra4 Anand follows the principle "rooks belong behind passed pawns". He could have restored material equality with 40...Rxe4, but after 41.a6 e5 42.a7 Ra4 White has a large advantage. The simplest plan is to march the King to b7 or b8 and queen the pawn. Some care must be taken since Black can try to create counterplay with a passed e-pawn by ...f5, Kg6, Kf6, g5, e4, f4, e3 etc. The position after 42...Ra4 is a good one to practice your technique versus your computer or a training partner. White should be able to win. 40.Rxe6 not 40.Ra7?! Rxe4 when the position of the white rook, blocking the advance of his pawn, is much worse than in the previous example. Black would have no problem holding the draw. Kasparov points out another posibility 40.e5 fxe5 41.Ra7 when Black's pawn structure is crippled. However, it is hard to see how White takes advantage of this after 41...Kg6 and 42..Kf6 or Kasparov's suggestion 41...e4 limiting White's King 42.a6 Ra2 threatening e3 43.Kf1 Kg6 now after 44.Ra8 Kf6 45.a7 Ke5 46.Ke1 e3 47 fxe3 Ke4 Black's active King will ensure the draw. 40...Rxa4 reaching the theoretically drawn ending of 4 pawns versus 3 on the same side of the board. Capablanca won this ending twice, so there is some practical hope. The defensive player is to exchange 3 sets of pawns reaching a drawn R+P vs. R ending. The ideal formation for this is the pawn structure f7-g6-h5 so that inorder to create a passed pawn White will have to play h3 and g4 allowing the h-pawns to be traded, followed by f5 which takes care of the g-pawns and finally e6 taking care of the f-pawns. The offensive player's hopes lie in creating a passed e-pawn with only one set of pawn exchanges (as could be the case here with Black's pawn on f6 instead of f7) or exchanging his h-pawn for Black's g-pawn (this was Capablanca's technique, but the key defensive move ...h5 stifles this plan). 41.Rd6 if Black now play's his standard defensive move 41...h5, White wins with 42.Rd5 41...Ra4 removing the rook from the fifth rank with a gain of tempo. 42.Kf3 Ra3+ 43.Ke2 White gets nowhere with 43.Kg4 Ra2 43...h5 finally the characteristic defensive move. 44.Rd3 Ra2+ 45.Ke3 Kg6 46.h3 Ra4 47.f4 Rb4 48.Ra3 White could create a passed pawn with 48.e5 but after 48...fxe5 49.fxe5 Kf5 50.Rd5 Ke6 the White passed pawn is firmly blockaded and the White King is cut off by the Black rook, so no progress can be made. 48...Rc4 49.g4 It isn't clear to me why Kasparov played this move so soon. There doesn't seem to be any rush and every pawn exchange eases the defender's burden. Possibly he was a bit fed up with himself for not converting the large advantage he held before the queen's were exchanged. Anand is not the type of player who would slip up in this ending, so Garry may just have decided to save some energy by not trying to torture him.49...hxg4 50.hxg4 Rb4 51.Ra6 Kf7 52.Ra7+ 52.f5 Rb7 53.Kf4 g6= 52...Kg6 53.f5+ Kh6 54. g5+ fxg5 White gets his passed pawn, but it cost him a pawn to do it. 55.e5 g4 56.e6 Kg5 1/2-1/2 After 57.Rxg7+ Kxf5 58.e7 Rb8 59.Rf7+ Ke6 60.Rf8 Rb3+ or 57.e7 Ra8 58.Rd7 Re8 the draw is clear. A model defensive show by Anand. The idea of ...h5 in such endings is an important one to know.
Sincere apologies for the gap between postings. I got stranded out of town by the weather and have been trying to catch back up. I should have time to look at the Corus games tonight and get some more rook endings up tomorrow. I see my Fantasy Chess standing has slipped to number 3. At the hotel in Tampa where I was staying they had a small coin and stamp show that I wandered into one afternoon. One of the stamp dealers had a selection of chess related stamps, envelopes, and post cards. For any collectors who are interested prices, stuff like a Hungarian envelope with Portisch on it was ~$5, Karpov-Korchnoi Merano items were ~$10, and Fischer-Spassky Rejkavik items were ~$15.
No real suprises yet from WAZ. After 4 rounds, Kasparov is tied for first in the tournament and Bereolos is tied for first in Fantasy Chess . Migstradomus and Caissandra better keep their day jobs. There were 2 more rook endings in rounds 3 and 4, but as we will see, not much of interest to analyze in them.
In round 3, Nikolic was white against Lputian. The following position was reached after 35...Rxc7
Nikolic restored material equality with 36.Rxa5 Double rook endings can sometimes be more tactical than single rook endings, but the material is so limited here that a draw is the obvious result. 36...Rcb7 37.Ra2 Guarding the pawn, although the ending would still be drawn if White didn't do this. 37...Rb3 38.Kg2 R8b6 Threatening 39...Rg6. Losing the g-pawn would cause White some trouble. 39.Rf2 Kh7 39...Rg6 40.Rf3 =. 40.Ra3 White has no way to improve his position so he offers a pawn to reach a dead draw. 1/2-1/2 Black can win a pawn with 40...Rxb2 41.Rxb2 Rxb2+, but the single rook ending with 2 pawns against 1 offers no chances without a passed pawn. The best Black could achieve is to make a passed pawn by ...h4 at some point, but after a set of pawns is exchanged, White has no trouble holding as his king is well-placed blocking the pawn.
In round 4, the number 2 and 3 seeds Anand and Kramnik squared off. They reached a R+P ending after 33...Kxd8.
This is a bit more unbalanced of a position than we have seen. Both sides have passed pawns, and Black's rook is a bit better placed. However, neither side has any real chances. Anand played 34.Kc1 activating the King towards the center and avoiding any back rank problems. Black would like to try and tie White's rook down by 34...Rg3, but this could be met by 35. Re1 with a counter attack on the Black e-pawn. 34...Re2 could be met with the simplifying 35.c3 so Black gets his King out of any potential pin on the d-file and brings it up to support his pawns 34...Ke7. If White defends passively now Black can improve his position with Ke7-d5 and Rg3, then e5-e4-e3 and maybe then he is getting somewhere. Anand nips any such thoughts in the bud with 35.Rd3 preparing c3 to exchange off a couple of sets of pawns. Black dare not exchange rooks because the pawn ending is hopeless on account of White's outside passed g-pawn. 35...Re2 36.Rd2 Re1+ 37.Rd1 Re2 38.Rd2 Re4 the only way to avoid the repetition, but after 39.Rd3 White is again ready for c3. 39...Ke6 40.c3 1/2-1/2 The position is balanced after 40...bxc3 41.bxc3 dxc3 42.Rxc3 and the material is so limited that neither side can hope to win.
So far my featured topic of rook endings has not yielded much in this tournament. However, there are still plenty of games to go, so maybe a gem will appear. Or maybe all rook endings really are drawn.
Congratulations to Russell Linnemann, who's performance in Clarksville finally propelled his rating over the 2000 mark. Despite being the lowest rated player in the open section, Russell only lost to IM Tim Taylor in a tough game and his 3.5 points was good enough to tie for third with yours truly. This was also good enough for the expert prize. Russell also showed that you can make money at chess since after entry fee, hotel, and state chess dues, he took home two shiny new quarters.
Russell collected his first career master scalp outside of G/30 against Todd Andrews. After outplaying Todd in the opening and middlegame, Russell began to slip in the endgame when time pressure began to take it's toll. The following position was reached with Black (Andrews) to play, and Russell's flag hanging.
Here, Todd played 1...Rg8+ unpinning his rook, so that after 2.Kf3 he could play 2...dxc3?? to which Russell replied 3.Rd6! and Black is helpless against R1d5# 1-0 An amazing helpmate in the middle of the board. Todd is going to have to be careful that this site doesn't become the Todd Andrews blunder site (see 11/21/99).
The Corus supertournament has begun at Wijk aan Zee. Korchnoi-Kasparov and Polgar-van Wely in round 1 were both interesting R+minor piece endings that are well worth studying. The only pure rook ending in the first two rounds was the first round encounter between Timman and Anand. The following position was reached after 41.Rxc6
This is about as dead a position you can get in a rook ending. Neither side has pawn weaknesses and the pawn structure is completely balaced. If Black made a waiting move like 41...Kg7 White might be able to claim a tiny advantage after 42.Rc7 Ra8 43.Rb7 a6 44.Rb6 when Black has a passive rook. Anand didn't even allow this with 41...Re6 ready to go to a6 if White retreats. Timman exchanged rooks with 42.Rxe6 and the draw was agreed. 1/2-1/2. Black's slightly fractured pawn structure after 42...fxe6 is inconsequential. The extra tempo Black has available with a7-a6 ensures that White can't get too frisky. A possible continuation that shows that there isn't much going on is 43.Kg2 e5 44.Kf3 Kf7 45.Ke4 Ke6 46.f3 Kd6 47.g4 Ke6 48.gxh5 gxh5 49.f4 exf4 50.Kxf4 Kf6=
I went to the first North Tennessee Regional Open this past weekend in Clarksville, TN. I think this tournament may have been created to fill the void created by the cancellation of the Fairfield Glade tournament last year. Unfortunately, this new tournament comes up very short in comparisons with Fairfield Glade. Where Fairfield Glade had large guaranteed prizes, this tournament had an overly optimistic basis for it's prize structure. The prizes were based on 30 players per section, and none of the 6 sections came close to that. Consequently, almost all prizes were reduced by half. I know better than to expect to make alot of money playing chess, but a $150 first prize for a $60 entry fee is severely unbalanced. Throw in $50 a night for a couple of nights of hotel and even the biggest winners couldn't break even. If they hold this tournament again next year, I hope they make serious changes to the prize structure including guaranteed prizes for the top places. This kind of bait and switch is very frustrating to players who have to travel and is a reason that I rarely play at the Atlanta Chess Center (which if they ever got close to the numbers they base their prizes on, they would probably violate the fire code).
The other major problem with the tournament was the pairings. There were two major problems and one minor problem. The first problem came in the second round. Russell Linnemann received a bye in the first round, which somehow got entered into the computer as a zero point bye. When this was pointed out to the TD, Harry Sabine, about 5-10 minutes into the round, at which point none of the affected games had begun , Harry ruled that the round had begun so the pairings must stand. This has got to be one of the worst decisions I have ever seen a TD make, especially in light of what happened later. The next hiccup came in round 4, when Linnemann, with 3 points, was paired with IM Tim Taylor, who had 2.5. Linnemann's colors had been bye-B-W, while Taylor had played W-B-W. Now here I think the rulebook would rule that the colors could be paired either way, but I think the prefered pairing would be to balance Taylor's colors and give Linnemann 2 whites in a row. I think most human TD's would pair it this way as well. However, the computer (Pair Plus program) spit out Taylor as white, since Linnemann had the higher score, he got his due Black. I was somewhat the benficiary of this since in the last round I got white against Taylor. This shouldn't seem odd since one wouldn't expect Taylor to get 3 whites in a row. However, for the first time in my tournament history I received 3 whites in a row and 4 whites in 5 games! The final pairing disaster also occurred in the last round. There were 6 players left for the bottom 3 boards have scores of 2.5,2,2,2,1,and 0.5. Normally if players hadn't played each other the pairings would go 2.5-2, 2-2, and 1-0.5. Indeed, these pairings would work. However, since the first 2.5-2 would have given the player with 2 Black in consecutive games (who cares when you're giving a guy on Board 1 three whites in a row) the computer made some switches and wound up with 2.5-2, 2-0.5, and 2-1. This time the players convinced Sabine to make the switch even though the game 2.5-2 had begun and was far enough along that White (Doug Eckert) could rightfully claim an advantageous position. I think the decision to switch to the obvious pairings was correct, especially since it could have major impact on the prizes, but why wasn't this done in Round 2 as well? Eckert probably has the most right to be upset with what happen since he lost badly to Doug Hyatt after the pairings were switched. I'm also a little disturbed by the heavy reliance on computerized pairings, especially in small to medium sized tournaments. I realize it makes reporting of the results much more efficient, but the TD's need to at least glance at the pairings to make sure they pass the ho-ho test. Also, how good is the quality control on these programs? Does USCF have some sort of certification process?
The chess portion of things went better. I lost the thread from a better position against Gerald Larson in round 1 and lost. I rebounded with 3 wins and drew with Taylor in the last round to take 3rd. My travelling partners, Linnemann and Hyatt, also finished in the money, so it was a pleasant trip home. Despite a small section, numerous withdrawals, and funky pairings, we managed to avoid playing each other. I'll give some analysis later in the week. Also, as part of the series on rook endings, I'm going to try to discuss all the rook endings that come up in the Corus supertournament.
Things turned out even worse for Alexander Khalfiman than I predicted. Peter Leko dominated their match winning 3 times with white and drawing 3 times with Black for a convincing victory. Looks like taking Peter Leko as the "Win With White" player in Fantasy Chess for Wijk aan Zee could be a good choice. Despite Leko's win, and a recent big win by Nigel Short in Spain, I can't pick anyone besides The Boss to rule in Wijk aan Zee. For Khalifman, he still has time to send me an autograph before Linares.
In chess news from England, GM Sam Palatnik came on strong in the Hastings Challengers, but fell half a point short and had to settle for a tie for third place. In the Premier, WIM Irina Krush had a promising start (1.5/2), but is limping towards the finish sitting at -3 after 8 rounds. This type of international debut seems common for young American players as witnessed by similar struggles by Tal Shaked and Gata Kamsky.
As far as Feng-hsiung Hsu's letter claiming that Garry Kasparov turned down a return match with Deep Blue, I only have 2 comments: 1. The challenge seems not to have been to Deep Blue, but some future machine Hsu would build with the Deep Blue chip. 2. I'll reserve further comment until we hear the other side of the story.
Pal Benko points out in this month's Chess Life, "Again we see that two extra pawns do not automatically guarantee a win in rook endgames." Often, this is because the defending side can stir up trouble with a passed pawn. However, even without a passer, the defending side can still put up resistance. Even the best players in the world are not immune from this as can be witnessed by the conclusion of the 9th game of the PCA Championship in 1993. Garry Kasparov achieved the following completely winning position after 45 moves against Nigel Short.
Here, Kasparov went seriously astray with 46.e4? White needed to make a waiting move like 46.Ra2 or 46.Ke2 46...Ke6? Short returns the favor! After 46...Rc5 47.a5 Rc3+ 48.Ke2 Ke4 49.a6 Rc8 or 47.Ra3 Rc4 48.a5 Re4 49.a6 Rf4+ and 50...Rf8 Black is in time to stop the a-pawn. Kasparov made no further errors in the execution: 47.Ke3 Kd6 48.Kd4 Kd7 49.Kc4 Kc6 50.Kb4 Re5 51.Rc1 Kb6 52.Rc4 1-0 With the Black King cut off, White can bring his King back to support the march of the e-pawn.
I've been on both sides of similar misadventures in this type of ending, which I'll share over the next few columns. In the first case, I was Black against Dennis Gogel in the 1996 Rutherford County open. Here is the position after 50...Rxg3+
It seems like Black should be winning here with two extra pawns. However, I found myself with an awkwardly placed rook 51.Ke4 Rg5 52.Ra7 h6 52...h5 53.Ra6+ cuts off the Black King, but this may have been worth a try. Black threatens ...h4 and ...Rh5. If White continues 54. Ra7+ Kh6 (54...Rg7? 55.Rxg7+ =) 55.Ra6+ Rg6 56.Ra5?! Black has the finesse 56...Rg4+ 57.Ke3 (56.Kxe5? Rg5+ -+) Rg5 57.Ke4 (checks don't help since Black can hide at f5 and Black still meets Ke4 on later moves with h4) h4 and White still can't take the e-pawn. Therefore, White must try something else on move 56. 56.Ra7 or Rb7 both seem to fail to 56...h4 when 57.Kxe5 is a theoretical win for Black. However, 56.Ra3 seems to be a tougher nut to crack. Now, 56...h4 57. Kxe5 is a draw, 56...Rg5 57.Ra6+ repeats the position and 56...Re6 allows 57.Kf5 Re8 58.Ra6+ Kg7 59.Kg5 = so maybe it is a draw after all. 53.Rb7 I think 53.Ra6+ Kh5 54.Ra1 is more accurate. If 53...Kh7 54.Ra7+ Black doesn't have h6 available and I don't see how he can make progress. 53...Kh5 53...h5 would transpose to the previous note 54.Rb1 Rg4+ Unable to find a good plan I give up my center pawn hoping that cutting off the White King along the 4th rank might give some winning chances, but it is a theoretical draw. Backing off with 54...Kg6 trying to reach one of the earlier positions could be met by 55.Rb6+ as in the note to White's 53rd move 55.Kxe5 Kg5 56.Rb8 h5 57.Rh8 Ra4 58.Rg8+ Kh4 59.Rg1 Kh3 60.Kf5 h4 61.Rb1 Ra2 62.Kg5 Rg2+ 63.Kf4 Kh2 64.Kf3 h3 65.Rb8 Ra2 66.Rg8 Ra6 67.Rg7 Rf6+ 68.Ke2 I stopped recording here, but White had no trouble holding the draw 1/2:1/2 I would be very interested if any reader can find a win for Black from the diagram. I'll probably take a look at earlier points in this endgame in a later column, because I am sure that I was winning at some point.
I'm still a bit slow getting back up to speed in the new year. I've archived December below and promise fresh analysis tommorrow. For now, I'll just bask in the glow of my first FIDE rating at long last: 2308. This is a bit lower than I expected, apparently they didn't rate my results for the 1997 or 1998 Chicago Opens. Oh, well, at least I'm still rated higher than Karpov, Fischer, Spassky, Petrosian, Aliekhin, Botvinnik, and Kasparova 8-) If only I could get into one of those Burmese tournaments, I'd get the IM title and a 2500 rating no time!
My prediction on the Leko-Khalifman match has held up so far. Leko won game 2 and the other 2 were drawn. I guess it's time to reveal the true reason for Alexander Khalifman's struggles of late. While it is true that he hasn't won a game since becoming FIDE world champion, his winless streak goes back a little farther than that. He has been unable to win a game since blowing off my autograph request in Las Vegas!!! Just goes to show you what happens if you forget about the little people.
Happy New Year! I've selected my top 10 games for 1999. I'll try to post links to them later.
10. A. Ivanov-Akopian, World Open. Black sacs a piece for two pawns in the opening and an endgame blunder by White moves the decimal place on his check two places to the left.
9. Nisipeanu-Khalifman, Las Vegas. The unknown Nisipeanu, with his back against the wall in the final regulation game produces a bishop vs. knight endgame gem in the FIDE Final 4.
8. Topalov-Kasparov, Linares. Garry knows Q+P endings.
7. Kamsky-Khalifman, Las Vegas. After 3 years away from the game, Kamsky returns with a tour de force.
6. Khalifman-Kamsky, Las Vegas. Facing elimination, Khalifman sacs an exchange and a pawn in the opening and gives black an uassailable position for his knight. But space and the bishop pair prove too much. A huge turning point in the FIDE World Championship.
5. Akopian-Khalifman, Las Vegas. A magical rook endgame by Akopian.
4. Anand-Kasparov, Linares. A titanic struggle between two giants. Facing deep home preparation, Anand finds a piece sacrifice to steer the game towards a drawish ending. Just when he is acheiving his goal, a shocking blunder lets Kasparov mate with a bishop and a knight.
3. Kasparov-The World, Internet. Probably the most analyzed game in the history of chess. Certainly the first of its kind on such a large scale. Garry knows Q+P endings!
2. Shirov-Short, Las Vegas. A magical game from Alexi Shirov. Sacrificing first a piece then his queen, Shirov drives Short's king to the middle of the board. Black's defense collapses and White collects all his material back with interest.
1. Kasparov-Topalov, Wijk aan Zee. What else could it be? A rook and knight sacrifice to send Black's king on a walk followed by a series of quiet moves then deflection sacrifices of a bishop and a rook. Perhaps the game of the century by the player of the century.
Looking ahead to the new year, there seem to be a lot of big events coming up after a relatively quiet end of 1999 after the Las Vegas tournament. First up on the agenda is the Leko-Khalifman match starting today. I'm predicting Leko 3.5-2.5 with Khalifman remaining winless as FIDE champion.
The annual Hastings festival has started in England. I see that Tennesee GM-in-residence Sam Palatnik is taking part in the Challengers. After being upset in Round 1, he stormed back to win his next 3. Hopefully Sam can keep up the pace and we will see him in next years Premier. Irina Krush is playing in this year's Premier, also starting this week. This should be a good test for her, since I don't think she has ever faced this level of competition