No. 1  Ashley-Kempinski (1997 Bayern-Championship I - Bad Wiessee, Germany)
This was an interesting game as after five rounds, both of these players had 5-0 in a strong German tournament… 21 Grandmasters. Ashley, who had not yet earned his GM stripes, sat down in a pivotal game against Polish GM, Robert Kempinski.  Ashley has discussed this game in great detail on the ChessWise website. After trotting out the Moscow Variation with 1.e4 c5 2.Nf3 d6 3.Bb5+ Ashley played a precise opening as black lagged dangerously behind in development. Fortunately for Ashley, he never got all of his pieces out. With tremendous time pressure, this game attracted a crowd and the next several moves were played in rapid succession. Ashley stated that he saw the entire mating combination after 33.Rxb7+! The black king walked the plank after 33…Kxb7 34.Nxc5+ Kb6 35.Qb7+! Kxc5 36.Bb4+ Kd4 37.Qb6+ Kc4 38.Re4+ Kxd5 39.Qe6#.  During the excellent audiocast, Ashley made reference to his friend FM Ronald Simpson who said that Ashley's sacrifices throughout the game were justified "by verification" since black pieces were undeveloped. (See game; listen to audio analysis by GM Ashley.)

No. 2 Gwaze-D. Gluckman (2002 Chess Olympiad, Bled; Zambia vs. South Africa)
At this point, Gwaze had won his first four games in his quest for a gold medal. He sat down to play his South African opponent who intended to stop the Gwaze streak. Out of a French Defense, black played well, except he allowed his king to remain in the center for too long. White ignored the rook and calmly played 43.Ng5! Suddenly white had a strange two-piece attack! Black blundered with 43…Re7?? (Better was 43…Qd5 44.Qxd5 Nxd5 45.fxe3 Bxe3+ and black can complicate matters.) 44.Qh7+ Kf6 45.Qh6+ and black resigned. (See game)

No. 3  Tate-Braunlich (2001 U.S. Open, Boston)
This shot was easy for a player of Tate's tactical acumen. One has to be cautious playing the Najdorf against Tate… especially when the margin of error is nil. Notice both of black's rooks begging to get into the game. It seems to be a custom that Tate will sacrifice at least one piece in every game. He started his chirades with 13.Nf5!? and after 13…exf5 a series of forced moves led to Tate's 20.Qb7+! Qd7 and the hammerblow 21.Rxe6+! Black resigned since on  21…Kxe6 black will lose a queen for a rook after 22.Re1+ and mate would not be far off.  Even if black had opted for 20…Kf8 instead of 20…Qd7, it's hopeless after 21.Rxe6.  (See game)

No. 4 Carlsson-Burgess  (2002 European Cup, Halkidiki, Greece)
This game lasted 132 moves with the last 37 being part of a classic rook and bishop vs. rook ending. Carlsson, who was born in Columbia, South America, but raised in Sweden, showed good technique in a winnable, but difficult endgame. There is one humorous story concerning this same ending. In a Chicago tournament many years ago, Daniel Vasto had a Rook and Bishop against FM Leonid Kaushansky's Rook. Kaushansky was visibly upset that Vasto wouldn't settle to a draw.  The game continued and Vasto apparently missed a win and exceeded the 50 moves required to force a win… the game was declared a draw (Thanks for the correction Mr. Vasto!). Ironically, Kaushansky showed Vasto the missed opportunity after the game. An article on the same ending had appeared in U.S. Chess Life magazine months earlier! In this position, Carlsson had played well to confine the King to the edge while shielding his King from annoying checks with his bishop. Black has just played 128…Rh5-h1 to give more checks, but after 129.Kc6! Rc1+ 130.Kb6 Rb1 131.Bb2 (131.Rxb1?? stalemate!) 131…Rg1 (131… Rf1 132.Rh5 Rf8 133.Ra5+ Kb8 134.Bc5+ Kc8 135.Ra8+ Kd7 136.Rxf8) 132.Kc7! and black will be served mate in two.  (See game)