All annotations by Harvard team members

Editing and additional comments in brackets [--JLM] by Jeremy L. Martin

This file is the property of the Harvard Chess Club. It may be reproduced only in its entirety, and if this information is kept intact. Please contact the Harvard Chess Club with any questions, comments, etc.

ROUND 1

(20) Brooklyn College "B"		(5) Harvard University

Alex Galper (1930)		0-1	Matthew Bengtson (2242)
Hammad Ahmed (1586)		1/2	Jeremy Martin (2212)
Sean Collymore (1500)		0-1	Timothy Hanke (2122)
Vladimir Kirichansky (860)	0-1	John Bronsteen (2067)

1. e4 d6 2. d4 Nf6 3. Nc3 g6 4. f3 c6 5. Bc4?! b5 6. Bb3 Bg7 7. Be3 a5 8. a4 b4 9. Nce2?! (9. Nb1 =+) 9... Qc7! 10. Qd2 c5 11. g4?! Ba6 12. g5 Nfd7 13. dxc5?! Nxc5 14. Bxf7+?? Kxf7 15. Qd5+ e6 16. Qxa8 Bb7? (16... Nd7! -+) 17. Qa7 Nc6 18. Bxc5 Nxa7 19. Bxa7 Qxc2?? (19... Bxb2 -+) 20. Rc1 Qxb2 21. Rc7+ Ke8 22. Rxb7 b3 23. Kf2! Bc3 24. Nh3 Bb4 25. Bd4 Bc5 26. Bxc5 dxc5 27. Rc1? (27. Nhf4 +-) 27... Qe5 28. Nhf4 Rf8 29. Rxb3 Rxf4 30. Nxf4 Qxf4 = 31. Rbc3 Qd2+ 32. Kf1 Qxg5 33. Rxc5 Qf4 34. Rc8+?? Kd7 35. R8c3 (35. Rc7+ Qxc7 -+) 35... Qxh2 36. Rd1+?? Ke7 -+ 37. Rcd3 h5 38. Rd7+ Kf6 39. R7d2 Qc7 40. Kg2 Ke5 (declining White's draw offer) 41. Rb1 Qc3 42. Rb5+ Kf4 43. Rf2 h4 44. Rb8 Qd4 45. Rf8+ Kg5 46. Kf1 Qd1+ 47. Kg2 Qxa4 48. Kh3 Qd4 49. Rg2+ Kh5 50. Rf4 g5 51. Rfg4 Qf6 52. f4 gxf4 53. Rg7 Kh6 54. R7g4 f3 55. Rc2 f2
[0-1]

Round 1, Board 2
White: Jeremy Martin (2213)
Black: Hammad Ahmed (1586)
Grunfeld Defense

1. d4 Nf6 2. c4 g6 3. Nc3 d5 4. cxd5 Nxd5 5. e4 Nxc3 6. bxc3 Bg7 7. Bc4 O-O 8. Ne2 c5 9. Be3 Nc6 10. O-O cxd4 11. cxd4 Na5 12. Bd3 Bg4 13. f3 Be6 14. Qa4 (14. Rc1!) 14... a6! 15. Bf2? b5?! (15... Nc6! -/+, with the threat 16... b5) 16. Qb4 Bc4 17. Bxc4 Nxc4 18. a4 Rfb8 = 18... Rfb8 19. Rfc1 h6 20. h3 Qd7 21. Rab1 Nd6! (Eliminating the pressure by removing the Knight from its increasingly unstable c4-post.) 22. axb5 (I offered a draw because I didn't like the ensuing endgame; perhaps I should have played on because of the difference in ratings, but that's gotten me in trouble before.)
[1/2-1/2]

Round 1, Board 3
White: Sean Collymore (1500)
Black: Timothy Hanke (2122)
Sicilian Dragon

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 g6 6. Bc4 Bg7 7. f3 O-O 8. Be3 Nc6 9. Qd2 Bd7 10. Bb3 Qa5 11. O-O-O Rfc8 12. Nce2 Qxd2+ 13. Rxd2 Na5 14. c3 Nc4 15. Bxc4 Rxc4 16. Kc2 b5 17. b3 Rc7 18. Rd3? e5 (Oddly, White's knight on d4 is now trapped in the center of the board. However, White gets a pawn for the piece and an awkward bind on Black's position. Black spends considerable time on the clock trying to avoid further loss of material. More incisive play by White from this point on might have given Black serious problems.) 19. Nxb5 Bxb5 20. c4 Rc6 21. Rhd1 Bf8 22. Nc3 Ba6 23. a4 Kg7 24. Na2 Bc8 25. Kc3 a5 26. Kc2 Be6 27. Bc1 Nd7 28. Bb2 Nc5 29. R3d2 Nxb3! (Black mates or wins more material, e.g., 30. Kxb3 Bxc4+ 31. Ka3 d5+ 32. Nb4 axb4#.)
[0-1]

Round 1, Board 4
White: John Bronsteen (2067)
Black: Vladimir Kirichansky (860)
Queen's Gambit
1. d4 d5 2. c4 Nf6?! 3. cxd5 Nxd5 4. Nf3 e6 5. e4 Bb4+ 6. Bd2 Bxd2+ 7. Qxd2 Ne7 8. Bd3 Nbc6 9. O-O Nxd4?? 10. Nxd4 O-O (+-; 10... Qxd4? 11. Bb5+) 11. e5 g6?! 12. Nf3 b6 13. Qh6 Nf5 14. Bxf5 f6 (14... gxf5 15. Ng5) 15. exf6 exf5?! 16. Qg7#
[1-0]

ROUND 2

(5) Harvard University			(14) Toronto University "B"
					
Matthew Bengtson (2242)		1-0	Peter Giarobbe (1958)
Jeremy Martin (2212)		1-0	Robert Roller (1812)
Timothy Hanke (2122)		1-0	Kit Sungg (1852)
John Bronsteen (2067)		1-0	Ian Carmichael (1798)

1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. Qc2 c5 5. dxc5 Na6
I'm a recent advocate of the 4. Qc2 variation. Both opponents I've faced opted for the 5... Na6 variation, which is an unfortunate choice in my opinion. I enjoy playing against it.
6. a3 Bxc3+ 7. Qxc3 Nxc5 8. b4
Formerly 8. f3 (battling for e4) was considered a respectable line, until the gambit with 8... d5 9. cxd5 b6! was discovered.
8....Nce4 9. Qc2
I prefer this to the main line with 9. Qd4. It looks like White is behind in development and is moving his queen and pawns around, but in fact White has the better position here. Black's knights are very awkward. The best try here is 9... d5 10. f3 Nd6 11. c5, but White stood better here in Euwe-Johner. A game of mine from the Harvard-Yale match this year went 9... b6? 10. f3 Nd6 11. e4 e5 12. Bb2 Qe7 13. Qc3, and Black was already lost!
9... h6?! 10. f3 Ng5 11. e4?!
Actually better is 11. h4!
11... d5?!
Black needs to play 11... e5, when his knight can return to e6, with a playable game for Black. After the text White is much better.
12. e5 Nd7 13. h4 Nh7 14. f4
Also 14. Bd3 is good for White, because of the variation 14... Nxe5 15. Bxh7 g6 16. Bb2 Qf6 17. Qc3 Nd3+ 18. Kd2 Qxc3+ 19. Bxc3 Nf2 20. Bxh8 Nxh1 21. Ne2 +-. I saw this but didn't trust it. Why bother with all of that mess? After the text, Black must play 14... dxc4 if he ever intends to breathe.
14... Qc7 15. Bd3 Nhf8 16. c5! f6
Actually the best try here is 16... Nxe5 17. fxe5 Qxe5+ 18. Ne2 (not 18. Kf2 Qd4+! 19. Ke1 Qe5+!) 18... Qxa1 19. Nc3 Bd7 20. O-O, but White must stand better here too because of his better development. The text gives Black a little breathing space, but opens up the position, not to mention the Black king position, for White's predatory bishops. Now that White has won the strategical battle, on move 17 he begins the development of his pieces! It's important for a chessplayer to keep his priorities straight.
17. exf6 gxf6 18. Nf3 e5 19. fxe5 fxe5 20. O-O Ne6
If 20... e4 21. Bxe4 dxe4 22. Qxe4+ Kd8 23. Bb2 wins for White.
21. Bb2 Rg8 22. Rae1!
Most precise. White allows the counterattack with 22... Nf4 because after 23. Nxe5 Rxg2+ 24. Qxg2 Nxg2 25. Kxg2 Black's days of pleasure will be numbered. [Noam Elkies pointed out the beautiful sideline 25. Nc6+ Nxe1 26. Bg6 mate! --JLM] Because White has hammered the central dark squares so heavily, Black must now advance a pawn, but this allows a devastating sacrificial breakthrough.
22... e4 23. Bxe4 dxe4 24. Qxe4 Nf8 25. Nd4 Bd7
To stop the threat 26. Nb5.
26. Rxf8+ Kxf8 27. Nxe6+ Bxe6 28. Qxe6 Rg7
Now White disdains a couple of obvious forced wins in favor of the most artistic conclusion, in the spirit of the Romantic Era.
29. Rf1+! Rf7 30. Qxh6+ Ke8 31. Qh8+ Ke7 32. Re1+!
[Watching the game, I honestly didn't see anything wrong with 32. Rxf7+ Kxf7 33. Qh7+, but Matt had seen a forced mate. --JLM]
32... Kd7 33. Qd4+ Kc6 34. Re6+ Kb5 35. Qd3+ Ka4 36. Qc2+ Kb5 37. a4+ Kxb4 38. Qc3+ Kxa4 39. Qc4+
[1-0]

Round 2, Board 2
White: Robert Roller (1812)
Black: Jeremy Martin (2213)
Benko Gambit
Annotations by Jeremy Martin

1. d4 Nf6 2. c4 c5 3. d5 b5 4. cxb5 a6 5. Nc3 axb5 6. e4 b4 7. Nb5 d6
This is the sharp Zaitsev variation of the Benko Gambit; White forgoes winning a pawn, choosing instead to place his QN aggressively and try to slaughter Black with a quick e4-e5. If Black can get out of the opening safely, he will likely stand better, but White has considerable attacking chances.
8. Bf4
The most common move, intending 9. e5, although 8. Bc4 (the so-called "Nescafe Frappe Attack") is popular as well.
8... g5
I knew that this was the theoretically best move, but thought for half an hour nevertheless, since both main lines involve sacrifices for Black -- 9. Bxg5 Nxe4 10. Bf4 and now either 10... Qa5 11. Bc4 b3+ 12. Kf1 Qxa2 13. Rxa2 bxa2 14. Bxa2 Rxa2 or 10... Bg7 11. Qe2 Nf6 12. Nxd6+ Kf8 13. Nxc8 Qxc8. We decided later that these positions are both good for Black, but I wasn't so sure at the time. Anyway, quieter moves seem to give White too much, e.g., 8... g6 9. e5 or 8... Nbd7 9. Nf3 intending 10. e5.
9. e5? gxf4 10. exf6 e5?!
Fedorowicz gives 10... Nd7! as much better for Black, who will have the two Bishops and a lot of play on the dark squares.
11. dxe6
Otherwise Black will just stop Knight checks, grab the f-pawn, and be a pawn up for nothing.
11... Bxe6 12. Qf3 Rxa2
I had looked at a lot of messy lines with 12... Qa5 13. Qxf4 b3+ etc., but luckily realized that 12... Qa5?? would lose to 13. Qxa8! So the text move was all but forced -- if the Rook is going to move along the a-file, I may as well grab the pawn!
13. Rxa2 Bxa2 14. Qxf4?
This "obvious" move is actually inferior. A much better move is 14. Qe4+!, when 14... Be6 15. Qb7 Bh6 16. Nc7+ is very unpleasant. So Black would play 14... Kd7, whereupon 15. Qxf4 is unclear -- both sides have very little piece coordination!
14... Nd7!
Black must eliminate the f6-bone in his throat no matter what.
15. Nxd6+ Bxd6 16. Qxd6 Qxf6
Now that the dust has settled, it is apparent that Black has a considerable advantage after having traded the weak d-pawn and inactive KB for White's strong f-pawn and dangerous Knight. The major factors in the position are now Black's advanced Queenside majority and a large lead in development.
17. Qxf6
Nor is the middlegame after 18. Qd2 Be6 pleasant for White.
17... Nxf6 18. Bb5+ Ke7 19. Nf3 Rb8 20. Ba4?
The Bishop is misplaced here. Better 20. Be2.
20... Bd5 21. Kd2 Ra8
Activating the Rook and keeping the threat of ...Bxf3 in reserve. Now White has no good moves: 22. Bc2 Ra2 23. Kc1 Ra1+ 24. Bb1 Be4 25. Nd2 Bxg2 and 22. Bd1 Ne4+ are crushing.
22. b3 c4 23. bxc4??
A time-pressure blunder (White had six minutes for 25 moves) but after anything else Black gets a protected passer on c3. The rest of the game requires no comment.
23... Bxf3 24. Re1+ Kd6 25. gxf3 Rxa4 26. Rb1 Kc5 27. Kd3 Ra3+ 28. Kd2 Kxc4 29. f4 Ne4+ 30. Kc2 Ra2+ 31. Kc1 Nc3
[0-1]

Round 2, Board 3
White: Timothy Hanke (2122)
Black: Kit Sungg (1852)
Sicilian Defense, Sveshnikov-Alapin Variation

This game is notable for the concluding combination.
1. e4 c5 2. c3
All three of my games with White began in this Sicilian variation.
2... d5 3. exd5 Qxd5 4. d4 e6 5. Nf3 Nc6 6. Be2 Nf6 7. O-O Be7
According to IM Gary Lane in "The 2.c3 Sicilian," White could now get a favorable endgame by 8. c4! Qd8 9. dxc5 Qxd1 10. Rxd1 Bxc5 11. Nc3.
8. Bf4 O-O 9. Ne5 Nxe5 10. Bxe5 cxd4 11. cxd4 Qa5 12. Nc3 Bd7 13. Bf3 Nd5 14. Re1 Rfe8 15. Be4 g6 16. Rac1 Rac8 17. Qd2 f6 18. Bg3
Played after 10 minutes thought. I was considering a double-bishop sacrifice starting with 18. Bxg6 but rejected it as too risky. However, in postgame analysis it seemed to both players that 18. Bxg6 would have won. There are many interesting variations and it would have been easy for either side to go wrong. Now White's position begins to fray.
18... Bf8 19. h4 Bc6 20. Bb1 Nxc3 21. bxc3 Qd5?!
Played after 10 minutes thought. Black could have won a pawn starting with 21... Ba3, but rejected the idea because he did not like the resulting position -- certainly an unwise decision.
22. Be4 Qb5 23. c4 Qb4 24. Qxb4
White might have been expected to keep queens on in this position. However, White is looking ahead to the rook ending. The play from here to the end of the game is sharp and interesting. Watch White's c- and d-pawns march forward!}
24... Bxb4 25. Bxc6 Rxc6 26. d5! Ra6
Practically forced. The poor position of this rook is crucial to the resulting play.
27. Re2 Bd6 28. c5 Bxg3 29. fxg3 Kf7 30. d6! Rd8 31. Rb2 Rd7 32. R1b1 b6 33. c6!!
The start of a long winning combination, my best of the tournament. White sacrifices the d-pawn to force the c-pawn through.
33... Rxd6 34. Rc1!
This rook must go to c1 so White can offer the other rook on d2 without being in check.
34... Rd8 35. Rd2! Ke7 36. R1d1! Rxd2 37. Rxd2 b5
Much better is 37... Ra5! but I believe 38. Rd7+ Ke8 39. Rxh7 Rc5 40. c7 f5! 41. Kf2 should win eventually, since Black's rook is tied to the c-file and his king has no moves at all. Sample variation to show the main ideas: 41... b5 42. Ke3 e5 43. g4! fxg4 44. Ke4 a5 45. h5! gxh5 46. Kf5 +-. After 37... b5 as played, Black did not see White's win nor did one of White's teammates, who at this point reported to other Harvard team members that White was probably losing!
38. Rd7+ Ke8 39. Rd6!!
Suddenly it is obvious: nothing can stop the c-pawn, as after 39... Ke7, 40. c7 forces a new Queen.
[1-0]

Round 2, Board 4
White: Ian Carmichael (1798)
Black: John Bronsteen (2067)
Queen's Gambit Declined, Exchange Variation

1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Bg5 Nbd7 5. cxd5 exd5 6. Nc3 Be7 7. e3 Nf8 8. Bd3 c6 9. O-O Bg4 10. Re1 Ne4 11. Bxe7 Qxe7 12. Be2 Nd7 13. Nd2 Bf5 14. f3?! Nxc3 15. bxc3 Qxe3+ 16. Kh1 O-O-O!? (16... Qxc3!?; 16... O-O?!) 17. Bb5 Qf4 18. g3 Qd6 (=/+) 19. Ba4 Rhe8 20. Qb3? b5 21. Bxb5 cxb5 22. Qxb5 Nf6 [White has insufficient compensation -- JLM] 23. c4 Qb6 24. Qa4 Rxe1+ 25. Rxe1 Qxd4 26. Qa6+ Kb8 27. c5 Qxc5 28. Nb3 Qb6 (-+) 29. Qa3 Qf2 30. Qb4+ Ka8 31. Na5 Rb8 32. Qc3 Qxe1+ 33. Qxe1 Rb1
[0-1]

ROUND 3

(5) Harvard University			(1) New York University "A"
					
Matthew Bengtson (2242)		1-0	GM Ilya Gurevich (2665)
Jeremy Martin (2212)		0-1	IM Mark Ginsburg (2476)
Timothy Hanke (2122)		1/2	Dimitriy Leykekhman (2304)
John Bronsteen (2067)		0-1	Yuan Xu (2075)

1. d4 Nf6 2. c4 g6 3. Nc3 d5 4. cxd5 Nxd5 5. e4 Nxc3 6. bxc3 Bg7 7. Be3 c5 8. Qd2 O-O 9. Rc1
Being a Grunfelder myself, I opted for the line which I find most unpleasant for Black to face, especially when needing a win. Karpov favored it against Kasparov in New York / Lyon. Actually 9. Nf3 is more precise, as the text allows Black to set up nicely with 9... Nd7.
9... Qa5 10. Nf3 Rd8?!
Theory gives ...e6 or ...cxd4, in both cases with a slightly more comfortable game for White. I should have reacted with 11. d5! right away, as my slow response allows Black to reach the characteristic endgame after ...cxd4 with an extra tempo. Black avoided this, still scraping for complications, I guess.
11. h3?! b6?! 12. d5! e6 13. c4 Qxd2+ 14. Bxd2?!
I was correct to avoid 14. Kxd2 because of Black counterplay against the White king on d2, with ideas like ...b5, ...f5, and ...Nc6. If anyone is really interested e-mail me and I'll send a page worth of analysis of this position - it is pretty fascinating actually. Anyway, 14. Nxd2! is correct, bolstering the center and obtaining slightly more than a normal opening advantage.
14... exd5!
Now I realized that Black was getting serious play, and I began to get worried.
15. exd5 Re8+ 16. Be2 Bf5 17. Be3
An oversight, but a lucky one. Black gains the exchange, but not necessarily an easy advantage! The only way to keep material here is 17. Kf1; this move deserves consideration as well.
17... Bb2 18. O-O Bxc1
Ilya thought for a while here before taking the rook. He is losing a critical bishop, leaving behind a trail of weaknesses. Black must take the exchange to justify his play, but having grabbed it he must defend very carefully - making a careful plan to guard the weaknesses and neutralize White dangerous bishop pair. However, he seemed not to take the game or his opponent too seriously at this point, and committed several serious inaccuracies. I got a vicious attack as a result. (Everyone should be aware that in the Harvard Chess Club, a rook is considered to be worth less than a bishop or a knight - many of our blitz games are won the exchange down!)
19. Rxc1 Bd7?! 20. Bd3 Kg7? 21. Bd2! f6 22. Bc3 Kf7 23. g4!
Not even allowing 23. Nd2 Bf5! White is threatening 23. g5 and Black is thrown onto the defensive against the menacing bishop pair.
23... h6 24. Nh4! Rg8 25. Re1 Na6 26. Bxg6+!!
A beautiful shot which, if followed up correctly, would have been my ticket to immortality.
26... Rxg6 27. Nxg6 Kxg6 28. Re7
Black's extra piece is of no consequence, since White's bishop, being "opposite," is essentially extra, and Black's rook can't move, while his knight is hopelessly out of play.
28... Ba4?!
A slightly better try is 28... Be8, but then 29. f4 or 29. Re6 would be powerful for White.
29. f4 h5 30. f5+ Kg5 31. Bd2+ Kh4 32. Kg2
A beautiful position to sit and contemplate in one's first encounter with a GM! Unfortunately I got excited here and hallucinated, realizing too late that the f5 pawn hung in some critical variations.
32... Bd1 33. Be1+ Kg5 34. Kg3??
With this move I declined a draw, missing the win that was to be had with 34. Bd2+ Kh4 35. g5!
34... hxg4 35. hxg4 Bxg4 36. Rg7+ Kxf5 37. Rxg4 Re8 38. Rf4+ Kg6 39. Kf2 Nb4 40. Bxb4 cxb4 41. Rg4+! Kf5 42. Rg7 a5 43. Rc7 Re4 44. d6!
With perfect defense White has forced a barely drawn position - with even more perfect defense, that is! My opponent's reaction here was an astonishing blunder. I couldn't believe my eyes. Best play from here is 44.... Rd4 45. c5! Ke4! 46. Re7+! Kf4 47. cxb6 Rxd6 48. b7 Rb6 49. Re6! Rxb7 50. Rxf6+ Ke4 51. Ra6 Rb5 52. Ke2 and White barely clings onto a draw.
44... Ke6???? 45. Re7+ Kxd6 46. Rxe4 Kc5 47. Ke3 b5 48. cxb5 Kxb5 49. Kd3 Ka4 50. Kc2 Ka3 51. Kb1
[1-0]
An amazing turn of events.

Round 3, Board 2
White: IM Mark Ginsburg (2476)
Black: Jeremy Martin (2213)
English Opening

No miracles here, unlike what happened on first board; I just got crushed.
1. c4 g6 2. d4 Nf6 3. Nc3 Bg7 4. e4 d6 5. Be2 c5 6. Nf3 O-O 7. O-O cxd4 8. Nxd4 Nd7 9. Bg5 Nc6 10. Nc2 Rc8 11. Rc1 Be6 12. b3 a6?! 13. f3 Re8 14. Qd2 Qa5 15. Rfd1 Kf8 16. a3 Kg8?! 17. b4 Qd8 18. Nd5 Bxd5 19. exd5 Ne5 20. Nd4 Rc7? (20... Qd7!) 21. Bf4 Nh5 22. Be3 Nf6 23. Nb3 Qc8 24. Na5 Ned7 25. Bf1 Qb8 26. c5 b6 27. Nc6 Qb7 28. Nxe7+ Rxe7 29. c6 Qd7 30. cxd7 Rexd7 31. Rc6 Rxc6 32. dxc6 Rd8 33. Qd3 (33. Qd4 is quicker but it doesn't matter by this point) 33... Qc7 34. Qxa6 Qxc6 35. Qxb6 Qe8 36. Bf2 d5 37. Bb5 Nd7 38. Qb7 d4 39. Re1
[1-0]

Round 3, Board 3
White: Timothy Hanke (2122)
Black: Dmitriy Leykekhman (2304)
Sicilian Defense, Sveshnikov-Alapin Variation
Annotations by Timothy Hanke

In this game White plays badly enough to lose, but Black, who went on to win the Board 3 prize with an overall score of +4=2-0, never delivers the knockout punch.
1. e4 c5 2. c3 d5 3. exd5 Qxd5 4. d4 Nc6 5. Nf3 cxd4 6. cxd4 e5 7. Nc3 Bb4 8. Bd2 Bxc3 9. Bxc3 e4 10. Ne5 Nxe5 11. dxe5 Ne7 12. Qxd5?
Not recommended by theory. White can get the advantage with other moves such as 12. Qa4.
12... Nxd5 13. Bb5+ Bd7 14. Rd1 Bxb5 15. Rxd5 Bc5 16. Rd4 Bxa2 17. O-O O-O 18. Rxe4 Rfd8 19. f4
The start of a bad plan to attack the kingside, which comes to nought. Now White begins to lose the thread of the game.
19... Bd5 20. Re3? (20. Rd4! =) 20... Bc4 21. R1e1 Be6 22. Re4 h5 23. Rd4 Rxd4 24. Bxd4 a5 25. Kf2 Rc8 26. Be3? (26. Bb6! and 27. Re2 =) 26... b5 27. Rc1 Rd8 28. Rc5 Bc4 29. b3 Bxb3 30. Rxb5 a4 31. Rb4? (31. Ra5!) 32... Rc8 32. f5 Rc2+ 33. Kg3 Bd5?
Trying impatiently to find a tactical solution on the kingside, Black throws away his chances on the queenside by trading his a-pawn for White's g-pawn. I breathed a sigh of relief.
34. Rxa4 Rxg2+ 35. Kh3 Re2 36, Bf4
White has constructed a fortress position based on the presence of bishops of opposite color.
36... Be4 37. Kh4 Bxf5 38. Kxh5 Kh7 39. Kg5 Be6 40. Bg3 Re3 41. Rb4 Rf3 42. Rf4 Rb3 43. Rh4+ Kg8 44. Rd4 Rb5 45. Kf4 Kh7 46. Ke3 Kg6 47. Ra4 Rb3+ 48. Kd4 Kf5 49. Kc5 g5 50. Rd4 Rc3+ 51. Kd6 Ra3 52. Kc5 Ra5+ 53. Kd6 Ra3 54. Kc5 Rc3+ 55. Kd6 Rc2 56. Rb4 Rd2+ 57. Kc5
[1/2-1/2]

Round 3, Board 4
White: Yuan Xu (2075)
Black: John Bronsteen (2067)
Guioco Pianissimo

1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. d3 Bc5 5. O-O d6 6. c3 Bg4 7. Nbd2 O-O 8. h3 Bh5 9. Re1 Kh8 10. Bb3 g5?! (Probably too ambitious.) 11. g4 Qd7!? 12. Nh2 Bg6 13. Ndf1 h6 14. Be3 Bxe3 15. Nxe3 h5?! (Ditto.) 16. Qf3 Kg7 17. Nd5! Nxd5 18. exd5 Ne7 19. gxh5 Bf5 20. Ng4 Bxg4 21. hxg4 +/- f5 22. Qe3 fxg4? 23. Qxg5+ Kh7 24. d4 Rf3 25. Bc2+ Kh8 26. Qh6+ Kg8 27. Bh7+ Kf7 28. dxe5 dxe5 29. Rxe5 +- Rf6 30. Qg5 Rh8 31. Rae1 Rxh7 32. Rxe7+ Qxe7 33. Rxe7+ Kxe7 34. Qe5+ Kf7 35. Qxc7+ Kg8 36. Qc8+ Rf8 37. Qe6+ Kh8 38. d6 Rg7 39. d7 g3 40. fxg3 Rxg3+ 41. Kh2 Rg7 42. h6 Rgg8 43. Kh3 Rf3+ 44. Kh4 Rff8 45. Kh5
[1-0]

ROUND 4

(13) Rhode Island College "A"		(5) Harvard University

Oscar Maldonado (2280)		0-1	Matthew Bengtson (2242)
Jim Della Selva (2080)		0-1	Jeremy Martin (2212)
Joshua Anderson (1731)		0-1	Timothy Hanke (2122)
Roberto Campusano (1519)	0-1	John Bronsteen (2067)

1. d4 Nf6 2. c4 g6 3. Nc3 d5 4. Nf3 Bg7 5. g3? dxc4 =+ 6. Qa4+ Nfd7 7. Qxc4 O-O 8. Bg2 Nb6 9. Qd3 Nc6 10. O-O? (10. Be3 is forced but unpleasant -- if 10. e3 e5!) 10... Nxd4 11. Nxd4 Qxd4 -+ 12. Qf3 c6 13. Rd1 Bg4! 14. Qf4 Qxf4 15. Bxf4 Rfd8 16. h3 Be6 17. a4 Rxd1+ 18. Rxd1 Bxc3 19. bxc3 Nxa4 20. c4 Nc3 21. Rd2 Bxc4 22. Bf3 Bd5?! 23. Be5! Bxf3 24. Bxc3 Bd5 25. f3 Kf8? (Stupid. Just ...a7-a5-a4 etc. is an easy win.) 26. g4 Ke8 27. e4 Bc4 28. h4 a5 29. h5 b5?! (Simplest is ...g5) 30. Rh2 Kd7?! 31. Rd2+ Ke6? 32. f4 f5 33. exf5 gxf5 34. g5 b4 35. Be5 Rg8! (Finally Black wakes up and plays some good moves to earn the win.) 36. g6 Bd5 37. Kf2 h6 38. Rc2 a4 39. Bd4 a3! 40. Bc5 Rb8 41. Re2+ Be4 42. Rd2 Rb5! 43. Rd8 Rxc5 44. g7 Rc2+ 45. Ke3 Rg2 46. g8=Q Rxg8 47. Rxg8 b3
[0-1]

Round 4, Board 2
White: Jeremy Martin (2213)
Black: Jim Della Selva (2080)
King's Indian Defense
Annotations by Jeremy Martin

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Be2 O-O 6. Nf3 e5 7. O-O Nc6 8. d5 Ne7 9. Ne1 a5?!
Instead of the tried and true 9... Nd7, Black tries to set up a fortress on the Queenside. A questionable idea.
10. Nd3 Nd7 11. Be3!?
This move can be risky. The Bishop is well placed to attack Black's Queenside, but Black will gain a tempo with f5-f4 and White will not be able to put his Knight on f2, meaning that the g5-g4 break will be easier. However, I thought that Black's 9th was a waste of time and that I could also take liberties.
11... b6?!
Evidently Black wants to trade off the a- and b-pawns before letting me play c4-c5, on the principle that I won't have anything to attack on the Queenside. First of all, this isn't true: I can attack the c- or d-pawn. Secondly, this is another move made on the wrong side of the board; better 11... f5 at once.
12. a3 f5 13. f3 f4 14. Bf2 Ba6?
The Bishop is needed at c8 to support the g5-g4 push.
15. b3?!
White could gain a tempo on the game by playing 15. Nb5! immediately.
15... g5 16. Qc2 Rf7 17. Nb5!
Blocking Black's QB so that b3-b4 is playable. It also ties Black's Queen to the c-pawn.
17... Bf8 18. b4 Bxb5?
The QB should not be squandered like this, but if Black is insistent upon trading, he should play 18... axb4 first, with the ideas 19. axb4 Bxb5 or 19. Nxb4 Nc5, although White retains advantages in all lines.
19. cxb5 Ng6 20. Rfc1
First ascertaining the intentions of the d7-Knight...
20... Nf6 21. bxa5
... then opening Queenside lines now that Nd7-c5 is impossible.
21... Rxa5
Or 21... bxa5 22. a4, when White has all kinds of play on the Queenside (e.g., attacking the a-pawn or playing for b5-b6) and Black's Kingside attack has barely gotten off the ground.
22. Nb4 Ra8 23. Na6 Ne8 24. a4 Qc8?
Black's position is bad, but this blunder makes matters hopeless. A natural move like 24... Rg7 was a better chance.
25. Bxb6 g4 26. Nxc7
Not 26. fxg4?! f3! 27. Bxf3 Bh6 when the bad KB springs to life, winning material. Of course White would still be up after 28. Qc6 Bxc1 29. Rxc1, but the text is much clearer.
26... Nxc7 27. Bxc7 gxf3 28. Bxf3 Nh4 29. Kh1
All White has to do to defend is blockade f3 firmly. The rest is pretty easy.
29... Rg7 30. Bb6 Qe8 31. Qf2 Be7 32. Rc6 Qg6 33. a5 Kh8 34. Rac1 Rag8 35. Rc8 Qh6 36. Rxg8+ Rxg8 37. a6 Ng6 38. a7 Bh4 39. Qe2 Bg3 40. Bg1 Nh4 41. b6 Nxf3 42. Qxf3 Bh4 43. b7
[1-0, time]

Round 4, Board 3
White: Joshua Anderson (1731)
Black: Timothy Hanke (2122)
Sicilian Dragon
Annotations by Timothy Hanke

Inaccurately played by both sides, but in some ways my most interesting game of the tournament.
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 g6 6. Be3 Bg7 7. f3 O-O 8. Qd2 Nc6 9. Bc4 Bd7 10. O-O-O Ne5 11. Bb3 Qa5 12. g4 Rfc8
The Yugoslav Attack in the Dragon Variation is one of a very few openings that even B- and C-players know a dozen moves deep!
13. Qe2
I want to give this move a question mark because it looks so weak, but I looked it up and it does exist in GM play. However, it doesn't seem to give White much.
13... Nfxg4?!
An interesting but probably unsound attempt to take immediate tactical advantage of the white queen's position. [Actually, a piece sacrifice at g4 is a common theme in the Dragon, and it seems even better here since Black picks up an exchange as well, even if he does sack it right back. --JLM]
14. fxg4 Bxg4 15. Qf1 Rxc3!
This move seems essential if Black is to make progress on the Queenside.
16. bxc3 Bxd1 17. Kxd1 Qxc3 18. Qe1 Qb2
Black has three pawns for a piece and attacking chances. However, here it would be practically suicidal for Black to trade queens. Black must use the open lines to develop threats. White on the other hand must find a concrete plan to exploit his extra piece. [Stated another way, one of Black's advantages is the exposed position of White's King, which is irrelevant without Queens on the board. --JLM]
19. Rg1
A reasonable move that increases pressure on the Black kingside. Black must be careful not to allow sacrifices that blow open his pawn cover.
19... Rc8 20. Qf1 Bf6
Black thought for 22 minutes on this defensive move.
21. Rg2 Qc3 22. Qe1 Qb2 23. Qf1 a6 24. Nf3 Rc3!
Now all four of Black's pieces are "fianchettoed" on the long diagonal. Afterward one master joked, "Does this create an overpowering astrological force?" Actually the move pinpoints the faults in White's position, which now crumbles.
25. Nxe5 Rxe3 26. Nd3
[Note that 26. Nxf7?? Qa1+ loses the Queen and 27. Bxf7+? Kg7 does nothing about the skewer threat, while putting yet another piece en prise. Thus the text is forced. --JLM]
26... Qa1+ 27. Nc1 Qd4+ 28. Nd3 Rxe4 29. Qf3 d5
[Not only supporting the Rook, but blocking the White Bishop's diagonal. --JLM]
30. Re2 Qa1+ 31. Nc1 (31. Kd2 Bc3#!) 31... Rd4+ 32. Rd2 Rxd2+ 33. Kxd2 Bg5+ 34. Ke2 Qxc1 35. Qxd5 Qe3+ 36. Kf1 e6 37. Qxb7 Bh4
[0-1]
Perhaps slightly premature, but White's king has a short life expectancy after 38. Kg2 Qf2+ 39. Kh3 h5!, not to mention Black's "slight" two-pawn edge.

Round 4, Board 4
White: John Bronsteen (2067)
Black: Roberto Campusano (1519)
Queen's Gambit Declined

1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Bg5 Be7 5. e3 O-O 6. Qc2 Nbd7 7. cxd5 exd5 8. Bd3 c6 9. Nf3 Qc7 10. O-O Re8 11. Bxf6 Nxf6 12. Ne5 c5 13. f4 c4?! 14. Be2 +/- 14... g6 15. g4 Rd8 16. Bf3 Be6? 17. f5 gxf5 18. gxf5 Bd7? 19. Nxd5 Nxd5 20. Bxd5 Be8 21. f6! Bf8 22. Kh1! +- 22... Rxd5? 23. Qg2+ Kh8 24. Rg1
[1-0]

ROUND 5

(5) Harvard University			(8) McGill University "B"
					
Matthew Bengtson (2242)		1/2	Joel Ouaknine (2347)
Jeremy Martin (2212)		1/2	Pascal Bergeron (2100)
Timothy Hanke (2122)		1-0	Gilbert Cabana (2134)
John Bronsteen (2067)		1-0	Jean Efproxiadis (1600)

[Note that many of the McGill players' ratings are CFC or QFE (Canada and
Quebec respectively), which correspond to slightly higher USCF ratings.]

1. d4 g6 2. e4 Bg7 3. c4 d6 4. Nc3 Nf6 5. h3
My favorite anti-King's Indian system.
5... O-O 6. Be3 c6 7. Nf3
Avoiding 7. Bd3 because of ...e5 8. d5 b5! with equality according to the omniscient ECO. My opponent went for an unorthodox setup against it. I was expecting either ...e5 or some Robert Byrne kind of thing with ...a6 and ...b5, but that wouldn't be too effective: White just castles short and plays b4, playing for constriction as always.
7... Qa5 8. Bd3 e5 9. dxe5
One of my favorite plans, trying to exploit the hole on d6 and the queen on a5. Pushing d5 leads to problems after ...Nh5. Probably the best try for the advantage is just castling and then playing for b2-b4.
9... dxe5 10. O-O Nbd7 11. Rb1 Re8!
A very profound and farsighted move. It is interesting that I thought White was better because of the hole on d6, while Black thought he was better because of the hole on d4! It turns out that my own evaluation was closer to the truth. After my 13th move, the idea ...Nh5-f4 is harmless because of Bf1, Kh2, and g2-g3.
12. b4 Qc7 13. Qc2 Nf8 14. c5 Ne6 15. Rfd1 Bd7 16. Nd2 Rad8
Unfortunately Black was able to resist the temptation to hop into d4 with his knight, as I would chop it off, giving him a useless passed pawn that blocks a good square for his pieces, healing my own pawn structure. Then I could take control of the board with f2-f4, e4-e5, and Nc4-d6.
17. Nc4 Bc8
It is very frustrating for White that he has executed all of his plans and still can't find any way to make real progress. The best try here is 18. Qa4 Qb8 (not ...a6, which weakens b6) 19. b5!?, but that is risky. I had quite a bit of respect for my opponent at this point, and decided to liquidate to a boring position and offer a draw. This was a mistake, as I should have kept the "better half" of a drawn position when I offered the draw. When I offered it, Black amazingly enough found a way to keep play in the position, and seize the edge!
18. Nd6 Re7 19. Bc4 Ne8 20. Nxc8 Qxc8 21. Rxd8?!
Best is 21. Bxe6! keeping a minuscule plus.
21... Nxd8! 22. Bg5 Rd7 23. Bxd8
I want to take off everything that could get to d4. But Black avoided 22... Bf6 because he guessed that this piece would have good prospects - an amazing bit of foresight, as it arrives in d4 itself!
23... Qxd8 24. Rd1
Here I offered the draw. He wanted to take it, actually, but had to press for the win as our team was doing well on other boards.
24... Rxd1+ 25. Qxd1 Qxd1+ 26. Nxd1 Bh6 27. Nb2 Bd2 28. Nd3 Bc3
Now if I were aware of the danger I would have tried to blast the queenside with b5, leading the game to an inevitable draw.
29. Kf1 Kf8 30. Ke2 Ke7 31. Kd1?!
Better is 31. g4! blockading the position. I was hesitant to weaken my kingside like this, but I could defend with Kf3-g3, and chop off the Black knight when it activates itself.
31... f5 32. exf5 gxf5
Not ...e4 33. fxg6!
33. Kc2 Bd4 34. f3
White's defensive plan involves blockading on the White squares, of course, with a well-timed g2-g4. But Black succeeds in making it hard.
34... Nc7 35. Nc1 Nd5
Now I have to take.
36. Bxd5 cxd5 37. Nb3?
The losing move!! I admit that I thought I was better here, since the bishop doesn't work as well with center pawns. I figured I could blockade them with my knight and king (with the aid of g4) and then give him fits with my wing majorities and my tricky knight. He thought he was killing me with his raking bishop and massive pawn center. This time he was correct. This should have been the last chance to draw with 37. g4! This very complicated endgame was botched by both sides in time pressure.
37... Bf2 38. Na5 b6 39. Nc6+ Ke6 40. cxb6 Bxb6?
Throwing away the win that was to be had with ...axb6! Now White can draw with 41. g4! which bites a chunk out of Black's pawn mass before it becomes too dangerous.
41. a4? Kd6 42. b5 e4 43. fxe4 fxe4 44. g4
White loses by a tempo after 44. a5 and 45. Nxa7, hence the text. But now Black could probably win by biting a chunk out of White's pawn mass with the key move ...a6.
44... d4?? 45. a5 Bc5 46. Nxa7 Bb4 47. Nc6 d3+ 48. Kb2! Be1 49. a6 Kc7 50. Ne5?
The last move before the time control spoils an excellent chance to play for the win with 50. Ne7! Kb6 51. Nc8+ Kc7 52. a7 Kb7 53. Nd6+! (not 53. b6? Bb4! and *Black* wins!) Kxa7 54. Nxe4 Kb6 55. Nf6! h6 56. Nd7+! Kxb5 57. Ne5 d2 58. Kc2 Bg3! 59. Nf3 Kc6 60. Kxd2 Kd5 61. Ke3. I don't honestly know the evaluation of this position.
50... Bf2 51. Nxd3
After 51. Kc3? Bd4+! 52. Kxd4 d2 53. b6+ Kd8! Black can just eke out the full point. I was weary of the game here; we had already won the match, and I headed for the safe haven of a draw.
51.... exd3 52. Kc3 Bb6 53. Kxd3 Kxb5 54. Ke4 h6
[1/2-1/2]

Round 5, Board 2
White: Pascal Bergeron (2200)
Black: Jeremy Martin (2213)
Sicilian Najdorf
Annotations by Jeremy Martin

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 e6 7. Be2 b5??
Thoughtless and premature -- almost any calm developing move is better.
8. Bf3! e5
Forced to prevent mayhem along the long diagonal with 9. e5.
9. Nf5 g6?!
Leading to an unpleasant invasion, but how else is the annoying Knight to be ejected from f5? 9... Bxf5 10. exf5 gives White the two Bishops and opens the center, a dubious alternative.
10. Nh6!
The Knight now remains here essentially forever, a bone in Black's throat -- castling is impossible and f7 is weak.
10... Be6 11. O-O Be7 12. Qd2 Nbd7 13. Nd5 Bxd5 14. exd5 Rc8?
Further weakening Black's already barren Queenside; perhaps Black should have played Nf6-g8 at some point to oust the intruder, although even then White will be much better.
15. a4!
Of course: White opens the Queenside, where Black's premature 7th move has left nothing but weaknesses.
15... Qc7 16. axb5 axb5 17. Be2 Qb7 18. Ra7 Qxd5 19. Qxd5 Nxd5 20. Bxb5 Nc7 21. Bxd7+ Kxd7 22. Nxf7 Rf8 23. Ng5 Rh8?!
The ending is, of course, horrible for Black, but this retreat is nonetheless abject. Marginally better was 23... h5.
24. Rfa1 Rb8 25. b3 Bf6?! 26. Ne4 Be7 27. Nc3 Kc6 28. Rxc7+! Kxc7 29. Ra7+ Rb7
Since 29... Kd8 30. Nd5 Bf8 31. Bg5+ is nightmarish.
30. Nd5+ Kb8 31. Nxe7 Rxa7 32. Nc6+ Kb7 33. Nxa7 d5 34. Nb5 Ra8
Black is quite lost; all White has to do is activate his King and neutralize the center pawns, while pushing his own passed pawn.
35. Ba7??
A dreadful blunder which throws away the win. Of course Black cannot take the Bishop, as the pawn ending is lost, but the Bishop will never escape from a7! The simple 35. Kf1 was sufficient for the win.
35... d4! 36. g3 d3
Ruining White's pawn structure. White retains some winning chances in the case of 36... Kc6?! 37. Nxd4+ cxd4 38. Bxd4. However, even better was 36... Rd8!, preventing tricks on d4, followed by 39... Kc6.
37. cxd3 Kc6 38. Nd4+ Kd5!
Clearer than 38... cxd4 39. Bxd4, although this is a draw as well. 39. Nb5 Kc6 40. Nd4+ Kd5 41. Nb5 Kc6
[1/2-1/2]
White could have tried 40. Bd4 cxd4 41. Nxd4, but Black will be able to win one of the Queenside pawns, whereupon the active Rook will be sufficient to hold the draw.

Round 5, Board 3
White: Timothy Hanke (2122)
Black: Gilbert Cabana (2134)
Sicilian Defense, Sveshnikov-Alapin Variation
Annotations by Timothy Hanke except [JLM]

This game literally is a fight to the last pawn.
1. e4 c5 2. c3 d5 3. exd5 Qxd5 4. d4 cxd4 5. cxd4 Nc6 6. Nf3 e5 7. Nc3 Bb4 8. Bd2 Bxc3 9. Bxc3 exd4? 10. Nxd4 Nf6 11. Nxc6 Qxc6 12. Bxf6 gxf6
On a positional level Black is probably already lost, but as the subsequent course of the game shows, White lacks sufficient technique to demonstrate the proof.
13. Rc1 Qe4+ 14. Qe2 Qxe2+ 15. Bxe2 Be6 16. Bf3?!
16. a3 followed by 17. O-O should give White a comfortable advantage. Instead White goes in for a positional combination.
16... Bxa2 17. O-O O-O 18. Ra1 Bc4 19. Rfc1 Ba6 20. Rxa6 bxa6 21. Bxa8 Rxa8
Black's position looks horrible. He is a pawn up, but all five of his pawns are isolated and four of them are doubled. Each side has one rook. This is the position White envisaged on move 16, but is it winnable for White? If so, how? Should White keep rooks on or try for an advantageous king-and-pawn ending?
22. Kf1 Rb8 23. Rc2 Kf8 24. Ke1 Ke8 25. Rd2 Ke7 26. Kd1 Rd8 27. Rxd8 Kxd8
White now finds himself in a pawn ending that he should not win but cannot lose either.
28. Kd2 Kd7 29. Kd3 a5 30. Kc4 Kc6 31. g4!
This move gains several tempi for White by immobilizing Black's kingside pawns, and also sets up an eventual pawn breakthrough.
31... a6 32. h3 h6??
[Matt Bengtson and I analyzed this ending and concluded that this is the losing move. The basic reason is that White cannot break through on the Kingside if this pawn remains on h7, but the additional g4-g5 lever made available here will force a passed h-pawn. In view of this reason, the move would be bad even if it didn't lose by force: Black has increased his losing chances significantly by giving White this possibility. A simple road to the draw was 32... Kd6 33. Kd4 Kc6 34. h4 Kb5, when White must play 35. Kc3, as there is no forced win on the Kingside with this pawn structure. --JLM]
33. f3 Kb6 34. f4
[I think that White should push the h-pawn first so that Black's possible ...f5 is less forcing. E.g., 34. h4 f5 35. Kd5 Kb5 36. g5 h5 37. Ke5 Kb4 38. Kf6 Kb3 39. Kxf7 Kxb2 40. g6 a4 41. g7 a3 42. g8=Q a2 43. Qg2+, +-. However, the text wins also. --JLM]
34... Kc6 35. h4 Kb6?
[Note that White has carefully used his extra pawn tempi to ensure that the black king is on b6, not c6, when White initiates the kingside pawn break. This means that the White king will have the valuable option, in some variations, of bypassing the black king and reaching the kingside. By the way, 35... f5 36. gxf5 should also lead to a draw, but Black has losing chances and must play precisely. [Actually, White wins after 35... f5 36. g5 h5 37. Kd4 a4 (or 37... Kd6 38. b3!) 38. Ke5. A much better try is the obvious 35... Kd6, when White wins with some accurate play: 36. h5 Ke6 (forced in view of the threat 37. g5) 37. Kd4 and now:
(a) 37... f5?! 38. g5 hxg5 39. fxg5 with an easy win.
(b) 37... a4 (it is in Black's interest to keep his doubled a-pawns as far apart as possible so that if White decides to queen his b-pawn, it will take longer to capture both a-pawns) 38. Ke3! Ke7 39. Ke4! Ke6 40. Kd4! (White can triangulate to lose the move, since Black's King cannot go to the d-file in view of the g5 break) 40... Ke7 41. Kd5 Kf8 42. f5! Ke8 43. Kc6 and White's b-pawn will queen long before Black's King penetrates to the Kingside. --JLM]
36. g5?!
[A much clearer win is 36. h5! followed by 37. g5, and White queens first. --JLM]
36... fxg5
[White also wins after 36... hxg5?! 37. h5 g4 38. Kd4! or 36... h5 (the best chance) 37. Kd5 (not 37. f5? fxg5 38. hxg5 h4 =) 37... Kc7 (or 37... Kb5 38. Kd6 Kb4 39. Ke7 +-) 38. Kc5 (and not 38. f5? fxg5 39. hxg5 Kd7 40. Ke5 h4 41. Kf4 Kd6 =) 38... a4 39. Kb4 fxg5 40. fxg5 Kd6 41. Kxa4, and the b-pawn queens before Black's h-pawn reaches the 7th rank. --JLM]
37. fxg5 h5 (37... hxg5 38. h5 g4 39. Kd4, +-) 38. Kd5
Now the white king transfers to the kingside and White wins -- possible because of White's use of tempo moves to place the black king on b6 rather than c6.
38... Kb5 39. Ke5 Kb4 40. Kf6 Kb3 41. Kxf7 Kxb2 42. g6 h4 43. g7 h3 44. g8=Q h2 45. Qg2+ Kb1 46. Qf1+ Kb2 47. Qe2+ Kb1 48. Qxa6 a1=Q 49. Qxa1 Kxa1 50. Kg6 Kb2 51. Kxh5
[1-0]

Round 5, Board 4
White: Jean Efproxiadis (1600)
Black: John Bronsteen (2067)
Guioco Pianissimo

1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. d3 Bc5 5. Nc3 d6 6. h3 Na5 7. Bb3 Nxb3 8. axb3 a6 9. Qe2 Be6 10. Be3 Be3 11. Qe3 Qe7 = 12. Ng5?! O-O 13. f4 Nh5 14. f5 Nf4! 15. Qg3 Bd7 16. Ne2 Nxe2 17. Kxe2 f6 18. Nf3 d5! -/+ 19. Qh4 dxe4 20. dxe4?! Qc5 21. Rac1 Rad8 22. Nd2 Bb5+ 23. c4 Qd4 24. Rc2 Qd3+ 25. Kd1 Bc6 26. Kc1 Rd4 27. Qf2 Rfd8 28. Rd1 Bxe4 29. Rc3 Qb1+!
[0-1]

ROUND 6

(3) BMCC				(5) Harvard University

GM Gennadi Sagalchik (2568)	0-1	Matthew Bengtson (2242)
Nikola Duravcevic (2230)	1/2	Jeremy Martin (2212)
Jeffrey Mitchell (2158)		1-0	Timothy Hanke (2122)
Oleg Shalumov (1930)		1-0	John Bronsteen (2067)