The Gentleman or the Tireder?

Except for a few slips, my opponent consistently outplayed me in the following game, which is especially notable given that he used up much of his initial 2 minutes quickly and played the large majority of his remaining moves on each 12-second increment.

Just when he reached a "won" endgame, he offered me a draw. Was he simply tired from playing so long on the knife's edge of time? Or had he seen what was to come, but in a sporting way rewarded my resistance with a half-point?

 perrypawnpusher - dogofthesouth
blitz 2 12, FICS, 2010

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Kf8

7.Qxe5 Bd6

This move is better than it looks at first glance, and forces White's Queen to abandon the center or abandon the board.

8.Qf5+

Also seen has been 8.Qc3, in perrypawnpusher - Lark, blitz, FICS, 2009 (1-0, 30); 8.Qa5, in billwall - firewine, Chess.com, 2010 (1-0, 53); and 8.Qh5 in blackburne - Whiterose, ChessWorld, 2004 (1-0, 11).

8...Qf6 9.Qxf6+ Nxf6 10.Nc3

I was not happy exchanging Queens, but sometimes it has to happen. I continued to play confidently and quickly, and a moment of uncertainty on behalf of my opponent (plus his use of time) was my reward.

Also played was 10.d3, in perrypawnpusher - LeiCar, blitz, FICS, 2010 (1-0, 34).

10...Be5 11.0-0 Bxc3 12.bxc3 d6

The pawn on e4 was free for the taking. I should have protected it with 11.d3. My opponent, however, was suspicious and gave me time to recover.

13.d3 h6

Louis Marin ground down his opponent after 13...Kf7, mrjoker - PhlebasP, ICC, 2008 (1-0, 34).

14.f4 Ke7 15.Ba3 Bd7 16.e5 Nd5 17.c4 Ne3 18.Rf2 Bc6

Black's last move (instead of 18...c5, blocking the a3-f8 diagonal) is part of his plan to put pressure on g2, but proves to be inaccurate and allows White to pull even.

19.Re1 Nxg2 20.Rxg2 Bxg2 21.exd6+ Kf7 22.Re7+ Kg6 23.Kxg2 cxd6

Black is up the exchange for a pawn (soon two), but with the proper follow-up White can now develop winning chances.

24.Re6+

This is not the right way, however. After 24.Bxd6, White's Rook can threaten Black's pawns at b7 and g7 (especially after Be5). When one of them falls, three pawns would be great compensation for the exchange.

24...Kf5 25.Rxd6 Rad8 26.Rxd8 Rxd8

Over the next 30 moves White's compensation falls, like autumn leaves from a tree. My opponent's play is not always the most accurate, but, given his time constraints, he did well; and certainly better than I did

27.Kf3 Rd7 28.Bb2 b6 29.Be5 Re7 30.c3 g5 31.d4 gxf4 32.Bxf4 h5 33.Be3 Re4 34.c5 bxc5 35.dxc5 a5

Black could have saved himself a lot of trouble by playing 25...Ra4 here. His a-pawn would then Queen, while my forward c-pawn would not.

36.Bd4 Rg4 37.c6 Rg8 38.c7 a4 39.a3 Re8 40.c4 Ke6 41.Kg3 Kf5 42.h4 Ke4 43.Bb6 Rc8

44.Ba5 Kd4 45.Kf4 Kxc4 46.Kg5 Kb3 47.Bb4

White had a clearer chance to play for the draw here with 47.Kf6: 47...Kxa3 48.Ke7 Kb3 49.Kd7 a3 50.Bb6 Rg8 51.c8/Q Rxc8 52.Kxc8 a2 53.Bd4 Kc4! Black can now run his King over to attack White's h-pawn, causing the Bishop to protect it; play ...a2-a1/Q, forcing the Bishop to capture the Queen and un-protect the h-pawn; and capture the h-pawn; but the Bishop will stop the final Black pawn from Queening.

I was thinking, instead, that getting another Black pawn off of the board was the key to my survival. It might have been, but not the way that I played it.

47...Rxc7 48.Kxh5 Kc4

It is possible that 48...Rb7 is stronger, as it implements the threat of capturing White's Bishop and thus freeing Black's a-pawn to advance and Queen. White would have to try something like 49.Bf8. I am not sure how much more Black could achieve by chasing the Bishop here and there.

49.Kg6 Kd5 50.h5 Rc6+ 51.Kg5 Ke6

White must now play 52.Kg6 to keep Black's King away from the h-pawn and maintain drawing chances.

52.h6 Kf7 53.Kh5 Kg8 54.Be7 Kh7

Now the h-pawn is doomed, and with it, quite probably, my game.

55.Bf8 Rc3 56.Kg5 Rf3 57.Bb4 Rg3+ 58.Kh5 Rh3+ 59.Kg5 Rg3+

At this point my opponent offered a Draw, and with a sense of relief I accepted.

The question in my mind was "Didn't Black have a win after 59...Rxh6"?

It turns out that the people who have created Rybka have a website (http://chessok.com/?page_id=361) where there is a 6-man Nalimov Endgame Tablebase which gives the answer: after 59...Rxh6 Black would have a mate in at most 48 more moves: 60.Kf5 Rh5+ 61.Ke4 Rh4+ 62.Kd3 Rg4 63.Bc5 Kg8 64.Bd6 Kf7 65.Bb4 Rh4 66.Bc5 Ke6 67.Bf8 Rh8 68.Bc5 Kd5 69.Bb4 Rh3+ 70.Kc2 Rg3 71.Be7 Rf3 72.Bb4 Kc4 73.Bd6 Rf2+ 74.Kb1 Kb3 75.Kc1 Rc2+ 76.Kd1 Rc8 77.Be7 Re8 78.Bb4 Kb2 79.Bc5 Rd8+ 80.Ke2 Kb3 81.Ke3 Rc8 82.Bb4 Rb8 83.Bc5 Kc4 84.Bd6 Re8+ 85.Kf2 Re6 86.Bb4 Rb6 87.Be7 Rb7 88.Bd6 Kd5 89.Bf4 Rb3 90.Bc1 Rc3 91.Bg5 Rxa3 92.Be7 Rd3 93.Bf6 a3 94.Ba1 Rd1 95.Bc3 Kc4 96.Bf6 a2 97.Ke3 Rd3+ 98.Ke4 Rc3 99.Bg7 a1Q 100.Bxc3 Qxc3 101.Kf5 Qd4 102.Kg6 Qe5 103.Kh7 Qg3 104.Kh8 Kd5 105.Kh7 Ke6 106.Kh8 Kf7 107.Kh7 Qh4#

Whew! That line is very complicated and possibly not worth the effort finding in a casual blitz game. In all fairness, though, White would be more likely to err and lose while Black would be risking very little.