chess online

I don't know so much. I reckon Black is back in the game, here. White should not have exchanged rooks.

After 39.Kc5, Black could have got a counterattack by 39...g5! with this possible follow-up: 40.fxg5 Ke6 41.Ne2 Kxe5 42.Kb6 Nd5ch 43.Kxb7 Nf4 44.Nc1! Kd4 45.Kxa6 Ke3 46.Kb5! Ne6 47.Kc4 f4 48.Nd3 Nxg5 49.b4 f3 50.b5 Nf7 51.b6 Nd6ch 52.Kc3 Nb5ch and a draw seems likely. It is probable that White can improve on this, but chances are that Black can match those iimprovements. When defending difficult endgames, it is a very good idea to keep an eye out for possible counter-attacks, especially when the enemy is heavily committed to attack in one sector of the board.

Why not take the center passer with Nxe5? This way the knight and king can work together on the b and a pawns. If black can trade N for two pawns a draw by insufficient material can be accomplished. Also the knight is a short range fighter being split between two weakness will make for a harder game.

I think Black made a very slight slip that cost a vital tempo. Instead of 49...h2, ...Nxb2 ought to have been played, forcing White to take on h3. There was not a lot White could have done about it, so 49...Nxb2! 50.Kxh3 Nc4 51.Nxa6 Nxe5 (or maybe ....Ke6) and White will have the Devil's own job pushing home his a-pawn. It seems even at this late stage likely that the game could have ended in a draw. EDIT: I REVISIT THIS a few moves down the track, but I'll state it simply here: 49...h2! was a fine move and had she followed up with zwarwiz's suggestion 50...Nxe5!! the game might well have ended in a draw: See below. Jo, your play has been vindicated thus far at any rate: your counterattack idea was good enough to have secured the draw, but for the wrong capture at move 50. I have an idea that in the same situation I would have made the same capture, but a move earlier. On the other hand 49...Nxe5 ought also to have held comfortably, so ought to have got the preference a move earlier as well. It is remarkable that the split pawns should prove stronger than the joined ones.

This has cost Black far too much time. Is there any time since move 49 that Black could have improved her chances? Possibly, having wasted the tempo by playing 49...h2, Black might have been better off with zwarwiz's suggestion: 50...Nxe5!!, with the idea 52.Nxa6 Nd3 53.b4 Kd6! 54.b5 (to keep the Black King at bay) 54...Nb2! (preventing a2-a4 for just long enough...) 55.Kg3 Na4! 56.Kf4 Nc3 and White can not hold the pawns. I take it all back - 49...h2 was not a mistake, but a very fine move! The error was Black's 50th.

One tempo away from the draw

Well play dmaestro! Good annotation  .

Nice endgame!

It turns out that the endgame was a deal closer than might have been expected. Jo was in the fight at least to draw until move 50, but after that was too far behind to make good the deficit. For mine, dmaestro deserves credit for the win: keeping up the psychological pressure throughout the game, and maintaining a sufficient edge to persuade Black she was fighting to save the draw. But Jo gets some kudos for a tough and tenacious defence that - for one slight error in a complicated endgame situation - fell a whisker short of salvaging a draw. This is what chess should be like.

Nice and interesting endgame.