 |
| Black to move in Sam Coates (1589 Fide) - Spanton (1808 Fide) |
*****
*****
*****
*****
32...Nbxd3! 33.gxf4
Not 33.cxd3?? Rxc1 34.Rxc1 Ne2+ and 35...Nxc1.
33...Nxe1 34.Rxe1
*****
*****
*****
*****
*****
34...exf4?!
Probably better is 34...Rxc2. I was worried about the counterplay White gets with 35.fxe5, but Stockfish14 and Komodo12.1.1 reckon Black has an edge after 34...d3 35.Nf3 a5.
35.Ng5
Materially the position slightly favours Black as a rook and two pawns (7pts) are worth more than two knights (6.5pts), using the average points values worked out by Larry Kaufman using a computer to analyse 300,000 high-level games (see https://www.chess.com/article/view/the-evaluation-of-material-imbalances-by-im-larry-kaufman for details). Here the uncoordinated state of the white pieces should surely also favour Black. However the engines reckon the position is dead-equal. I am not clear why - it is not as if the black pawns are obviously weaker than the white ones.
35...Re8 36.Rd1
A relatively well-known rule-of-thumb in these unbalanced endings is that the player with the sole rook should not allow it to be exchanged for one of the opponent's rooks, even though here this would marginally increase the ratio of his material advantage in pieces from 11.5pt v 10pts (54.8%) to 6.5pts v 5pts (56.5%). The reason is there is a small but significant redundancy in having two rooks.
36...Rxc2 37.Rxd4 Rxb2 38.Rxf4
This looks natural but the engines prefer 38.Rd7+!? After 38...Kg8 39.Nhf3 they reckon Black cannot make progress, eg 39...a5 40.Ra7 a4 41.Nh7 Re6 42.Ra8+ Kg7 (not 42...Kxh7? 43.Ng5+ etc) 43.Ra7 Kg8 (Black gets mated after 43...Kh6?? 44.Nhg5 etc).
38...Re7?!
Passive rook-play is very often a mistake, and this is probably no exception. I was basically defending against shadows. The engines give 38...Rbe2!?, and if 39.Rf7+ Kh6 40.h4 then 40...R2e7 41.Rf6 b4!? 42.Rxa6 Rb8 looks good for Black.
39.Nhf3 a5!?
There does not seem to be anything better, although now White wins a pawn.
40.Nd4 Rb1+ 41.Kg2 Rbe1 42.Nc6 Rd7 43.Nxa5 Re5 44.h4 Rf5 45.Rb4
The engines slightly prefer 45.Re4!? Rd2 46.Nh3, when Komodo12.1.1 gives White a slight edge, although Stockfish14 disagrees.
45...Ra7 46.Nc6 Ra4 47.Ne6+
The engines prefer keeping rooks on with, say, 47.Rb3!?, even though this allows 47...Rxh4.
47...Kf6 48.Ned4!? Rxb4
Black gets a slight edge with 48...Rd5!?, according to the engines.
49.Nxb4 Rc5 50.Kg3?
After making this move, which is a mistake, SC pressed his clock and very soon offered a draw.
The engines reckon 50.Na6 leaves Black with only a slight edge, while 50.f4! maintains equality. If after the latter, Black plays, as in the game, 50...Rc4, the engines give 51.Nd5+ Kf7 52.Nxb5.
50...Rc4 51.Nxb5?
Now 51.Kd5+ fails to the simple 51...Ke5, preventing which was the point of playing 50.f4! But better than the text is 51.Ndc2, although after 51...Ke6 the active black king should probably be decisive.
51...Rxb4
This is an easy win for Black, according to Stockfish14, which has Black at +9.79, and also according to Komodo12.1.1 (Black +10.41).
The game finished:
52.Nd6 Rd4 53.Nb5 Rg4+ 54.Kh3 Ke5 55.Nc3 Kf4 56.Nd5+ Kf3 57.Ne3 Re4 58.Nd5 Kxf2 59.Nf6 Re1 60.Kh2 Kf3 61.Nh7 Kg4 62.Nf8 Rde2+ 63.Kg1 Kf5 64.Nd7 Re4 65.Kg2 Rxh4 66.Kg3 Rd4 67.Nf8 Rd3+ 68.Kg2 Rd8 0-1
No comments:
Post a Comment