 |
| Carving at Ferry Meadows |
Spanton (1944 ECF/2035 Fide) - Alan Hall (1689 ECF/1765 Fide)Petrov Defence
1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.c4!?
This is less popular than d3, Qe2, Nc3 and especially d4, but scores a better percentage than all of them in ChessBase's 2026 Mega database.
5...Be7 6.d4 0-0 7.Bd3
 |
| How should Black proceed? |
*****
*****
*****
*****
*****
7...Nf6
This is most common in Mega26, but 7...d5, supporting the knight in traditional Petrov-style, has also been played. However, after 8.0-0, reaching a position occurring 996 times in the database, the best continuation is 8...Nf6!?, according to Stockfish19 and Dragon1.
8.0-0
 |
| Now both sides have castled, how would you assess the game? |
*****
*****
*****
*****
*****
White's extra space gives a slight edge, according to the engines.
8...Nc6 9.Nc3 Bg4 10.Be3 Re8 11.h3 Bh5 12.g4!?
This apparent-novelty is liked by the engines, but they also like the known 12.Re1.
12...Bg6 13.Bxg6 hxg6 14.Qd3
The engines prefer 14.Bg5 or 14.Re1.
14...Qd7
 |
| How should White meet Black's threat to capture on g4? |
*****
*****
*****
*****
*****
15.Nh2?!
Probably too passive, especially when there are simple remedies such as 15.Bf4 and 15.d5.
15...d5!?
The thematic ...d5 thrust in the Petrov, and here it gives Black the upper hand, according to the engines.
16.c5 Bf8?!
Almost certainly too slow, whereas the immediate 16...b6, a typical reaction when White pushes on with c5, seems strong. Note, however, that 16...Bxc5!? 17.dxc5 d4 can be met by 18.R(either)d1, after which Black has at best a slight edge, according to the engines.
17.Bf4 b6 18.cxb6 axb6 19.Rfe1?
White should probably play 19.a3, to take control of the b4 square.
19...Bd6?!
Black has an edge after this, but much stronger is 19...Nb4, when the engines reckon best-play runs 20.Qf1!? Ne4 (20...Nc2 21.Rxe8 Qxe8!? 22.Rd1 c5 is also good), with a lot of pressure for Black, eg 21.Red1 c5 22.dxc5 Bxc5 23.Nxe4 Rxe4, after which storm clouds are gathering around White's king.
20.Bg3?
White should probably play 20.Bxd6 Qxd6 21.Nf3, with what the engines reckon is a slight edge for Black.
20...Bxg3 21.fxg3 Rxe1+
Both 21...Nb4 and 21...Qd6 are even stronger, according to the engines.
22.Rxe1 Re8?!
But this throws away most of the rest of Black's advantage, whereas both ...Qd6 and ...Nb4 are still very good for Black, according to the engines.
23.Rxe8+ Qxe8!? 24.Nf3
Probably not 24.g5?! Ne4 25.Nxd5?! Nxg5.
24...Qd7
AH offered a draw.
 |
| How would you assess the position? |
*****
*****
*****
*****
*****
The white king is less safe than its black counterpart, but I thought this should not prove critical with somewhat reduced material, and I hoped to pose problems for Black by creating a queenside passed pawn. However, the engines agree the game is completely equal.
25.Kg2 Kf8 26.Ne5 Qe6 27.Nxc6 Qxc6 28.Qb5 Qxb5
The engines prefer keeping queens on the board with 28...Qe6 or 28...Qd6.
29.Nxb5
 |
| How would you assess this knight-and-pawn ending? |
*****
*****
*****
*****
*****
It is equal, according to the engines, but not simple.
[Indeed, after inputting more moves, and then returning to the diagram position, the engines change their evaluation to White being slightly better.]
29...c6
The engines agree this is best, although Stockfish19 evaluates 29...c5!? as only slightly worse for Black. Dragon1, however, reckons ...c5 loses, or at least gives White the upper hand. Black could also play 29...Ne8?!, but that is probably too passive.
30.Na7 c5 31.Nc8 Nd7
Not 31...cxd4? 32.Kf3!, eg 32...Nd7 33.Ke2 Ke8 34.Kd3 Kd8 35.Nd6 Ke7 36.Nb5, with a winning advantage for White, according to the engines.
32.dxc5
The engines prefer centralising the king with 32.Kf3!?
32...bxc5 33.Kf2 Kf8 34.Ke3 Kd8 35.Nd6 Ke7 36.Nb7 Ke6 37.a4
 |
| How should Black proceed? |
*****
*****
*****
*****
*****
37...Ke5?
The engines reckon the diagram-position is one where centralising the king is not correct. They give 37...d4+ and 37...f6 as equal.
38.a5 d4+ 39.Kd3
39.Kd2! is much preferred by the engines.
39...Kd5 40.b3 Nb8 41.g5 Nc6!?
This may be a mistake, although the engines' evaluations fluctuate, with Dragon1 at one point reckoning the move loses, while at the same time Stockfish19 calls the game equal. The former settles on awarding White the upper hand, while Stockfish19 changes to giving White a slight edge, but finally calls the position completely equal. Nevertheless, the engines agree 41...Kc6 and 41...f6 give complete equality.
42.h4 Nb4+!?
This is an only-move, according to the engines.
43.Kd2
 |
| Black to play and be only slightly worse (Stockfish19), or at any rate to cede no more than the upper hand (Dragon1) |
*****
*****
*****
*****
*****
43...f6
It seems Black had to play 43...Kc6!, when 44.Nd8+ Kb5 45.Nxf7 Nc6!? 46.g4 Kxa5 47.h5 gxh5 48.gxh5 Ne7! holds, according to the engines. Everything else loses fairly trivially, eg 43...Na6 44.Nd8 f6 45.gxf6 gxf6 46.g4, which is similar to the game.
44.gxh6 gxh6 45.g4 Na6 46.h5 gxh5 47.gxh5 Ke6
 |
| White to play and win |
*****
*****
*****
*****
*****
48.Nxc5+?
Sacrificing the knight is the right idea, but this is the wrong way to go about it. Instead, 48.Nd6!? leaves Black helpless, eg 48...f5 49.h6 Kf6 50.Nxf5!, or 48...Ke7 49.h6 Kf8 50.Ne8! Kg8 51,Nxf6+. But other moves also win, including 48.Kd3 and 48.h6.
48...Nxc5 49.b4
 |
| Black to play and draw |
*****
*****
*****
*****
*****
49...Ne4+?
Black holds with 49...Na6 50.b5 Nc7 51.b6 Na6.
50.Kd3 Nd6 51.Kxd4 f5 52.h6 Nf7?
There is more resistance in 52...Kf6 53.Kd5 f4 54.a6! Kg6 55.Kxd6 f3 56.a7 f2 57a8=Q f1=Q, when the queen-and-pawn ending is won for White, but practice does not always follow theory.
53.a6 1-0
