David Flynn (1867 ECF/1816 Fide) - Spanton (1944 ECF/1908 Fide)
Sicilian Hyper-Accelerated Dragon
1.e4 c5 2.Nc3 g6 3.d4 cxd4 4.Qxd4 Nf6 5.Nf3 Nc6 6.Qa4 d6 7.e5 dxe5 8.Nxe5
 |
| This position occurs 1,784 times in ChessBase's 2026 Mega database - how should Black proceed? |
*****
*****
*****
*****
8...Bg7!?
Offering at least one pawn. The main line in Mega26 runs 8...Bd7 9.Nxd7 Qxd7 10.Be3 Bg7 11.Rd1 Qc8!? 12.Be2 0-0 13.0-0, when White's bishop-pair and lead in development outweigh Black having the only pawn on the two centre files, according to Stockfish17.1 and Dragon1.
9.Bb5
This is the most popular move in Mega26. It is also marginally the top choice of Dragon1, and will quite possibly give White the option of grabbing two pawns. Stockfish17.1, however, for quite some time marginally prefers 9.Nxc6, before coming to see the two moves as more-or-less exactly equal. After 9.Nxc6 bxc6 10.Qxc6 Bd7, Stockfish17.1 reckons Black has sufficient compensation for a pawn, but Dragon1 gives White a slight edge.
9...0-0 10.Nxc6 bxc6 11.Bxc6 Rb8!?
Offering a second pawn. Instead 11...Bd7 12.Bxd7 Nxd7!? divides the engines as before, with Stockfish17.1 calling the position equal, while Dragon1 slightly prefers White. At move 12, instead of capturing on d7, White can also castle.
 |
| What should White play? |
*****
*****
*****
*****
12.0-0
This is the engines' recommendation. After 12.Qxa7!? they suggest an apparent-novelty in 12...Qd6!? (12...Bf5 is known), when 13.Bf3 Ba6 gives a sharp position with what the engines reckon are equal chances.
12...Bf5
More popular, and possibly marginally better, is 12...Qc7, while 12...a5!? has scored very well in practice.
13.Rd1 Qc7 14.Bf3 Rfc8!?
This may be new, and is the top choice of Dragon1, whereas Stcokfish17.1 prefers the known 14...a5.
15.Re1!?
Presumably in the hope of playing Bf4 without Black being able to reply ...e5. The engines suggest 15.h3 or 15.Rb1, but are fine with the text.
15...e5 16.Ne4 Bd7
The engines reckon Black should regain the sacrificed pawn immediately with 16...Nxe4 17.Bxe4 Bxe4 18.Qxe4 Qxc2.
17.Nxf6+ Bxf6 18.Qa6!? Rb6 19.Qe2 Qxc2 20.Qxc2 Rxc2
 |
| The pawn has been regained anyway, and the game is equal, according to the engines |
*****
*****
*****
*****
21.Be4 Rc7 22.b3 Be6 23.Bd2 Rd6
Perhaps 23...Ra6!?, with queenside pressure, is a tad better.
24.Be3 Bg7 25.Rac1 Rxc1 26.Rxc1 Ra6 27.Bb1 Ra5?!
The engines suggest 27...Bf8, 27...f5 or 27...e4!?
28.Rc5 Rxc5 29.Bxc5 a6 30.Bd3 Bc8 31.f3
Trying to centralise the king as quickly as possible, but 31.Bd6 gives White at least a slight edge, according to the engines.
31...Bf8 32.Bb6 Bb4 33.Kf2 Kf8 34.Ke3 f5 35.Bc4 Bb7 36.g3 Ke7 37.g4!? h6 38.gxf5 gxf5
 |
| Giving Black a 2-1 central pawn-majority seems a little strange, but the engines call the position completely equal |
*****
*****
*****
*****
39.Bd3 Bc8 40.Bc4 h5!? 41.Bd5 Be6 42.Bc6
Exchanging bishops is also OK for White, according to the engines.
42...h4!? 43.Bb7 a5 44.Bc6 Kd6 45.Bb7 Bc5+!?
There seems no other way to make progress. Dragon1 at first reckons the text gives Black the upper hand, but eventually agrees with Stockfish17.1 that the game remains equal.
46.Bxc5+ Kxc5
 |
| How should White continue? |
*****
*****
*****
*****
47.Kd3!?
The engines reckon both 47.f4!? e4 48.Ba6 and the immediate 47.Ba6!? are completely equal. Stockfish17.1 is also fine with the text, at least at first, but Dragon1 instantly calls it losing. Given more time, the engines come to agree it leaves Black slightly better. One point about 47.Ba6!? is that after 47...Kb4 48.Bd3 Ka3 49.Bb5 Kxa2 White has the neat trick 50.Bc4!, with what the engines reckon is complete equality.
47...Kb4
 |
| White seems to have only one drawing move |
*****
*****
*****
*****
48.Kc2
This looks natural, but the engines show White has to play 48.Ke3! Ka3 49.Ba6!?, again relying on the neat trick 49...Kxa2 50.Bc4!
48...Ka3 49.Kb1 Bd7 50.Bd5 Bb5 51.Kc2!?
The best try, but it should not save White.
51...Kb4!?
This may win, but other moves, eg 51...f4!?, are more convincing. However, not 51...Kxa2? 52.b4+.
52.a4?!
The engines suggest 52.Be6 f4 53.Kb2!?, but agree 53...e4!? is one of several winning moves. However, after, 54.Bg4 e3 55.a3+ Kc5 56. Kc3 Bc2 57.b4+ axb4 58.axb4+ Kd6!?, the picture is not as clear as Dargon1's assessment of +3.48 for Black makes it appear. Instead Stockfish17.1's +1.29, ie the upper hand, rather than winning, may be nearer the mark.
52...Bd7 53.Kb2 Kc5 54.Bb7 Be6 55.Kc3 f4!?
This is probably the key move, which has to be played at some point to bring home the full point.
56.Be4 Bd7
Also winning is 56...Bd5!?, but the lines get hairy, eg 57.Bxd5 Kxd5 58.Kd3 h3! 59.Kc3 e4 60.fxe4+ Kxe4 61.Kd2 f3 62.Ke1 Ke3 63.Kf1 f3, when there is no stalemate as White has to play the losing 64.b4. Alternatively, 58.b4 axb4+ 59.Kxb4 Kd4! 60.a5 e4 61.a6 e3 62.a7 e2 also loses as after 63.a8=Q Black queens with check, 63...e1=Q+, and then 64.Kb5 Qe2+ 65.Kb6 Qe6+ 66.Kc7 Qe5+ 67.Kd7 Qd5+ forces queens off, leaving Black with a winning pawn ending. Engines see these lines in the blink of a (human) eye, but over-the-board for club players it is another matter.
57.Bb7
DF offered a draw.
57...Bf5 58.Ba6
Or 58.Ba8 Be6 with ...Bd5 to come. The text is the engines' top choice.
58...e4 59.fxe4 Bxe4 60.Bc4 Bg2 61.Kd3
This lets Black get bishops off, but 61.Kd2 is also hopeless after, for example, 61...Bd5.
61...Bf1+ 62.Ke4 Bxc4 63.bxc4
 |
| Black wins the pawn race |
*****
*****
*****
*****
63...Kxc4 64.Kxf4 Kb4 65.Kg4 Kxa4 66.Kxh4 Kb3 67.Kg5 a4 68.h4 a3 69.h5 a2 70.h6 a1=Q 71.Kg6 Qh8 0-1
No comments:
Post a Comment