Casimiro Ballina Amandi (1748) - Spanton (1895)
Spanish Berlin
1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.d3 Bc5 5.c3 d5!?
A sharp line that became possible once it was realised 6.Nxe5 0-0 7.Nxc6 bxc6 8.Bxc6, which at first glance looks very good for White, runs into 8...Bxf2+! 9.Kxf2 Ng4+, as seems to have been first played in 2017 in Dadi Ómarsson (2266) - Einar Hjalti Jensson (2323), Icelandic Team Championship (Reykjavik) 2018. The game continued 10.Kf1? (10.Kg1 is probably best, although 10...Qf6 wins the white light-square bishop) Qf6+ 11.Qf3 Qxc6 12.exd5 Qg6 13.h3 Ne5 14.Qe4 Nxd3 15.Nd2 Qg3 16.Qe2 Bg4!? 0-1.
6.0-0!?
The mainline in ChessBase's 2022 Mega database runs 6.exd5 Qxd5 7.Bc4 Qd6 8.b4 Bb6 9.a4, when White is slightly better, according to Stockfish15, although Komodo13.02 reckons the position is completely equal.
6...0-0
David Navara (2726) - Alexander Grischuk (2785), European Club Cup (Antalya, Turkey) 2017, went 6...dxe4 7.Nxe5 0-0 8.Bxc6 bxc6, transposing to a position known since at least 1862. The engines reckon it is equal, and the game was drawn in 30 moves.
7.d4?!
Lots of moves have been tried here by grandmasters, including the text, but the engines prefer 7.Nbd2 or 7.exd5.
7...exd4?!
Better, according to the engines, is 7...dxe4, as played by Adolf Anderssen in a friendly game in Breslau, Germany (now Wrocław, Poland) in 1877.
8.cxd4?!
Aleksandr Galkin (2590) - Luc Winants (2016), Playchess.com Blitz, saw 8.e5 Ne4 9.cxd4 Bb6, transposing to a position known since at least the 1846 Staunton-Horwitz match, but which more commonly arises from the Giuoco Piano or the Scotch Gambit. The engines reckon the position is equal, although Winants won in 27 moves.
8...dxe4 9.Bxc6!? bxc6
The engines prefer 9...Bd6!?, eg 10.Bxe4 Nxe4 11.Re1 Re8, when they reckon Black is at least slightly better.
10.dxc5 Qxd1 11.Rxd1 exf3 12.Nc3!?
The engines slightly prefer 12.gxf3, but like Black after 12...Nd5.
12...fxg2 13.Kxg2 Be6 14.Bf4
 |
| The opening is over (both players have connected rooks) - how would you assess the coming middlegame? |
*****
*****
*****
*****
Black is, at least temporarily a pawn up, but opposite-coloured bishops give middlegame attacking chances, while also boosting the odds of an ending being drawn. The engines reckon Black has the upper hand.
14...Rab8
Stockfish15 gives the passive-looking 14...Rfc8!?, while Komodo13.02 prefers 14...Rfb8!?
15.b3 Bd5+!?
The engines dislike this, preferring 15.Nd5, 15.Rfc8!? or 15.Rb7
16.f3!?
16.f3!?
Self-pinning, but the engines reckon this is best.
16...Rbc8 17.Rd4 Nh5 18.Nxd5!? cxd5 19.Re1 Nxf4+ 20.Rxf4 Rfe8!?
This is Stockfish15's top choice for a short while, but the engines come to agree 20...c6 is best, albeit they reckon the position is completely equal after 21.Re7.
21.Kf2?!
The engines are OK with this, but probably better is 21.Rxe8+ Rxe8 22.Ra4, when the passive 22...Ra8?! gives White strong queenside play despite being a pawn down. Almost certainly better is 22...a5!? 23.Rxa5 d4 with an unclear position that Stockfish15 reckons is completely equal, but that Komodo13.02 reckons slightly favours White.
21...Rxe1 22.Kxe1
 |
| How would you assess this rook-and-pawn ending? |
*****
*****
*****
*****
Black is a pawn up and will have a protected passer after ...c6, but the black c and a pawns are weak. The position is completely equal, according to the engines.
22...c6 23.Ra4 Rc7
22...c6 23.Ra4 Rc7
The black rook is passive but not as bad as in the line given at move 21, where it ends up on a8.
24.Kd2 Kf8 25.Kd3 Ke7 26.Rh4 h6
 |
| White to play and draw |
*****
*****
*****
*****
27.b4?
This arguably natural-looking move loses, whereas 27.Rb4, taking what for White is effectively an open file, and 27.Ra4, pressuring a7, seem to draw. The point seems to be that the text lets the Black rook become active, which, combined with Black's extra (passed) pawn, gives a decisive advantage.
27...Kf6?!
Almost certainly better is immediately activating the rook with 27...Rb7.
28.a4?!
Probably better is 28.b5!? cxb5 29.Kd4 Ke6 30.f4, when Black is two pawns up but White has active play, although the engines reckon Black is at least slightly better.
28...Ke5 29.Rd4?!
The rook is not very active here. The engines suggest 29.f4+!?, when Stockfish15 reckons Black is winning but Komodo13.02 calls the game completely equal! I have not got time to look into this properly now but plan to return to it later.
29...a6 30.Kc3 f5?
Black is only slightly better after this, according to the engines, which reckon 30...Rb7 and 30...g5, among other moves, leave Black well on top.
31.Rh4?
Missing a chance to block the kingside with 31.f4+ and 32.h4, after which it seems Black cannot win.
31...Re7 32.b5!?
This does not work, although, once shown the move, it becomes Komodo13.02's top choice.
32...axb5 33.axb5 cxb5 34.Kb4 d4
The d pawn cannot be stopped.
The game finished:
35.Kxb5 d3 36.Kc4 Rd7 37.c6 d2 0-1
No comments:
Post a Comment