An oft-quoted maxim of Wilhelm Steinitz is that “the king must be treated as a strong piece both for attack and defence”. While sceptical about the truth of this statement while queens are on the board, I thought of it more than once while analysing a series of tactical queenless middlegame from a book “Damen sind Luxus” (“Queens are a luxury”) by W. Heidenfeld which my friend Steve Giddins leant to me over Christmas. In all of these queenless middlegames, the king’s ability to ride blows and fearlessly meet his foes head on would have made Steinitz proud! In this blog, we take a look at an example of Emmanuel Lasker’s amazing resilience. The game is available in playable form at http://cloudserver.chessbase.com/MTIyMTYx/replay.html
Spielmann,Rudolf – Lasker,Emanuel
Moscow International 1935
1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Nf6 5.Nc3 Bb4 6.Nxc6 bxc6 7.Bd3 d5 8.exd5 Qe7+ 9.Qe2 Qxe2+ 10.Kxe2 cxd5 11.Nb5 Kd8 12.Rd1 c6 13.c3 Re8+ 14.Kf1 Bf8 15.Nd4 Kc7 16.Bf4+ Kb6 17.a4 a5 18.b4 axb4 19.a5+ Kb7 20.cxb4 Ne4 21.Nxc6
Dramatic, but White had a much simpler way to win in 21.Rdc1 Bd7 22.a6+ Kb6 23.Be3 Kc7 24.Nxc6 Bxc6 25.b5.
I haven’t been able to find any other annotations of this game except for the Kmoch annotations in the tournament book quoted (and improved on) by Heidenfeld. John Nunn doesn’t examine the game in his wonderful examination of Lasker’s Games (“John Nunn’s Chess Course”) and it doesn’t make it into Verkhovsky’s “Draw!: The Art of the Half-Point in Chess”
21…Kxc6 22.Rac1+ Nc5 (22…Kd7 23.Bb5+; 22…Kb7 23.Rc7+) 23.Be3 wins
22.Be5 is even stronger according to Komodo 22…Kxc6 23.Rac1+ Kd7 24.Bb5+
22…dxe4 23.Nd8+ Ka6 24.Bxg5
23.Nd8+ Ka6 24.Bxg5
was suggested as winning at the time, but Heidenfeld casts doubt on this assessment, pointing out Black’s strongest defence. Time to analyse!
stops White from playing Rc1–c7+xf7 as after 24…gxf4, but releases Black access to White’s back rank which removes some of Black’s counter threats against the White king
25.b5+ Ka7 26.Rc1 Kb8
a1) 27.b6 Ba6+ 28.Kg1 Rc4!
The defensive scheme pointed out by Heidenfeld is surprisingly effective
29.Rxc4 Bxc4 30.Rxg5
I thought this might be very strong for White, but Komodo is unimpressed.
30…Ra6 31.Rg8 Ba3
was my desperate defensive setup. Komodo thinks equality has been reached!
32.Nxf7+ Kb7 33.Nd8+ Kc8 34.Rh8 Bb2 35.Re8 Ba3
My first choice. The threat is Nc6+–e7+ winning the bishop on c8.
Possible as Black no longer has any threats against White’s back rank. Black is defenceless.
On to the strongest defence:
25.b5+ Ka7 26.Rc1
Threatening Rc7+xf7 and then a later Nc6+
b1) 26…Ba3 27.Rc7+ Kb8 28.Rxf7 Rxa5 29.Nc6+ Ka8 30.Rd8 Komodo (30.Nxa5 Bb4 31.f3 Re1+ 32.Kf2 Bxa5 33.b6 also wins) ;
b2) 26…f3 27.Rc7+ Kb8 28.Rxf7 Rxa5 29.Nc6+ Ka8 30.Nxa5 Bb4 31.Rd1 Bxa5 32.Ra1 was my win. Komodo goes just a bit further 32…Re1+ 33.Rxe1 fxg2+ 34.Kxg2 Bxe1 35.Rxh7;
The best defence as pointed out by Heidenfeld.
Threatening primarily Rxc8+ and b7. b7 followed by a pin on the b-file is also a dangerous option
27…Ba6+ 28.Kg1 Rc4
Heidenfeld’s resource again which annoyed me just as much the second time around! Actually Komodo says the position is now equal
29.Rxc4 Bxc4 30.Rd7 Bb5 31.Rb7+ Kc8 32.Ra7 Kb8 (32…Rxa7 33.bxa7 wins!) 33.Rb7+ Kc8 34.Ra7
My line, and Komodo’s main line too!;
b321) 27…Ra4 28.a6 (28.b6 I also thought this was good now, but it doesn’t address the main problem… 28…Ba6+ 29.Ke1 Rc4 Black can still block the 4th rank) ;
b322) 27…Re3 28.b6 was the idea 28…Ba6+ 29.Kg1 No …Rc4 blocking option now we have forced the rook off the 4th rank 29…Bb7 30.Rd7 My line. Komodo goes a bit further 30…Re7 31.Rxe7 Bxe7 32.Nxb7 Kxb7 33.Rc7+
b323) 27…Re8 28.a6 Ka7 29.Rc7+ Kb8 30.Rxf7 Rxa6 31.bxa6 Bxa6+ 32.Kf2 Re2+ 33.Kf1 Re8+ is a draw by repetition;
Komodo’s #1 preference. The pawn on a6 restricts the king on b8 which makes Nc6+ suddenly very dangerous
27…Re8 The only way really to defend against the threat of Nc6+–e7
27…Ba3 28.Rc3 Rxa6 29.Rxc8+
28.g4 f3 29.h3 Be6 30.Nxe6 fxe6 31.Rd3 is Komodo’s #1 line. White is better, but Black is still fighting. All in all, 24.Rxd5 offers excellent chances but it’s nothing like a clear forced win.
Back to Spielmann’s very practical choice 24.Bxg5
Lasker was renowned as a fantastic defender. So far he’s been outclassed and has simply had to roll with the punches. Typically however, he’s emerged into a position with defensive possibilities. After White struck with 21.Nxc6, it seemed that Black’s position would be blown away, but Lasker has managed to hold on to some of the basis of his position, albeit in a somewhat tattered state. First of all, the downside: Black has lost 2 pawns, and his remaining pawns are all isolated. His king is also denuded of pawn cover which makes it vulnerable to attack along the open c-file. On the positive side, Black has maintained his central pawn on d5 (which blocks an entry file for the White rooks) and has gained the advantage of the 2 bishops. White can capture one of Black’s bishops, but this would improve Black’s pawn structure greatly so White needs to consider the exchange carefully. Moreover, after Nxe6 …fxe6, while White’s (extra) queenside pawns are on dark-squares and form easy targets for the Black pieces, White’s remaining minor piece (the dark-squared bishop) has no targets on the kingside!
How would White proceed then? My favourite idea was to destroy Black’s advantage of the 2 bishops and then to invest one of my extra pawns to create fresh weaknesses in Black’s kingside pawn structure. This is not immediately winning, but it takes play into an ending where White is a pawn up with a better pawn structure and – at the very least – equivalent activity. Reasonable technique should be sufficient in such circumstances to finish off the game. The game continuation is still winning of course but requires White to maintain a higher level of calculation.
25.Nxe6 fxe6 26.Rdc1
Komodo’s 2nd best move was my line. 26…Bxb4 (26…Rxe1+ 27.Rxe1 Bxb4 (27…Re8 28.Bd2 White is excellently organised now and he has exchanged off Black’s most active rook on e4) 28.Rxe6+ Kxa5 29.Re5 wins a second pawn) 27.Rxe4 dxe4 28.Be3 with Rc1 to follow should be enough to win in the long run.
26…Rc4 27.Bd2 Rb8 28.Re1 Rc6 29.Rab1 Kb5 30.f4 exerts control over Black’s position. 2 extra pawns, the rook on e4 has been sent back to c6 and a new kingside weakness (e6) has been created and fixed.
27.Rc6+ Kb5 28.Rb6+ Kc4 29.Rb1 is Komodo’s cleanest win. These options seem like the best technique to me. The game gets gradually more complicated…
To stop a knight establishing itself on d4
looking for Rc5 followed by b4–b5+
Blocking White’s threat and preparing …Kb5 challenging the knight on c6 and thus weakening White’s defence of his b4 pawn.
Looking to cement White’s grasp of d4 without allowing Black any destabilising factors in the position. Lasker continues to shake and pull at White’s position to stop White from achieving the solid structure he is seeking. He sends his king amongst White’s pawns and introduces the spectre of Black reestablishing material equality by grabbing both White’s queenside pawns. He makes a number of tactical mistakes along the way, but the approach is spot on. He plays with maximum aggression and gives White something to lose if his calculation fails.
It’s around here by the way that I understood Stockfish’s very sensible preference for 24.Be3 earlier on, leaving the pawn on g5 to focus on the essential business of getting the b-pawn moving 24…Be6 25.Nc6 Bg7 26.Rab1 Kb5 27.Na7+ Kc4 28.Bb6 and b5 will follow. Still looks messy though!
27.Rxc4 dxc4 28.Nd4 Bxd4 29.Rxd4 is also good for White but gives Black some hopes of counterplay with 29…c3
Very dangerous tactically for Black, but a positional risk as the knight on a7 is poorly placed. Things have to work now or White might end up losing a pawn with offside pieces if one line of White’s calculation fails.
28.Nd4+ Bxd4 29.Rxd4 Rxd4 30.Bxd4 Kxb4 The first moment that White has to evaluate: is this the right way to give a pawn back to force Black into a cheerless endgame? Considering Black’s split kingside pawns (and thus White’s ability to invade there with his king) this should provide White with excellent winning chances. It still requires a fair amount of technical work and White avoided a similar decision on the 25th move. It is thus not surprising that Spielmann continues aggressive tactical play.
28…Ka4 was better according to Komodo. The text should lose by force.
a) 29…Ka4 30.Rd2 Rc3 31.Ra2+ (31.Nb5 (my line) also wins but much less convincingly) 31…Ra3 32.Bc5 Komodo;
b) 29…Kxa5 30.Rd2 Ra4 (30…Bd7 31.Rxd5+) 31.Nc6+ Ka6 32.Rb6#;
30.Rd2 threatening Nb5+ (30.Nb5+ Ka4 31.Rd2 Rb4 keeps Black alive!) 30…Rxa7 31.Bxa7 should win for White
A mistake in return. Komodo thought Black if anything was better after 29…Bf5! The players trade mistakes now until the 40th move. Quite appropriately however, Black’s active king saves the day!
30.Rb1+ Rb3 31.Nc6+ Ka4 32.Bd4 Rxb1 33.Rxb1 Bxd4 34.Nxd4 Ra6 35.Ra1+ Kb4 36.Ke2 Bd7 37.Nc2+ Kc3 38.Ne3 Bb5+ 39.Ke1 d4 40.Rc1+ Kd3 41.Rd1+
A great save by Lasker!
While searching for Moscow 1935 on the Internet, I came across this interesting (Dutch language) article (from which the accompanying picture is taken) http://oudzuylenutrecht.nl/spielmann-lasker-1935/
Dutch Grandmaster Hans Bouwmeester takes a look at this game too and suggests some alternative lines to mine.