Sämisch–Nimzowitsch: the Immortal Zugzwang Game
Aron Nimzowitsch’s 1923 win over Fritz Sämisch is known as the Immortal Zugzwang Game. The final position is remarkable: White has plenty of material and legal moves, but almost every move concedes something decisive.
Complete game score
25 moves. Result: 0-1.
- 1.d4 Nf6 2.c4 e6 3.Nf3 b6 4.g3 Bb7 5.Bg2 Be7 6.Nc3 O-O.
- 7.O-O d5 8.Ne5 c6 9.cxd5 cxd5 10.Bf4 a6 11.Rc1 b5 12.Qb3 Nc6.
- 13.Nxc6 Bxc6 14.h3 Qd7 15.Kh2 Nh5 16.Bd2 f5 17.Qd1 b4 18.Nb1 Bb5.
- 19.Rg1 Bd6 20.e4 fxe4 21.Qxh5 Rxf2 22.Qg5 Raf8 23.Kh1 R8f5 24.Qe3 Bd3.
- 25.Rce1 h6.
What happened in this game?
Black first restrains White’s center, expands with ...b5 and fixes the queenside. The maneuver ...Nh5, ...f5 and ...b4 takes useful squares away from White’s pieces, especially the knight stranded on b1.
After 20.e4 fxe4 and 21.Qxh5, Black does not rush. The rooks invade f2 and f5, the bishop lands on d3, and 25...h6 completes the bind. White resigned because moving a major piece or pawn unguards a critical square or loses material.
Strictly speaking, the finish is not stalemate: White can move. It is a positional zugzwang in a middlegame, created by restriction rather than checks. That distinction makes the game especially valuable for teaching prophylaxis and piece mobility.
Restriction from the opening
The game was played in Copenhagen in March 1923, and Nimzowitsch annotated it himself. His notes describe a plan of taking squares away rather than attacking: ...a6 and ...b5 secured the c4-square, and his knight on c6 headed for it like a ghost, as he put it.
Sämisch spent two tempi exchanging his active e5-knight for it with 13.Nxc6. Stockfish 18 shows how small concessions added up: 9.cxd5, marked dubious in the annotated score, cost almost a pawn against 9.e4, and 14.h3 about half a pawn more.
A piece for the seventh rank
After 17...b4 18.Nb1 the white knight was back on its starting square, and 20.e4? allowed the famous 20...fxe4!!. Black gave a knight for two pawns, the open f-file and a rook on the seventh rank, a trade Nimzowitsch said was based on sober calculation.
Stockfish agrees with both marks. 20.e4 cost more than a pawn compared with 20.Bf3, and 20...fxe4 is far stronger than the alternatives: about −2.6 for White, against about −0.2 after 20...Nxg3.
The bind
After 21.Qxh5 Rxf2, the moves 22.Qg5? and 23.Kh1?! each cost more, and in the final position Stockfish rates White more than four pawns worse. 25...h6!! announced the zugzwang: White has no useful move, and Black can simply wait with his king.
The final position is unusual for a zugzwang. White has 27 legal moves with 11 different pieces and pawns, yet none of them helps; if 26.Kh2 or 26.g4, then 26...R5f3 wins.
Is it really zugzwang?
Not everyone accepts the name. Andrew Soltis pointed out that Sämisch could have moved a piece, returning the sacrificed material and playing on for some time, and that Black still had a threat to win the queen, which a true zugzwang should not need. Wolfgang Heidenfeld made the same point about the threat of ...R5f3.
Raymond Keene preferred to call it total paralysis, the ultimate expression of prophylaxis. Stockfish backs the practical verdict either way: its best line for White, 26.a3 R5f3 27.Bxf3 Rxf3, gives back material and stays more than four pawns down.
Soltis offered Podgaets–Dvoretsky, USSR 1974, as his ideal zugzwang game, and the Tomb Game, Harper–Zuk, Burnaby 1971, is another contender. In both, White has only one or two pieces able to move; here White has over 20 legal moves by 11 different pieces and pawns, which is what makes the position remarkable. Nimzowitsch wrote in 1925 that the name came from Danish chess circles.
What else was playable?
Stockfish 18 at the moments that decided the game, from White’s point of view:
- 9.e4 (instead of 9.cxd5): About +0.6 against −0.1 for the game move; the exchange on d5 released the tension in the centre.
- 20.Bf3 (instead of 20.e4): About −0.9. The game move is about −2.2 and allowed 20...fxe4.
- 25...Qf7 or 25...a5 (instead of 25...h6): About −4.5 as well. Several waiting moves win; 25...h6 is the one that made the point.
What to take away
- Restricting the opponent’s pieces can be worth more than attacking them, as Black’s ...a6, ...b5 and ...b4 showed.
- A sacrifice for lasting factors, such as two pawns and the seventh rank, can be calculated rather than gambled.
- Zugzwang is not only an endgame idea: when every move worsens a position, a quiet waiting move can win.
Want to turn one of your own games into a video? Try the ChessMotion Chess Video Maker.
Sources
- Immortal Zugzwang Game — Nimzowitsch’s notes, the final position and the objections to the name.
- Aron Nimzowitsch.
Read next: Karpov–Unzicker: the bishop on a7, Botvinnik–Capablanca, AVRO 1938, Capablanca–Tartakower: the active king on g3 and Full Game template.