Chessy
Blackmar-Diemer Gambit (1.d4 d5 2.e4)
A d4-player's answer to sharp, tactical chess — White offers a pawn on move two to get rapid piece activity against a closed-minded defender. Engines are unimpressed, but club players have tortured each other with it for 150 years. Face it from both sides in the drill below.
Play the Blackmar-Diemer Gambit against the engine
Free, no signup — you play white, the engine adapts to your level.
You just drilled the Blackmar-Diemer Gambit against the engine. Sign up free on Chessy and get AI coaching that explains when White's piece activity is real compensation and when it's just a pawn down.
Create a free account →The engine verdict: a half-pawn concession
After 1.d4 d5 2.e4, Stockfish at depth 16 gives Black −0.49 — just under half a pawn. White is betting rapid development (Nc3, Nf3, Bg5 after 2...dxe4 3.Nc3) against a material deficit. It's not a refuted mess like some gambits, but it's also not sound: a careful Black player holds an edge from move two.
Scoreboard across 10.9 million games
Lichess data across 10.9 million games shows White at 50.4% and Black at 46.1% — White barely edges it. The most important detail is hidden in the top-moves table: Black's engine-best reply 2...dxe4 actually yields lower win rates for Black (49.7% for White) than declining with 2...Nf6 does (White 54.4%). The move that looks sharpest is the safest response.
How to meet it: take the pawn
Stockfish says 2...dxe4 — accept, then develop carefully. The alternatives cost Black real ground:
- 2...e6 — inaccuracy, 87 cp loss vs dxe4
- 2...c6 — inaccuracy, 82 cp loss
- 2...Nf6 — mistake, 130 cp loss; ironically lets White get the Nf6 attacks he wants
Grab the pawn, meet 3.Nc3 with 3...Nf6, and White's compensation evaporates against correct play.
If you want to play the BDG as White
Know what you're buying: practical complications, not objective equality. The BDG is at its best when Black plays a natural-looking but slightly wrong move — 2...Nf6, for instance, scores a remarkable 54.4% for White across half a million games. Against well-prepared opposition it's a half-pawn deficit from the start; against anyone unfamiliar with ...dxe4 development, it punches well above its weight.
Results across 10,950,420 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| dxe4 | 7,485,231 | 49.7% |
| e6 | 1,131,752 | 51.0% |
| c6 | 766,791 | 51.9% |
| Nf6 | 551,173 | 54.4% |
| Nc6 | 456,938 | 50.0% |
| c5 | 195,331 | 50.2% |
Frequently asked questions
Is the Blackmar-Diemer Gambit refuted?
Not sharply refuted, but objectively dubious. Stockfish at depth 16 evaluates the position at −0.49 (Black better). It's a real practical weapon at club level, not a theoretical wreck.
What is the correct response to the BDG?
Accept with 2...dxe4 and develop naturally. Declining with 2...Nf6 is the most common mistake — Stockfish rates it 130 cp worse than taking, and White scores 54.4% against it across 551k Lichess games.
What is White's best move after 2...dxe4?
Stockfish's principal variation continues 3.Nc3 Nf6 4.Bg5, trying to recover the pawn while posting pieces aggressively. That said, facts file best_move_san refers to Black's reply at move 2 — the engine's recommendation for White's continuation is 3.Nc3 per the PV.
Does White actually score well with the BDG?
Barely. White scores 50.4% overall across 10.9 million Lichess games — ahead of Black's 46.1%, but the margin is thin. It's a coin-flip in practice, not a dominant attacking weapon.
How many games feature the Blackmar-Diemer Gambit?
Over 11 million Lichess games have reached the Blackmar-Diemer Gambit position. White wins 50.4%, Black wins 46.1%, with 3.5% draws — based on real rated games.