Chessy
Elephant Gambit (1.e4 e5 2.Nf3 d5)
Black plays 2...d5 to immediately counter-attack in the centre — bold, risky, and extremely misunderstood. The engine awards White nearly a full pawn, but the scoreboard barely shows it. Find out why in the drill below.
Play the Elephant Gambit against the engine
Free, no signup — you play black, the engine adapts to your level.
You just navigated the Elephant Gambit traps against the engine. Create a free Chessy account and let AI coaching explain why 3.Bc4 scores 29.2% — and how to avoid the trap yourself.
Create a free account →Black throws a pawn into White's centre
The Elephant Gambit is Black's idea: after 1.e4 e5 2.Nf3, Black plays 2...d5 instead of defending the e5-pawn normally. It's White to move (the FEN confirms it), and Stockfish gives White +0.92 at depth 16 — nearly a full pawn of objective advantage. Black is banking that the unbalanced pawn structure creates enough confusion to compensate. The engine says it doesn't.
Why White barely wins despite the +0.92
Across 8.7 million Lichess games White scores only 49.6%, Black 46.9% — a whisker of a gap for a nearly-pawn advantage. The explanation sits in the top-moves breakdown. White's engine-best reply 3.exd5 scores 50.6% across 5.4 million games, close to expectation. But 3.Bc4 — a plausible-looking developing move — scores a startling 29.2% for White, dragging down the overall average. White fumbles the refutation constantly.
White's best answer: just take the pawn
Stockfish says 3.exd5 — grab the pawn cleanly and develop. The PV runs 3.exd5 e4 4.Qe2, neutralising Black's pawn push. Mistakes White makes that give away the advantage:
- 3.Nc3 — mistake, 128 cp loss vs exd5; lets Black rebuild the centre
- 3.d3 — mistake, 111 cp loss; passive and gives Black time to consolidate
- 3.d4 — inaccuracy, 56 cp loss; reasonable but not optimal
And 3.Bc4 — not in the common_mistakes list but visible in lichess.top_moves scoring 29.2% — is a serious practical danger zone for White.
What Black is actually hoping for
The Elephant Gambit is a practical device, not a theoretically correct counter. Black is betting White chooses 3.Bc4 or 3.Nc3, plays into the resulting tactical chaos, and stumbles. If White simply plays 3.exd5 correctly, Black has handed over a pawn with little compensation. It's a trap for the opening phase, not an opening system.
Results across 8,673,155 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| exd5 | 5,379,316 | 50.6% |
| Nxe5 | 1,455,035 | 50.5% |
| Nc3 | 471,661 | 47.2% |
| d3 | 457,037 | 46.2% |
| d4 | 388,552 | 52.0% |
| Bc4 | 153,619 | 29.2% |
Frequently asked questions
Is the Elephant Gambit good?
Objectively no — Stockfish gives White +0.92 after 2...d5. But in practice White scores only 49.6% across 8.7 million Lichess games, mostly because White often plays 3.Bc4, which scores just 29.2% for White.
What is the best reply to the Elephant Gambit?
3.exd5 — simply take the pawn. Stockfish's PV continues 3.exd5 e4 4.Qe2, maintaining the material advantage. 3.Nc3 and 3.d3 are both mistakes that cost White over 100 centipawns each.
Why is 3.Bc4 so bad against the Elephant Gambit?
The Lichess data is striking: 3.Bc4 scores only 29.2% for White across 153k games. The move walks into tactical tricks after ...d4 and ...e4, giving Black active counter-play that White did not bargain for.
Does the Elephant Gambit have any tricks to know?
Yes. The main trap is luring White into 3.Bc4, then unleashing ...d4 and ...e4 pawn pushes to gain time. Once White knows to play 3.exd5, the tricks disappear — but most players don't drill this line and step in anyway.
How many games feature the Elephant Gambit?
Over 9 million Lichess games have reached the Elephant Gambit position. White wins 49.6%, Black wins 46.9%, with 3.5% draws — based on real rated games.