Chessy
Cochrane Gambit (1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nxf7)
White ignores the textbook Petrov and hurls a knight into f7 on move four, ripping open Black's king before Black can castle. Stockfish scores it −0.83 — Black is better — yet across 303,139 Lichess games White scores 49.8%, more than Black's 46.8%. Try it on Chessy and discover why the exposed king matters more than the material count.
Play the Cochrane Gambit against the engine
Free, no signup — you play white, the engine adapts to your level.
You just tried the Cochrane Gambit against the engine. Create a free Chessy account for AI coaching that tracks exactly where Black's king runs out of safe squares.
Create a free account →The engine verdict: dubious but not losing
After 4.Nxf7 Stockfish evaluates −0.83 at depth 16 — White is down roughly a knight for a pawn, which is clearly material-negative. The engine's antidote is 4...Kxf7 5.Nc3 c5 6.d4, where White chases queenside space while Black consolidates around the exposed king. At nearly a full pawn behind, no engine will play this — but −0.83 is far milder than a typical piece sacrifice, reflecting the real pressure the knight sac generates.
Why White outscores in practice
The Lichess data is striking: White scores 49.8% vs Black's 46.8% across 303,139 games — the gambiteer actually outscores despite the engine evaluation. The reason is the exposed king on f7: Black must solve concrete defensive problems every move, and most players can't. Panic replies collapse fast — Nxe4 (5,309 games) gives White 86.4%, Qe7 (2,285 games) gives White 69.4%, and Ke7 (299 games) gives White 90.3%. Only Kxf7 keeps Black's edge intact.
How to face it: king takes f7, then consolidate
The only correct reply is 4...Kxf7 (played in 294,370 games), after which White scores 48.9% — slightly under parity. From there, develop quickly, resist counterattacking immediately, and let the extra piece decide a longer game. Every alternative — Nxe4, Qe7, Ke7, Bg4 — blunders hundreds of centipawns and invites a direct attack on a king that has nowhere to hide.
If you play it as White on Chessy
This is not a pure trap weapon like the Jerome — the Cochrane has real practical value even after the correct reply. The initiative and the open king give White attacking themes that take serious preparation to defuse. Train the lines where Black panics (Nxe4, Qe7) and learn the follow-up. Against a well-prepared defender it's still objectively dubious, but −0.83 is close enough to generate real winning chances in rapid and blitz chess.
Results across 303,139 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| Kxf7 | 294,370 | 48.9% |
| Nxe4 | 5,309 | 86.4% |
| Qe7 | 2,285 | 69.4% |
| Ke7 | 299 | 90.3% |
| Bg4 | 275 | 73.1% |
| Be7 | 165 | 89.7% |
Frequently asked questions
Is the Cochrane Gambit sound?
Not by engine standards. Stockfish evaluates 4.Nxf7 at −0.83 (Black's favour) at depth 16, meaning White is objectively worse. But it's far milder than most piece sacrifices, and White scores 49.8% across 303,139 Lichess games.
What is the best reply to the Cochrane Gambit?
4...Kxf7. After Nc3 and d4, Black must play precisely and accept a structurally awkward king. All other moves — Nxe4, Qe7, Ke7 — are blunders giving White 69–90% scores.
Why does White score well in the Cochrane Gambit despite losing material?
Black's king is stripped of castling rights and forced to f7 immediately. Most players below master level cannot navigate the resulting pressure accurately — the engine can, but a human often can't over the board.
Is the Cochrane Gambit good for blitz?
Yes, it's well-suited to blitz and rapid. The attacking themes are concrete enough to execute quickly, the defensive task for Black is time-consuming, and most opponents are unfamiliar with it — a dangerous combination on the clock.
How many games feature the Cochrane Gambit?
Over 303K Lichess games have reached the Cochrane Gambit position. White wins 49.8%, Black wins 46.8%, with 3.4% draws — based on real rated games.