Fried Liver Attack
Named variation
1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 is where the Two Knights Defense splits into two entirely different openings depending on a single recapture. 5...Na5 attacks the bishop and, after 6.Bb5+ c6 7.dxc6 bxc6, leads to a roughly balanced middlegame most strong players prefer. 5...Nxd5 is the move that actually creates the Fried Liver Attack — it looks like the more natural recapture, and it walks straight into 6.Nxf7!?, a knight sacrifice that drags Black's king to f7 whether Black wants it there or not. Nothing about this page is about the Two Knights Defense in general; it is specifically about what happens once that one recapture has been played.
What the sacrifice actually buys
After 6.Nxf7 Kxf7, White is down a full knight, and modern engines do not consider this a genuinely clean advantage — the compensation is practical, not permanent material or a forced mating attack. What White actually gets is a king that has lost castling rights for good, sits stranded in the open, and has to survive a whole sequence of forcing checks starting with 7.Qf3+. Black has real, concrete defensive tasks to solve over the next several moves, and a single inaccurate square for the king turns White's temporary initiative into a real attack. That is the entire trade the Fried Liver offers: a whole piece, for a hunted king and a series of only-moves Black has to find under pressure.
Starting position. Fried Liver Attack — the forced king hunt.
The forced sequence that gives the Fried Liver Attack its name.
The name itself is old and specific to this exact idea: it comes from the Italian fegatello, a dish of liver wrapped and skewered before cooking, a metaphor early Italian analysts used for a king that gets dragged out and cooked in the center of the board. Lines resembling this sacrifice were already being written about centuries before the modern opening name settled, which is part of why the position still carries a distinct, memorable identity separate from the wider Two Knights Defense it grew out of — it was recognized as its own idea long before ECO codes existed to classify it.
Why this matters more in practice than in theory
Because 5...Na5 is the theoretically preferred move for Black, the Fried Liver Attack is, at the highest level, largely a historical curiosity rather than a live weapon — well-prepared players simply avoid the position that creates it. Club play is different. 5...Nxd5 is the intuitive, natural-looking recapture, and plenty of players reach for it without knowing what follows. That gap between top-level theory and typical club knowledge is the entire reason this line still matters: White does not need the sacrifice to be objectively sound, only for the opponent across the board to be unfamiliar with the defense, which happens constantly below master level.
The follow-up White actually needs to know
After 7.Qf3+ Ke6 (the only reasonable square — Ke8 walks back into more checks and Kg8 loses the d5 knight for nothing), 8.Nc3 attacks the d5 knight a second time and keeps Black's king in the center under fire. This is the moment where memorizing the sacrifice itself, without the follow-up, stops being enough: having given up a full piece, failing to keep finding forcing moves here just leaves White down material with nothing to show for it. The sacrifice is not the hard part of this opening for White — converting the initiative into something concrete over the next four or five moves is.
On the defending side, the recurring club-level mistake isn't a single bad move so much as panic: Black players who reach this position under-the-board for the first time often try to save every piece at once, shuffling the king back and forth or moving the attacked knight to a square that loses more material to a follow-up check, instead of accepting that some further concession is unavoidable and looking for the move that gives up the least. A calm defender who accepts giving back some material to finish development usually survives; a defender trying to hold everything usually doesn't.
For the branch that avoids all of this with 4...d4 instead, see the Italian Gambit, and for the tactical pattern behind the sacrifice itself, see how forks work — 8.Nc3's double attack on the d5 knight is exactly this idea in practice.
Frequently Asked Questions
Is 6.Nxf7 actually a good move, or just a trick that only works against weaker players?
Modern theory rates the resulting position as close to balanced with best play from both sides, not as a clear advantage for either. It's a real, respectable try rather than a pure trick, but it isn't a forced win — its practical strength comes from how much harder Black's defensive task is than White's attacking task.
What should Black play instead of 5...Nxd5 to avoid all of this?
5...Na5, attacking the bishop on c4 immediately, is the well-established main line. After 6.Bb5+ c6 7.dxc6 bxc6, Black gets a slightly loose pawn structure but a safe king and full material — the position most strong players prefer over walking into the sacrifice.
If White reaches this position but doesn't remember the follow-up after 6...Kxf7, is the sacrifice already ruined?
Largely, yes. The sacrifice only pays off if White keeps generating threats — 7.Qf3+ and 8.Nc3 specifically. A White player who takes the knight on f7 and then plays quiet developing moves is simply down a piece with no compensation, which is worse than never sacrificing at all.