Légal’s mate: the oldest opening trap
Named after Sire de Légal, who is said to have played it in eighteenth-century Paris. White appears to blunder the queen — and mates with three minor pieces instead.
The position
White to move.
- 1.e4 e5 2.Bc4 d6 3.Nf3 Bg4 4.Nc3 g6 5.Nxe5! Bxd1 6.Bxf7+ Ke7.
- 7.Nd5#
If 5...dxe5 (declining):
- 5...dxe5 6.Qxg4 — White has simply won a pawn.
Why does this work?
The trap turns on the pin that isn’t. Black’s bishop on g4 “pins” the f3-knight against the queen — but a pin against anything except the king is only a suggestion. Nxe5 breaks it, and the greedy 5...Bxd1 loses to a forced mate: the bishop check drags the king to e7, and Nd5 covers every flight square with help from the e5-knight and Black’s own d6-pawn.
The honest defense 5...dxe5 avoids mate but concedes a clean pawn after Qxg4 — the bishop on g4 was only ever defended by the illusion of the pin.
This line is the classic argument for checking every capture: material counting said Bxd1 wins a queen; the mating net said otherwise. As a replay with both branches recorded, viewers see the trap and the refusal side by side.
The oldest trap still worth knowing
The trap is named after Sire de Légal, an eighteenth-century Parisian player, which makes it one of the oldest named opening traps in the game. It still catches players today, which is remarkable for an idea this old.
The mechanism is a queen offered as bait and a mate delivered by three minor pieces — bishop, knight and knight — against a king that never castled.
The pin that was never real
Black pins the f3 knight against the queen with ...Bg4, and White ignores it: Nxe5. If Black takes the queen with ...Bxd1, White mates with Bxf7+ and Nd5#.
The point is that a pin against a queen is not a pin at all. A pinned piece may not move only when the piece behind it is the king. Everything else is a cost-benefit question, and here the queen is worth less than the mate.
What the engine says about the whole line
Checked move by move, White’s play in the recorded line is sound but not forced: the early moves give up a few hundredths, and 5.Nxe5 evaluates at about +2 — a good move on its own terms even if Black declines the bait.
That is worth stating plainly. Taking the queen is the losing move; declining it leaves Black merely worse rather than mated. The trap does not win by force, it wins by temptation.
Why 4...g6 is the real mistake
In the line shown, the move that loses material is not the queen capture but the quiet move before it. After 4...Nc6 or 4...Nf6, Stockfish 18 rates White only about +0.6. After 4...g6, 5.Nxe5 is worth about +2, because Black’s best reply, 5...dxe5 6.Qxg4, simply leaves White a clean pawn up.
With a knight on c6 instead, the same trick fails. 5.Nxe5 is then met by 5...Nxe5, which recaptures, still guards the bishop on g4, and leaves Black about three pawns better. The trap only works when nothing can take back on e5.
The original game, and its doubts
The usual story is Légal against Saint Brie in Paris in 1750, with Légal giving the odds of a rook. Later research suggests the year is wrong, with 1787 more likely, and that the published move order was altered afterwards to hide a flaw: in the probable original, Black had a knight on c6, and the combination could have been refuted.
Légal is also said to have dressed the trap up with a piece of theatre. He touched the f3-knight, withdrew his hand as if he had only just noticed the pin, and played Nxe5 when his opponent insisted on the touch-move rule; the opponent took the queen at once. Joseph Henry Blackburne set the same trap so often on his tours that it is also called the Blackburne Trap.
A pure mate
The final position is a pure mate: every square around the black king is ruled out for exactly one reason. The bishop on f7 covers e6 and e8, the knight on d5 covers f6, and the knight on e5 covers d7 and protects the bishop. Black’s own pawn on d6, queen on d8 and bishop on f8 fill the rest.
It is also a mate by three minor pieces, delivered while Black’s queen and both rooks are still on their starting squares and Black has just won White’s queen. That contrast, heavy pieces at home against three light ones in contact, is what makes the trap so memorable.
What else was playable?
Where the game is actually decided, per Stockfish:
- 5.Nxe5: About +2.0 and the engine’s top choice — sound whether or not Black takes.
- 5...Bxd1 (taking the queen): Loses immediately to 6.Bxf7+ Ke7 7.Nd5#, which the engine confirms as forced mate.
- Declining the queen: Black stays in the game a pawn down. The trap is a practical weapon, not a refutation.
What to take away
- Only a king behind a piece creates an absolute pin. Against a queen it is a suggestion.
- Before capturing material offered in the opening, count the attackers around your king.
- Three minor pieces can mate an uncastled king without any heavy pieces at all.
Want to turn one of your own games into a video? Try the ChessMotion Chess Video Maker.
Sources
- Légal Trap — Origin, the disputed 1750 date, Blackburne and the pure mate.
Read next: Open game: early tactics on e5, The royal fork, The Fried Liver Attack: why ...Nxd5 loses and Classic Replay template.