In the Scotch Gambit, 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Bc4, White does not recapture on d4 at once and develops the bishop towards f7 instead. Black can keep the pawn with ...Bc5 or ...Bb4+, or return it for development with ...Nf6. The play is fast, and the bishop on c4 makes f7 the target of almost every white idea.
That is also the source of the most common mistake in our data, by a wide margin. After 4...Bb4+ 5.Bd2 Bc5, White plays 6.Bxf7+?, hoping for a quick attack on the king. But after 6...Kxf7 the follow-up checks lead nowhere and the king walks to safety. White has given a bishop for a pawn. The same sacrifice fails in several other lines, for example after 4.Nxd4 Nxd4 5.Qxd4 Ne7 6.Bc4 Nc6 7.Bxf7+?.
The bishop on g5 is the second theme. After 4...Nf6 5.e5 Ng4 6.O-O d5 White plays 7.Bg5?, but the knight on g4 still covers nothing important and 7...f6 traps the bishop: after 8.Bb5 fxg5 Black is a piece up. In the Danish-style line 4.c3 dxc3 5.Bc4 d6, the same 6.Bg5? drops the b2 pawn and soon the rook.
Black's own mistakes are usually loose knight moves. After 4...Bb4+ 5.Nbd2 Qe7 6.O-O the jump 6...Ne5? looks active, but 7.Nxe5 starts a sequence that leaves Black material down: trading pieces while behind in development rarely helps the defender.
The Scotch Gambit Trap with ...Be7 and Qd5 has its own page. The puzzles below train the rest: when f7 really falls, and when the sacrifice only gives away a piece.
| Mistake | Side | Count | Avg loss |
|---|---|---|---|
| 6...Bxf7+ | Black | 115 | 4.2 pawns |
| 7...Bxf7+ | Black | 75 | 4.1 pawns |
| 7...Bg5 | Black | 44 | 6.4 pawns |
| 6.Ne5 | White | 32 | 4.5 pawns |
| 6...Ng5 | Black | 31 | 4.8 pawns |