I was playing over the following Abrahams Jerome Gambit game with Stockfish looking over my shoulder, when the computer made a comment that I didn't understand. It turns out that both players had overlooked something along with me and that only one of them was able to recover, and take advantage of the particular situation...
Korpav - Dagestan
standard, FICS, 2015
1.e4 e5 2.Bc4 Bc5 3.Bxf7+
3...Kxf7 4.Qh5+ Kf8 5.Qxe5 d6 6.Qc3 Qg5 7.Qf3+ Nf6
8.Ne2 Nc6
To my eyes, Black was doing a good job of dealing with the gambit, and his last move seemed solid and useful. However, Stockfish didn't like 8...Nc6 and suggested, instead, 8...Bg4 9.Qf4 Qxf4 10.Nxf4 Nc6.
Hmmm... First exchange Queens, then bring out the Knight. Okay... Why?
9.c3
The first thing both players (and the annotator) overlooked was 9.d4!, which allows White, after a bit of exercise, to win back his sacrificed piece with 9...Bb4+ 10.c3 Qg6 11.Nf4 Qxe4+ 12.Qxe4 Nxe4 13.cxb4 Nxb4 when
analysis diagram
Black's active pieces would give him only an edge.
Having missed the opportunity to take advantage of his opponent's Queen's danger, White Queen now experiences her own troubles at the hands of the enemy Knights.
9...Ne5 10.Qg3 Bxf2+ 11.Kxf2 Nxe4+ White resigned
