Technically, the game isn't even a Jerome Gambit... I need to expand upon my musings in the note to move 4.
Wall, Bill - Rube
1.e4 e5 2.Nf3 Na6
Already I am not sure what to say. Is Bill's opponent a chess-playing program?
3.Bc4 Bc5 4.Bxf7+
Why not? The "Jerome Solution" is often a fun way to deal with surprises.
With a Knight on c6 (instead of a6) The Database shows White scoring 55% against this declination, which I find very odd, especially in light of the fact that The Database statistic for the Jerome Gambit offered - 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ - is White scoring 45%. This suggests that when White moves from offering a piece (or two) for a pawn (or two) to accepting a pawn, he only improves his scoring chances 10%.
Returning the psychological "shock" - but, here, White takes the piece.
6.Kxf2 c6 7.Nxe5 d6 8.Nf7 Qf6+ 9.Qf3 Qxf3+ 10.gxf3 Nc5 11.Nxh8
'Tis a puzzlement.