When it comes to the Blackburne Shilling Gambit – 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4!? – White has several moves that can lead to good play: 4.Nxd4 (followed by 5.c3), 4.c3, or even simply 4.0-0.
Or there is the Jerome Gambit-ish 4.Bxf7+
Master Tim McGrew discussed the latter line in one of his "The Gambit Cartel" columns at ChessCafe, titled "A Shilling in the Mailbag".
Dennis has produced a number of chess videos – which I recommend highly, for both their instructional and entertainment value – including "Master Lesson - Improvisation in the Italian Game" which has more recent analysis on the Jerome-ized Blackburne Shilling Gambit.Dennis Monokroussos sent a note...
Dennis analyzes 4.Bxf7+ Kxf7 5.Nxe5+, “another fun possibility,” he writes, “when Black has to play very accurately to get even a small edge.”His analysis runs 5...Ke7! 6.c3 d6! 7.Nc4 Nc6 8.d4 Nf6 9.0-0 Kf7. “White probably doesn't have enough for the piece,” Dennis concludes, “but White is better here than Black is after anything normal but 4.Nxe5.”
Does anybody feel like trying Black’s position after 5...Ke6 instead of Dennis’s 5...Ke7? Truly, this begins to look like a line that only a computer could love!
Believe it or not, this has been tried in a tournament game. Wieteck - Dutschke, Lahnstein, 1999 saw 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 4.Bxf7+!? Kxf7 5.Nxe5+ Ke8? Allowing the check at h5 is unwise. 6.Qh5+ (White could also consider 6.c3 Nc6 7.Qh5+ g6 when both 8.Nxc6 and 8.Nxg6 come into consideration.) 6...g6 7.Nxg6 and now Black committed suicide with 7...Qf6??, but after 7...hxg6 8.Qxg6+ Ke7 9.Qg5+! Nf6 10.Qc5+ d6 11.Qxd4
White has four (!) pawns for the bishop and Black’s king will be exposed long-term.I cannot resist the opportunity to point out the (distant but discernable) analogy between Dennis’s second line and the Jerome Gambit 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+. The Jerome is, of course, completely unsound; it is a kind of miracle, and a tribute to Jerome’s tenacity, that it was analyzed seriously at all.
By the way, I took up Tim's challenge, and gave the position after 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 4.Bxf7+ Kxf7 5.Nxe5+ Ke6 to the new Rybka 3.0 for a minute or two. (One of these days I'll let it run overnight.)
Black could probably eat another pawn with 7...Kxd4 (instead of 7...Ke6) but look at what might happen: 8.Nc3 (8.d3 Ke5 9.Qg4 Ne7 10.Bg5 d5 11.Qh5 Ke6 12.0-0) c6 9.d3 Kc5 10.Bf4 Kb6 11.Qb3+ Ka6 12.a4 b6 13.Nd5 cxd5 14.Qb5 Kb7 15.Qxd5 Ka6 16.0-0 Ne7 17.Qxa8 Nc6 18.a5 bxa5 19.Rfc1 Bb4 20.Rxc6 dxc6 21.Qxc6 and we have a typical drawn position... (Ha!)
All very, very complicated - a line that only a computer could love, as Tim said.
Game examples are very welcome.