Jerome Gambit: GM Game

Recently Dan Middlemiss sent me a collection of Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) games, and pointed out one that was played by a grandmaster, Miroslav Miljkovic.

I looked at a number of the GM's games at lichess.org. In some bullet and ultra bullet games where he sacrificed his Bishop on f7 he seems to have had a lot of fun

MiroMiljkovic - Dostoyevsky99

1 0 bullet, lichess.org

1.e4 

In a game that shows "anything goes" when the time control is a fraction of a minute: 1.Nc3 b5 2.e4 b4 3.Bc4 bxc3 4.Bd5 cxb2 5.Bxf7+ Kxf7 6.Bxb2 e6 7.Qh5+ Ke7 8.Qf3 Ke8 9.Ne2 Qe7 10.Ba3 Qxa3 11.O-O Qc1 12.Nxc1 Black resigned, MiroMiljkovic - SimonHSV, lichess.org, 2020. 

1...e5 

Another example: 1...d6 2.c3 e5 3.Nf3 Be7 4.Bc4 Nf6 5.Bxf7+ Kxf7 6.Ng5+ Ke8 7.d4 Rf8 8.Qb3 h6 9.Ne6 Kd7 10.Qc4 Qe8 11.d5 c6 12.dxc6+ Nxc6 13.Nc7 Kxc7 14.Qb5 Be6 15.Qd3 Bc4 16.Qe2 Bd3 17.Qxd3 Nb4 18.Qe2 d5 19.cxb4 Ng4 20.Bd2 Ne3 21Qb5 Nxg2+ White resigned, MiroMiljkovic - Therafik, lichess.org, 2020. 

2.Nf3 

Or, even faster: 2.Bc4 Nc6 3.Bxf7+ Kxf7 4.Nf3 Ke8 5.O-O Qf6 6.d4 exd4 7.e5 Nxe5 8.Nxe5 Qxe5 9.Re1 Qxe1+ 10.Qxe1+ Kf7 11.Qe2 Nf6 12.Bg5 Bd6 13.Bxf6 gxf6 14.Qh5+ Kg7 15.Qh4 Re8 16.Qg4+ Kf7 17.Nd2 Re2 18.Ne4 Rd2 19.Nxd6+ cxd6 20.Qh5+ Ke7 21.Qg4 Rxc2 22.Qf4 Rxb2 23.Qe4+ Black resigned, MiroMiljkovic - Jhonortega123, lichess.org, 2020. 

2...Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.d4 

This is not a strong Jerome Gambit line, although it meets the requirements of a bullet game - open lines, please.

As I wrote to Dan

It is good to see a GM Jerome game, although Miljkovic made a real hash of it, early, I am afraid. The database has a hefty 2,600 games with 5.d4, with White scoring 38%. Even if you adjust for the fact that the average White was rated about 100 points below the average Black, that's still a bit anemic. 

Or is it? The Jerome - 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ - is scoring 50 % in The Database, with the average Whites and average Blacks rated about equal. If White's average were -100, as in the above example, would that be enough to drop the score to 38%? Actually, a quick look at the InterWeb suggests that -100 rating would give you only a 34% winning chance. Hmmmm.... 

GM Miljkovic is listed as a coach; perhaps he was playing a student - albeit, a highly rated one - and was giving "Jerome Gambit odds"? Bullet time control is madness, IMHO. 

5...Bxd4 

I suppose he would have answered 5...Nxd4 with 6.Nxe5+, and 5...exd4 with 6.c3.

6.O-O Nf6 7.c3 Bb6 8.Bg5 d6 9.Qb3+ Be6 10.Qc2 h6 11.Be3 Rf8 12.Nh4 Ng4 13.Nd2 Qxh4 14.Bxb6 Qxh2 checkmate

Ouch.