What I recalled in my notes to move 6 got me chuckling, but what I found in my notes to move 7 sent my mind reeling: The "Jerome treatment" leads to this - Unbelieveable!
Take a look and see.
todotranquilo - SrNoth
blitz, FICS, 2014
1.e4 e5 2.Nf3 Nc6 3.Bc4 h6 4.d4
Sometimes when I am expecting Black to reply with 3...Bc5, I am met with 3...h6, the Semi-Italian Opening, instead. I usually continue with my development with 4.0-0 and hope for Black to fall in with my wishes. Occasionally I see 4...Nf6 and have to try 5.Nc3 before I see 5...Bc5 (and am then able to play 6.Bxf7+).
The earlier series on this blog, "A Jerome Look At The Semi-Italian Opening (Parts 1, 2, 3, 4 5 and 6)", is well worth visiting for history, analysis and games.
I took a quick look at The Database, and noticed:
There are 1,569 games with (or transposing to) 4.d4, with White scoring 64%; there are 352 games with 4.0-0, with White scoring 70%; and there are 349 games with 4.Nc3, with White scoring 61%.
The Database showed only 153 games with this move (not the best), with White scoring 70%. The more popular (and stronger) alternative, 4...exd4, appeared in 1,062 games, where White scored 57%.
5.dxe5 dxe5 6.Bxf7+
Of course, the straight-forward 6.Qxd8+ Kxd8 (or 6...Nxd8 7.Nxe5) 7.Bxf7 would win a pawn for White, but todotranquilo prefers to play like Jerome.
(Due to sampling bias, The Database does not have any games with 6.Qxd8+, but a quick look at ChessBase's online database gives 63 games with that move; White scores 86%. The two databases are vastly different, making comparisons dicey, but this peek gives an indication of the validity of 6.Qxd8+.)
Interestingly enough, 6.Bxf7+ was embraced in Chess Master vs Chess Amateur (1963), by Max Euwe and Walter Meiden
White could act even more energetically [than 6.Qxd8+ etc.] by playing 6.Bxf7+ Kxf7 7.Nxe5+, and Black is in great trouble, for (a) 7...Nxe5? 8.Qxd8 or (b) 7...Ke7 8.Ng6+, or (c) 7...Ke8 8.Qh5+ Ke7 9.Ng6+. After (d) 7...Kf6, the continuation is less clear. This does not mean that the sacrifice would be incorrect, since after 8.Nd3, White has two Pawns for a piece and the Black King is badly placed. On the other hand, a safe win of a Pawn as in the other variations, might be preferable.Stay with me, there is more.
For starters, due to the magic of transpositions, there are 2,025 games in The Database with this position (after 6.Bxf7+), and White scores 41% - a reason, from a practical point of view, that Readers might consider the mundane 6.Qxd8+ over the dashing 6.Bxf7+. (The online ChessBase database has only one game with the position, a win for White.)
Of course, "from a practical point of view" it might also be wise to avoid the Jerome Gambit altogether; but where is the fun in that??
6...Kxf7 7.Nxe5+ Ke8
As mentioned above, 7...Kf6 is probably Black's strongest reply. Surprisingly, after the suggested pragmatic 8.Nd3, Stockfish 6 recommends 8...Kg6 and 9...Kh7 for Black, maintaining that the defense holds, and that the second player would have the advantage!
To be fair, the authors of Chess Master vs Chess Amateur were tidying up an offhand line in an offhand variation, not preparing for a tournament or writing an opening manual. But, what is the best response to 7...Kf6 ?
The best move seems to be 8.Qf3+!?, sacrificing a piece in true Jerome Gambit style! It turns out that anything else other than taking the piece with 8...Kxe5 for Black leads to checkmate. White's best move then is keep-the-enemy-King-in-the-middle 9.Qf7!?
I gave the crazy position - White has sacrificed two pieces, Black's King is in mortal danger - to Stockfish 6, and here is just a surface view of what it found (the game is even) 9...Nf6 (9...Bb4+ 10.Nc3 Bxc3+ 11.bxc3 Nf6 12.f4+ Kxe4 13.0–0 Qg8 14.Qg6+ Kd5 15.Rd1+ Nd4 16.Rxd4+ Kc6 17.Qd3 Be6 18.f5 Rd8 19.fxe6 Rxd4 20.cxd4 Qxe6 White has an edge) 10.f4+ Kd6 (10...Kxe4 11.Nd2+ Qxd2+ the only move to avoid mate 12.Bxd2) 11.e5+ Nxe5 (11...Kc5 12.Be3+ Nd4 13.Nc3 Bg4 14.h3 White is better) 12.fxe5+ Kc6 13.Qc4+ (13.exf6? Bc5 Black is better) 13...Bc5 14.Be3 b6 15.Bxc5 Qd5 16.Qxd5+ Nxd5 17.Bf2 Ba6 18.Na3 Rhe8 19.0–0–0 Rxe5=
Please. Even that is too much. Is there any wonder why I play 4.0-0 ?
The line Euwe and Meiden mention, 8...Ke7 9.Ng6+ is clearly good for White, who wins a Rook. They can be forgiven for not having a computer program which would have told them that the alternative 9.Qf7+ would lead to checkmate in a dozen moves.
9.Qxg6+ Black resigned