Jerome Gambit: What Can We Learn? (Part 1)

There is more to annotating a lost chess game than discovering which moves to attach the "?!" and "?" symbols.

When that game starts out as a Jerome Gambit, 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+, it can always be tempting to give the 4th and 5th moves "?"s and sign off with "...and the rest is a matter of technique (for Black)".

But, I play the Jerome Gambit (376 games from the 5th move given above; scoring 81%), and a lot of other club players do, as well. 

So the question is: What can we learn from this defeat?

The following game took place in the 4th round of the 2022-2023 "Giuoco Piano Game" tournament at Chess.com. My opponent had scored 8 - 0 - 0 in his group in the 1st round, 5 - 0 - 3 in the 2nd round, and 6 - 1 - 1 in the 3rd round, for a total of 19 wins, 1 loss and 4 draws.

We had met once before, 7 1/2 years ago, and the blog post "Toward Disaster" gives a clue as to how perrypawnpusher - auswebby, "Giuoco Piano" tournament, Chess.com, 2015 (0-1, 37) turned out. He was rated about 60 points higher than me back then, maybe twice that for this game.

perrypawnpusher - auswebby

3 days/move, "Giuoco Piano Game", Chess.com, 2022

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+

If I am playing in a Giuoco Piano or Italian Game tournament, regardless of the time control, I will  play the Jerome Gambit every chance I get.

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Kf8 

The Jerome Defense to the Jerome Gambit (see "Jerome Gambit / Jerome Defense"), was mentioned by Alonzo Wheeler Jerome in his analysis published in the July 1874 issue of the Dubuque Chess Journal.

A half dozen years later Jerome used the defense against his own gambit in two games: Jaeger - Jerome, correspondence, 1880 (0-1, 14) and Jaeger - Jerome, correspondence, 1880 (0-1, 40)

To take a broader perspective, The Database has 1,965 games with this move. White scores 48%.

I have played 40 games games against 6...Kf8, scoring 76%.

7.Qxe5 Qe7 

The analysis engine at lichess.org, commenting on a different game with the same position, fussed that 7...Qe7 was "inaccurate", preferring 7...d5.

Consulting The Database - which has almost 25,000 games starting with 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ plus an additional 55,000 related games - I found 2,171 examples with 7...Qe7. White scored 42%. (That means Black scored 58%, which is not bad for an inaccuracy.) 

By contrast, in the 10 games where I faced 7...Qe7, I scored 70%. 

Although 7...d5 is very rare in the games in The Database, of the 5 games it contained, White scored only 20%, so the silicon beast may be onto something. That would make the move "Jerome Gambit Secret #17", I think.

8.Qf4+ 

I am not sure that 8.Qxe7+ would come to mind for the Jerome Gambiteer, but I note that The Database has 80 such games, with White scoring 59%.

8...Ke8 9.O-O d6 

10.c3 Qf7 11.Qg3 Ne7 12.d4 Bb6

White has two pawns for his sacrificed piece and a formidible pawn center. Black's extra piece gives him the Bishop pair. White's King is castled, Black's King is in on a center file with no quick escape.

Stockfish 15, which usually takes a somewhat skeptical view of the Jerome Gambit in its analysis, sees Black now as about a pawn and a half better (34 ply). That is about a two-pawn improvement for White since 4.Bxf7+, but there is plenty of work still to be done. 

[to be continued]