Jerome Gambit for Dummies 2.0 (Part 2)

                               

[continued from the previous post]

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

The Jerome Gambit

This is the first Critical Position in what is generally referred to as the Jerome Gambit. (There are variations, which will be noted below.)

The Database has 22,156 games with this position. White scores 50%. This number has been creeping upwards over the years. 

Looking back, 12 years ago, according to "Opening Reports on the New Year's Database", this position scored 44% for White.

Improving, 4 years ago, according to "The Database Upgrade", this position scored 46%.

Why the increase?

There is always the possibility of "sampling bias" - that Readers of this blog send me their wins, but not their losses, so that success is over-represented. Likewise, that published wins uncovered by my research are more likely than published loses.

I have worked to overcome this bias by including all 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ games (wins, losses and draws) from the FICS database, from 1991 through November 2021, to give a representative (i.e. of online club play) selection. To be fair, over time the portion of The Database that is FICS-derived has dropped from about 95% (see "What Exactly is 'The Database' ?") to a current 56% - primarily due to an influx of games from lichess.org (Thank you, Dan Middlemiss!) although, again, they seem balanced between wins, losses and draws.

It is also possible that Jerome Gambit players, over time, have gotten better, and that is reflected in the improved scoring. Or it could be that increased interest has attracted stronger players, who, in turn, have been more successful with the Jerome.

Before moving on, I want to make a couple of comparisons. If you look at the Italian Four Knights Jerome Gambit - 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Bc5 5.Bxf7+

Italian Four Knights Jerome Gambit

The Database has 2,969 games with this position, with White scoring only 39%. This is comparable to 38%, 4 years ago. This difference in results from the main line Jerome Gambit (39% vs 50%) occurs even though Stockfish 14.1 sees Black's position in the Italian Four Knights Jerome Gambit as only a little more than a half-pawn better (at 30 ply depth) than in the regular Jerome Gambit.

Perhaps this is why, of late, I have been seeing more Jerome-like Bishop sacrifices against the Two Knights Defense, i.e. 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Bxf7+, not waiting for the Italian Four Knights Jerome Gambit to develop. The Database is not representative, here, but notably has 735 game examples, with White scoring 46%.

Clearly, I need to do more resarch on the Italian Four Knights Jerome Gambit.

Finally, The Database has 447 Semi-Italian Jerome Gambit games, 1.e4 e5 2.Nf3 Nc6 3.Bc4 h6 4.Nc3/0-0 Bc5 5.Bxf7+. 

Semi-Italian Jerome Gambit

Semi-Italian Jerome Gambit

 White scores 51% with 4.Nc3, and 62% with 4.0-0. That is a significant difference, even though Stockfish 14.1 (at 30 ply) sees 4.0-0 as less than a 1/2 pawn better.

(It is likely coincidental that in the two examples given, above, an increase of 1/2 pawn in Stockfish's evaluation is accompanied by a roughly 10% in scoring results.)

[to be continued]