Jerome Gambit: Stockfish 12 Defeats Magnus Carlsen's Jerome Gambit?!

It has happened before.

Very early in the life of this blog I was taken in by a reference to the book "All or Nothing! The Jerome Gambit" by Chiam Schmendrick. I searched, but of course could not find the text, as neither it nor its author existed. 

Years later, I tried to track down a game that Alexander Alehine supposedly played, defending against the Jerome Gambit. Nothing.

Early last year I was trying to find the truth of the allegation in a newspaper chess column that Wilhelm Steinitz had lost to the Jerome Gambit the first time that he had faced it. Again, nothing.

Of course, I once did find a newspaper report of Emanuel Lasker's simultanous exhibition, where it was reasonably reported that he defeated a club player's Jerome Gambit, even if the game score was absent.

So, I was a bit unsure when I received the following email

Found this game from today before it was deleted off of Carlsen's page. Found it very instructional and interesting that stockfish 12 decided to go with the Jerome Gambit. Was wondering if you could write about it on your blog.

Well...

Let's suspend disbelief for a moment or two, and check out the game. Couldn't hurt, right?

The idea of the computer program being given "Jerome Gambit" odds by the World Champion is quite intriguing. 

Carlsen vs Stockfish 12
Chess.com, 2021

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

 

I am shocked to find the Jerome Gambit in Carlsen's opening repertoire, but he does like to explore unusual openings...

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Kf8 

Solid and sensible. You don't have to know much about the Jerome to make this choice. It might even be in Stockfish 12's book.

7.Qxe5 Qe7 

Alonzo Wheeler Jerome used this line to win two correspondence games from Daniel Jaeger in 1880. 

8.Qf4+ Nf6 9.d3 

A bit odd. I gave the game to my Stockfish 12, and with a minute per move in "blunder check" mode, it chose 9.Nc3. In fact, when given the position after move 8 and set to "infinite analysis" it immediately chose 9.Nc3 and did not waver.

I don't know what time control the game was played at but it had to be quite fast for Carlsen to "overlook" the best move.

9...Kf7

Practical and cautious. White's best move is not taking the pawn at c7, so this might be another reason for Black's choice.  

10.Qxc7

For the record, according to The Database this is a novelty, if not a particularly good one. Definitely not world class.

10...a6 11.f4 

This is the kind of Jerome-ish move that I would make, but it's a blunder. My version of Stockfish 12 again finds 11.Nc3 instantaneously. Strange. 

Is this the "Play Magnus" software, set to an early age?

11...Ng4 

My skepticism rises. Stockfish 12 overlooks 11...Nxe4!? ?

12.f5 Qh4+ 13.g3 Qe7 14.h3 Nf2 

15.O-O 

Cute - and foolish, what they call "castling into it". Perhaps he could hold on with 15.Rh2.This is beginning to look like a "banter blitz" game where White is wandering off topic to discuss his suggestions for the Fornyings- og administrasjonsdepartementet.

15...Nxd3+ 

Stronger was 15...Nxe4+, but it doesn't matter, as White steps into a forced checkmate.

16.Kg2 Qxe4+ 17.Kh2 Qe2+ 18.Kh1 Qxf1+ 19.Kh2 Qg1 checkmate

I'm always interested in seeing Jerome Gambit games, but I'm still a bit suspicious about the provenance of this one. More evidence, please.