Facing the Jerome Gambit: Artificial Non-Intelligence

Programmers who create chess-playing engines usually try to make them as strong as possible. They then develop ways to scale back the strength, creating levels of play that can give a human opponent a chance to have a decent game. How, then, to make a smart computer program weaker? This question comes up a few times in the following game, as Bill wall takes on a program at a playable level.

Wall, Bill - GarboChess JS 6.0
2017

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

4...Kxf7 5.Nxe5+ Nxe5 6.d4 Qh4

This move is likely "book" for the program.

7.O-O

There are 150 games with this position in The Database.

7...Bb4

The computer saves an endangered piece, but the move is hardly consistent with the counterattacking ideas of the line. It is not surprising that this is the only game in The Database with the move.

8.dxe5 Qxe4 9.a3 Bc5

Keeping the Bishop on an active square, but the program "overlooks" a tactic. The safer move was 9...Be7 with Black still better.

10.e6+ dxe6 11.Qh5+ g6

Kicking the enemy Queen with the pawn is all-too-human an idea; better was probably 11...Qg6. Black has frittered away its advantage with the last three moves.

12.Qxc5 c6 13.Nc3 Qxc2

Pawn-grabbing is a known failing of early and weaker chess programs. Lack of development and King safety are more important issues.

14.Re1 Nf6 15.Bg5 Nd5 16.Qd4 Bd7

17.Nxd5 cxd5 18.Qf6+ Kg8 19.Bh6 Qxb2 

Putting off checkmate for a very short time.

20.Qxb2 Kf7 21.Qg7+ Ke8 22.Bg5 Rf8 23.Qe7 checkmate

Jerome Gambit: The Queen Goes Where She Will

One thing that seems to give defenders a sense of ease is the number of times White's Queen moves in the Jerome Gambit. It is important that such a feeling be translated into further strong moves, not inattention or overconfidence. View the following game.

Wall, Bill - PerGranBom
lichess.org, 2017

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

4...Kxf7 5.Nxe5+ Nxe5 6.d4 Bxd4 7.Qxd4 Nc6

An interesting alternative to the more usual 7...d6 (although the game transposes).

8.Qd1

Bill has moved the Queen to different squares in other games:

8.Qd2 d6 9.Nc3 Be6 10.O-O g6 11.f4 Nge7 12.Nd5 Bxd5 13.exd5 Nb8 14.f5 gxf5 15.Qe2 Qd7 16.Bh6 Ke8 17.Rae1 Rg8 18.Rxf5 Kd8 19.Rf7 Re8 20.Bg5 c6 21.Qh5 cxd5 22.Rfxe7 Rxe7 23.Rxe7 Qb5 24.Qf7 Nd7 25.Rxd7+ Kc8 26.Rd8 checkmate, Wall,B - 1063314, lichess.org, 2017;

8.Qd5+ Kf8 9.Nc3 Nf6 10.Qc4 d6 11.O-O Ne5 12.Qb3 c6 13.f4 Nf7 14.Be3 Qe7 15.Rae1 Be6 16.Qb4 a5 17.Qd4 Ng4 18.Bc1 c5 19.Qd3 h5 20.h3 Ngh6 21.f5 Bd7 22.Nd5 Qd8 23.f6 g6 24.Ne7 Ne5 25.Qxd6 Nhf7 26.Qxc5 b6 27.Qd5 Be8 28.Bf4 Qxd5 29.exd5 Bb5 30.Rf2 Nd7 31.Nxg6+ Kg8 32.Nxh8 Kxh8 33.Re7 Kg8 34.Be3 Rc8 35.b3 Nde5 36.Bxb6 Black resigned, Wall,B - Guest7503555, PlayChess.com, 2017;

8.Qd3 Nf6 9.Nc3 d6 10.O-O h6 11.f4 Re8 12.Bd2 Kg8 13.Rae1 a6 14.Nd5 Be6 15.Bc3 Bxd5 16.exd5 Nb8 17.Bxf6 gxf6 18.Qg6+ Kf8 19.Qxh6+ Kg8 20.Qg6+ Kf8 21.Re5 dxe5 22.fxe5 Nd7 23.exf6 Black resigned, Wall,B - Number555777, lichess.org, 2017; and

8.Qc4+ Ke8 (8...Kf8 9.O-O Qf6 10.Nc3 Qd4 11.Qe2 Nge7 12.Nb5 Qb6 13.Be3 Qa5 14.c4 a6 15.Bd2 Qb6 16.Qf3+ Kg8 17.Qb3 axb5 18.c5+ Black resigned, Wall,B - Guest526975, PlayChess.com, 2016) 9.O-O d6 10.Nc3 Qe7 11.Re1 Ne5 12.Qd4 c5 13.Qd2 Nf6 14.f4 Nc4 15.Qe2 Be6 16.e5 Bg4 17.Qxc4 Black resigned, Wall,B - Kas55, lichess.org, 2016

8...Nf6 9.Nc3 d6 10. O-O Re8 11.Bf4

Black is doing well, but he is tempted by the "free" e-pawn. The discovered attack on White's Queen that evolves turns out to be harmless.

11...Nxe4 12.Qd5+ 

On revient toujours a son premier amour.

12...Be6 13.Qxe4 g6 14.Rae1 Bxa2 15.Qxe8+ Qxe8 16.Rxe8 Rxe8 17.Nxa2

This was not what Black had envisioned. Now he is the one with a pawn for a piece.

17...Nd4 18.Be3 Nxc2 19.Rc1 Nxe3 20.Rxc7+ Re7 21.Rxe7+ Kxe7 22.fxe3 d5 23.Nc3 Kd6 24.Nb5+ Black resigned

Jerome Gambit: Fighting The Annoying Defense

Chessfriend Vlasta faces the "annoying defense" and shows that while White does not have have a lot to work with, neither does Black. The game is an extended battle, after which the point is split. It is not clear who was satisfied with the result, the defender who nicked his half point against a refuted opening, or the attacker who "survived" playing a refuted opening.

Vlastous - NEWMAN 1982
Internet, 2017

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

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6

So annoying. Black gives back the piece and drains a lot of dynamism out of the position.

8.fxe5 dxe5 9.Qh3+ Ke7 10.Qg3 Ke8

Vlasta has faced 10...Kf7 as well, in Fejfar,V - Goc,P (1/2-1/2, 70); Fejfar,V - Chvojka (0-1, 32); Vlastous 2456 - irinat 2597, ChessManiac.com, 2016 (0-1, 38); and Fejfar,V - Svoboda/corr Czech Cup, 2016 (1-0, 30). 

11.Qxe5+ 

A small improvement over his earlier 11.Nc3 in Fejfar,V - Kyzlink, corr Czech Republic, 2015 (1-0, 20). 

11...Qe7 12.Qxe7+ Nxe7 

Now we have a battle between the extra two pawns and the extra piece in a Queenless middlegame. White cannot claim an advantage - but how is Black to win?

13.c3 Ng6 14.d4 Be7 15.Be3 Rf8 16.Rf1 Rxf1+ 17.Kxf1 b6 18.g3 Ba6+ 19.Kf2 Bd3 20.Nd2 Kd7

21.Bf4 Rf8 22.Ke3 Ba6 23.Nf3 Bb7 24.h4 h5 25.c4 Rd8 26.Rf1
Re8 27.b3 Bf6 

Black's Bishop pair is focused on White's pawn center which has a protected passed pawn.

28.e5 Be7 29.Ng5 Rf8 30.Rf2 Ba3 31.e6+ Kc8 32.Kd3 Nxf4+ 33.gxf4 Kd8 34.f5 Ke7 35.Rf1 Bd6 

Black seems unsure what to do. One idea, suggested by Stockfish 8, is to undermine the support of White's advanced passer, while exchanging Rooks, i.e. 35...g6 36.fxg6 Rxf1 37.g7 Rf8 38.gxf8=Q+ Kxf8. 

 36.Nh7 Rh8 37.Ng5 Rg8 38.Ke3 Bg3 39.d5 Bxh4 40.Kf4 c6 

Interestingly enough, this attempt at undermining support is not successful, in that it opens a line for White's Rook. The piece activity allows White to give up a second pawn.

41.d6+ Kxd6 42.Rd1+ Kc5 43.Ne4+ Kb4 44.Rd7 Bc8 45.Rxa7 b5 46.cxb5 cxb5 47.Rc7 Ba6 48.Rf7 Ka5 49.Ke5 Kb6 

50.e7 Re8 51.f6 gxf6+ 52.Nxf6 Bxf6+ 53.Rxf6+ Kc5 

Black decides to surrender his Bishop for the advanced passed pawn, submitting to the draw.

54.Rxa6 Rxe7+ 55.Kf4 Kb4 56.a4 Rb7 57.axb5 Rxb5 58.Kg3 Kxb3 59.Kh4 Kb4 60.Rh6 Rc5 61.Rxh5 Draw

Jerome Gambit: Too Much of A Good Thing?

In the particular Jerome Gambit line in the following game, Black is faced with returning one of the two pieces that White sacrificed. He has two general ways to respond - choose a piece to withdraw from danger, or ignore the attack and develop another piece. Either option could work. However, trying both leads to great danger. 

Wall, Bill - Shookspear, William
lichess,org, 2017

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.d4 Bb4+ 

I like to imagine that Black is feeling clever here, knowing that he has to give back either the Knight or the Bishop, and deciding to have the Bishop capture on c3, taking a pawn, before saving the Knight. There are 255 game examples in The Database.

7.c3 Nf6

Black reverts to option number two: focus on development and let White mess up his pawn structure by capturing a piece. The move ...Nf6 would be better played while leaving the Bishop on c5, however.

8.dxe5 Bxc3+ 

According to plan, but it is not going to be successful.

9.Nxc3 Ne8 

I was surprised to find a related game in The Database, with 9...Ng8, and the outcome - a win for Black - was a bit of a shock, but play could be improved for both sides:  10.Qd5+ Ke8 11.O-O Ne7 12.Qc4 c6 13.Bg5 d5 14.exd5 cxd5 15.Nxd5? (15.Qb5+) Be6? (15...Qxd5) 16.Nxe7 (16.Nc7+) Bxc4 White resigned, Idealist - HarryPaul, FICS, 2002. 

10.O-O Rf8 11.Qh5+ Kg8

Feeling good. White has helped Black castle-by-hand.

12.Bg5 g6 13.Qh4 Black resigned

Oh, dear. Black can try 13...Nf6 14.Bxf6 Rxf6 (if 14...Qe8 then 15.Nd5) 15.exf6 but that would leave him down a Rook and a pawn.

Chess is Difficult: All You Have To Do Is Win

Chess is difficult.

The following game quickly reaches a position which has a score of 72% for White in The Database. There is a rating difference of about 600 points between the players - favoring White.

You would think - especially in a blitz game - that the first player would have an overwhelming advantage. Indeed, he checkmates his opponent in 16 moves.

But - if you turn the game over to Stockfish 8 on its "blunder check" setting, the story quickly becomes more complicated.

Please note, I mean no offense to the players. None of us are computers. I use Stockfish 8 to show just how deep apparent shallow waters are... And to show how a checkmate attack can be either pursued or avoided.

lcuartas - diegogs
10 0 blitz, FICS, 2016

1.e4 Nc6 2.Nf3 e5 3.Bc4 Na5 

Anyone who plays the Jerome Gambit has to be prepared for this offside move. I usually recommend the typical Jerome response, although the alternative, 4.Nxe5, is fine, too.

4.Bxf7+ Kxf7 5.Nxe5+ Ke6 

White has sacrificed a piece for two pawns, and the computer gives him a small advantage.

6.Qg4+ Kxe5 7.d4+ Kd6 

Black can afford to take the second piece, as scary as it seems, but he would choke on the d-pawn: 7...Kxd4 8.Be3+ Ke5 9.Qf5+ Kd6 10.Bc5+ Kc6 11.Qd5+ Kb5 12.Nc3+ Ka6 13.Qd3+ Nc4 14.Qxc4+ b5 15.Qxb5 checkmate.

You can tell that Black's King is in danger, but working it out during the game can be quite a challenge - especially with the clock ticking. 

8.e5+ Kd5 

This is an error. Instead, the computer recommends a convoluted line to a surprising equality: 8...Kc6 9.d5+ Kb6 10.Be3+ c5 11.d6 Nh6 12.Qe4 Bxd6 13.exd6 Re8 14.Qd5 Qh4 15.Qxc5+ Ka6 16.Nc3 b6 17.Qb5+ Kb7 18.O-O-O Qc4 19.Qxc4 Nxc4 20.Bxh6 gxh6 21.Rd4 Na5 22.Rf4 Re6 23.Rd1 a6 24.b3.

I don't expect club level humans to be able to play that way. Note, though, the White pawn advance to d6 to keep the c7 square covered, to avoid allowing Black's King to escape. Also, that White seems to be scampering to keep the game even and avoid falling behind.

9.c4+ 

What is going on? White continues to harass the enemy King, overlooking a chance to acquire the enemy Queen: 9.Qf3+ Ke6 10.d5+ Ke7 11.Bg5+ Ke8 12.Bxd8 b6 13.e6 Nf6 14.Bxf6 gxf6 15.d6 Bxd6 16.Qxa8 Kd8 17.exd7 Kxd7 18.O-O Re8 19.Qf3 Ba6 20.Rd1 Kc8 21.Qa8+ Kd7 22.Qd5 Kc8 23.Nc3 Kb8 24.Ne4 Rd8. Nice cooperation between pawns and pieces. 

9...Nxc4 

Now the pendulum has swung again. The King needed to move, it wasn't time for a pawn grab: 9...Kc6 10.Qd1 a6 11.c5 d5 12.Qa4+ b5 13.Qxa5 Kb7 14.a4 Bd7 15.Be3 Ne7 16.Qd2 Nf5 17.O-O Be7 18.Nc3 Bc6 19.axb5 axb5 20.Na2 Qe8 21.Nb4 Qg6 22.g3 Rxa1 23.Rxa1 Ra8 24.Rxa8 Kxa8 25.Qd3 and Black has an edge, as White has not been able to corral the enemy monarch. 

10.Nc3+ Kc6 11.d5+ Kc5 

Those pesky "Jerome pawns"! The King is now in a mating net.

12.a3 

Planning to add the b-pawn to the mix, but it could have been added directly, as capturing it would allow the White Rook into the game, too: 12.b4+ Kb6 13.Qxc4 c6 14.d6 Bxd6 15.exd6 Qe8+ 16.Be3+ Qxe3+ 17.fxe3 a5 18.Qc5+ Ka6 19.Qxa5 checkmate.

12...Nxe5 

This should lead to immediate disaster, as opposed to longer-term disaster.

13.b4+ 

As planned. Also 13.Qb4# 

13...Kb6 

Chess is not fair. Black continues along the path to destruction, while he had a chance to escape it: 13...Kd6 14.Qe2 c6 15.Bf4 Kc7 16.Bxe5+ d6 17.Rc1 Qg5 18.Nb5+ Kd8 19.f4 Qg4 20.Nxd6 Bxd6 21.Qxg4 Bxg4 22.Bxd6 cxd5 23.Bc7+ Ke8 24.Be5 Nf6 25.O-O Rc8 26.h3 Be6 27.Rxc8+ Bxc8 28.Rc1 Kd8 29.Bc7+ with advantage to Black. 

14.Be3+ 

A perfectly reasonable move, adding a piece to the attack. The soulless comuter says that the move leads to a Black advantage, however, and prefers: 14.Na4+ Kb5 15.Qe2+ Nc4 16.Nc3+ Kb6 17.Qxc4 Qe7+ 18.Be3+ c5 19.d6 Qxd6 20.bxc5+ Kc7 21.cxd6+ Kd8 22.Nb5 Nf6 23.Bxa7 Rxa7 24.Nxa7 Bxd6 25.Qxc8+ Ke7 26.Qxh8 Kf7 27.O-O Bc5 28.Nb5 b6 29.Rfe1 Kg6 White is winning.

14...Ka6 

Of course. Still, the silicon beast prefers: 14...c5 15.dxc6+ Kc7 16.Nb5+ Kb8 17.c7+ Qxc7 18.Qd4 Nc6 19.Bf4 d6 20.Qe3 Qb6 21.Qg3 Bg4 22.Nxd6 g5 23.Bxg5 Qc7 24.Nf7 Qxg3 25.hxg3 Bg7 26.Rc1 a6 27.Nxh8 Bxh8 28.Rxh7 Bb2 29.Bf4+ Kc8 30.Rc2 when Black is better. 

15.b5+ Ka5 16.Qa4 checkmate

There you have it. The result is exactly as expected. Another example of a bold, sacrificial attack winning. 

How often do we wander through life, feeling propelled, overlooking alternatives we are not even aware are there?

Chess is a peek at the multi-verse.

Fascinated by the Jerome Gambit

In the book review of The Art of the Checkmate by Georges Renaud and Victor Kahn at the online Chessentials site, the  Jerome Gambit game NN - Blackburne (given as taking place in "London tournament, 1880", but this is likely in error) was presented, with the note

The author calls this opening a Jerome’s Gambit. It is of course nonsense in modern chess, but I still remember how I played it almost regulary immediately after I read this book.

Of course, upon reading this I immediately left a note requesting to see any of the site master's games!

I suspect other chess players did not take Joseph Henry Blackburne's "lesson" to heart, and, instead, were fascinated by the play of "NN" and his Jerome Gambit.

I will share what games turn up.

Jerome Gambit: Winning Ugly is Still Better Than Losing Ugly (Part 2)

perrypawnpusher - IlToscano, Chess.com, 2017

Adopting a "Come and get me" attitude, I decided to let my opponent do the attacking. After all, he is playing against a refuted opening. Piece of cake, right?

16.h3 Be6 17.b3 Rhf8 18.Bd2 Rae8 19.O-O-O Kd7 20.Rhf1 Kc8

So, who is the Jerome Gambit "expert"? Black has developed all his pieces and has even castled-by-hand on the Queenside. He clearly is better.

It was hard for me to be patient (keep the psychological onus on my opponent) and keep the position as closed up as possible. With his next move, my opponent indicated he wanted to move on the Kingside.

21.Kc2 Nh5 22.Ng5

Starting my own incursion. Plus, I had been reading James Schuyler's Your Opponent Is Overrated, and he had been writing about the "superfluous piece" - and I decided that my Rook on d1 fit that category, so I thought a Rook exchange would help a bit.

22...Rxf1 23.Rxf1 Bg8 

So, the "extra" White Rook has "disappeared" from d1, and Black only has one developed Rook, instead of two. Actually, I'm not sure how much that matters - and see my 26th move.

24.c4 h6 25.Nf3 Ng3 26.Rd1

Black's Knight goes back to its old post, White's Knight goes back to its old post, and White's Rook goes back to its old post. So, what are we going to do next?

26...g5 27.Bc3 Be3 28.Re1 Bf4 29.Bd2 Bxd2 30.Kxd2 c6 

I couldn't figure out what to do with my dark-squared Bishop, and it seemed less effective on the a1-h8 diagonal than Black's Bishop on the c1-h6 diagonal, so I wound up trading them off. At least Black's remaining Bishop is limited by my pawn wall.

31.a3 Kd7 32.b4 a6 33.a4 Kc7 34.a5 Nh5 

Oh, bother, not again.

35.h4 

Either the Kingside will get closed up again, or a few pawns will get exchanged. Despite my earlier decisions, I took some comfort in the age-old recommendation: When ahead in material, exchange pieces; when behind in material, exchange pawns.

35...g4 36.Nd4 Nf4 37.Rf1 Nxg2 38.Rg1 Nf4 39.Rxg4 Ne6 40.Nf5 h5 

The Kingside is closed up again. If anything, it is more open to White's pieces infiltrating.

41.Rg6 Kc8 

An oversight. My opponent had taken several "vacations" from play, and may have come back to the board mixing up variations in his head. Or maybe he figured there wasn't much to the Jerome Gambit, so why sweat it...

42.Nxd6+ Kd7 43.Nxe8 Black resigned

Unfortunate in a couple of ways. Black had played very well up until his 41st move. He also missed a chance (he hardly could have known) to allow me to make a mistake, myself. After 43...Bf7!? I thought I would have to return the exchange, and had planned 44.Rxe6?!+ Kxe6 when White would be a couple of pawns ahead, but the game would still be interesting.

Looking deeper after the game, with the help of Stockfish 8, I know now that 44.Nf6+ Ke7 45.Rh6 would allow White to stay the exchange (plus a couple of pawns) up, but the locked-in position of the Rook had kept me from looking further at that line.