Saturday, April 15, 2023

Jerome Gambit: What the Rest of the Game Is For

If chess players were guided by "objectively best" assessments of the positions they were playing, there would be no Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) or its related openings: after White's Bishop sacrifice, he could just resign.

In whatever way the positions are evaluated, however, play continues - that is what the rest of the game is for.

The following game is a good example. White persists, despite facing prevailing wins, because he is well aware that in club level blitz play, anything can happen.


Robepersky - JohnInWales

10 0 blitz, FICS, 2023

1.e4 e5 2.Nf3 Nc6 3.Bc4 h6

The Semi-Italian Opening.

4.O-O Bc5 5.Bxf7+ 

The Semi-Italian Jerome Gambit.

Black's hope to protect his himself from the Fried Liver Attack - 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.Nxf7 - by making the g5 square "off limits" for enemy pieces is foiled by White's Jerome Gambit treatment.

For some background see "A Jerome Look At The Semi-Italian Opening (Parts 1 & 2)" and "Semi-Italian Jerome Gambit"

According to The Database, Robepersky is 17 - 8 with this sacrifice. 

5...Kxf7 6.Nxe5+ Nxe5 7.Qh5+ 

By adding O-O and ...h6 to the normal Jerome Gambit line, 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Qh5+, White has improved his position by about a pawn's worth, according to Stockfish 15.1. 

7...Ke6 8.c3 

This preparatory move is Robepersky's choice, and The Database shows that it has scored 5 - 5, but more dynamic for White would be 8.Qf5+ Kd6 9.b4!?

8...Qg5 

Black calmly offers to exchange Queens, to dampen White's attacking chances.

Robepersky has also faced

8...Qf6 9.d4 Bxd4 10.cxd4 Nf7 11.Qd5+ Ke7 12.e5 Qe6 13.Qc5+ d6 14.Qxc7+ Qd7 15.exd6+ Nxd6 16.Re1+ Kf8 17.Qc5 b6 18.Qb4 Bb7 19.Bf4 Re8 20.Bxd6+ Kf7 21.Qc4+ Kf6 22.Be5+ Ke7 23.Bxg7+ Kd8 24.Rxe8+ Qxe8 25.Bxh8 Bc6 26.Nd2 Kc7 27.Be5+ Kb7 28.a4 Ne7 29.a5 Nd5 30.a6+ Kc8 31.Rc1 Ne7 32.d5 Kd7 33.dxc6+ Nxc6 34.Qxc6+ Ke7 35.Bd6+ Kf7 36.Qf3+ Ke6 37.Re1+ Kxd6 38.Rxe8 Black resigned, Robepersky - EddyTorrance, 5 0 blitz, FICS, 2021;

8...Be7 9.Qf5+ Kd6 10.d4 Bf6 11.dxe5+ Bxe5 12.Rd1+ Ke7 13.Qxe5+ Kf8 14.Nd2 d6 15.Qf4+ Nf6 16.e5 dxe5 17.Qxe5 Nd7 18.Qf4+ Kg8 19.Qc4+ Kf8 20.Nf3 Qf6 21.b3 Ne5 22.Ba3+ Ke8 23.Nxe5 Rh7 24.Nf3 Be6 25.Re1 Kd8 26.Rad1+ Ke8 27.Rxe6+ Qxe6 28.Qxe6 checkmate, Robepersky - Tenet, 10 0 blitz, FICS, 2022; and

8...Nf6!? 9.Qh3+ Kf7 10.d4 d5!? 11.dxc5?! (11.Qg3 Re8) 11...Bxh3White resigned, Robepersky - tutekk, blitz, 5 2 blitz, FICS, 2023 

9.Qxg5 hxg5 10.d4 


White achieves his pawn fork, but, to be fair, it is not enough to alter the fact that Black is better.

That is what the rest of the game is for.

10...Nf3+ 

A spririted idea: if Black needs to return a piece, why not do so to break up White's castled position?

11.gxf3 Be7 12.Nd2 Bd6 

Yes, the Bishop is safer on e7, but more aggressive on d6.

13.d5+ 

I would have sent the Bishop scurrying back to e7 with 13.e5.

13...Ke7 14.f4 

Clearance sacrifice.

14...gxf4 15.Nf3 


The Knight takes up guard duty on the Kingside, keeping a special eye on h2, which is threatened by the enemy Rook.

15...Nf6

Business as usual. In a blitz game, this kind of thing happens. More in line with the demands of the position was 15...Kf8. 

16.e5 Bxe5 17.Nxe5 Ng4 


Clock? Nerves? Caffeine deficiency disorder?

Instead, after the routine 17...Re8 18.c4 d6 19.Ng6+ Kf7 20.Nxf4 the game would be even, according to Stockfish 15.1 (34 ply) despite White's extra pawn.

Black resigned


Friday, April 14, 2023

Jerome Gambit: And Then There Were Four


The fifth and final round of the "Giuoco Piano Game" tournament at Chess.com - see "Jerome Gambit: Rescued?" for an earlier reference - has started, and there are four players left: myself (perrypawnpusher), auswebby, DouglasEngle, and xyz7.

I am playing the Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) in my games with White, as I have throughout the tournament.

In the past, asuswebby had defeated my Jerome twice (see "Toward Disaster" and "Jerome Gambit: What Can We Learn? (Part 1)") so I am hoping to show that third time's the charm.

As always, I wish good chess to my opponents.

Thursday, April 13, 2023

Jerome Gambit: Has ChatGPT Learned Anything?

 


A couple of months ago I visited the online website for ChatGPT, an artificial intelligence resource, and I asked it about the Jerome Gambit and the Blackburne Shilling Jerome Gambit. The results were mixed. (Of course, knowing about these openings is not clearly a sign of "intelligence", artificial or otherwise.)


I visited again today. Strangeness still invaded our discussion. Here is how it went. I have added some comments in blue.

Please analyze the Jerome Gambit. Is it good for White? 
The Jerome Gambit is a chess opening that arises after the moves 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.b4. [Of course, this is not the Jerome Gambit, it is the Evans Gambit - Rick]
This gambit is named after the American chess master Jules Jerome, who popularized it in the late 19th century. [Last month ChatGPT attributed the opening to "the French player Jerome Michel". The Jerome Gambit was popularized by Alonzo Wheeler Jerome - Rick]
In this opening, White sacrifices a pawn in order to gain control of the center and develop their pieces quickly. The idea behind the gambit is to create pressure on Black's position and limit their options, while also potentially opening up lines for White's pieces to attack. 
Whether the Jerome Gambit is "good" for White or not depends on a few factors, including the skill level of the players, their respective styles, and the specific variation played. [Yes - Rick]

 At the highest levels of play, the Jerome Gambit is generally not considered a very strong opening for White, as Black has several strong responses that can neutralize White's initiative and potentially leave them with a weakened position. [Yes, again - Rick] 

One common response is 4...Nf6, which attacks the pawn on e4 and forces White to either defend it or allow Black to capture it with their knight. [Sorry, this makes no sense. In the line ChatGPT gives, above - the Evans Gambit - 4...Nf6 loses a piece to 5.bxc5. In the Jerome Gambit, 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7, of course, 4...Nf6 is not possible - Rick]  

Another option for Black is 4...dxc6, [Not possible - Rick] which gives up the bishop pair but also gains a material advantage. [Huh? - Rick] 

However, at lower levels of play, the Jerome Gambit can be an effective surprise weapon, as many players are not familiar with the nuances of the opening and may make mistakes that allow White to gain an advantage. In particular, if Black is not familiar with the opening and does not respond accurately, they may find themselves under early pressure and struggling to find good moves. [Very true! - Rick] 

Overall, while the Jerome Gambit may not be considered the strongest opening for White at the highest levels of play, it can still be a viable option for players looking to mix things up and catch their opponents off guard. However, it's important to study the opening carefully and be prepared for a variety of responses from Black. [Wise counsel - Rick]

Wednesday, April 12, 2023

Jerome Gambit: Search and Find



I am always on the lookout for Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) videos, usually on YouTube.com

The latest is "italian game jerome gambit (as black).#chessgame #chessvideoplus #checkmate #playchess #chessboard" presented by @ChessNovice who has a website with a collection of chess videos.

The game is a bit odd, with White looking a bit like someone who absent-mindedly steps into traffic... There is no notification as to who won, although the final diagram surely favors Black.

Still, wins or losses, it is always educational to play over games and quiz yourself, What would I do here?, Why did he do that there?, What would be a good plan?, etc.


Anonymous - Anonymous

2023

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 

7.Qf5+ Kd6 8.Nc3 c6 9.O-O Qf6 

10.d3 Qxf5 11.exf5 Bd4 12.Ne4+ Kc7 13.Be3 Bb6 14.Bxb6+ axb6 15.Rfe1 Nf7 16.Rad1 Nf6 17.Nxf6 gxf6 18.Re7 Nd6 

Black has two pieces for two pawns.


Tuesday, April 11, 2023

Jerome Gambit: The Quick and the Dead



Apologies to Joy Williams and Sharon Stone, but the following game can be referred to as "the quick and the dead".

Blitz chess can be deadly. Quickly.


gabrielebattaglia - Sumy

5 3 blitz, lichess.org, 2023


1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+  

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ng6 

7.Qd5+ Ke8 8.Qxc5 N8e7 9.O-O b6 10.Qe3 Rf8 


At first it looks like Black is planning on castling-by-hand, but that never comes about. 

11.Nc3

The lichess.org computer labels this as an inaccuracy, preferring 11.f4

My guess is that the text move was played to prevent 11...d5. Indeed, in the game Black is dissuaded from playing the move, although Stockfish 15.1 sees it as still okay, i.e. 11...d5 12.Nxd5 Nxd5 13.exd5+ Kf7 and the King castles-by-hand after all. With Black's 3 Queenside pawns holding back 5 of White's, at least for a while, Black with the extra piece, is better.  

11... d6 12.d4 Bb7 13.Qg3 Nc6  

Despite the iffy position of Black's King, the game is in balance, or even slightly better for the second player.

14.d5 Nd4

The Knight is up to mischief. It can grab the c-pawn and escape, but the time spent would allow White to improve his attack on the King.

15.Qd3

Trapping the Knight. It is hard in a 3-minute game for it to find a way out, but there is...

With more time White might have investigated the recommended mess 15.Bg5 Qd7 16.Rad1 Nxc2 17.f4 Kf7 (17...Ba6 18.Rf3 Kf7 19.h4 h6 20.h5 Ne7 21.Bh4 Kg8 22.Rd2 Nb4)18.h4 h6 19.f5 Ne5 20.Bf4 Nb4.

I would play the text in a heart beat.

15...Ne5

Giving up a bit too soon. 

There was 15...Qf6 16.f4 c5 17.dxc6 ep Nxc6, saving the piece.

Even better was giving up the piece with 15...Nf3+!, e.g. 16.gxf3 Ne5 and White's King is at great risk.

16.Qxd4 


Having gained the material advantage, White is now fine with exchanges.

16...c5 17.dxc6 Bxc6 18.f4 Ng4 19.h3 Nf6 20.e5 dxe5 21.Qxe5+ Qe7 22.Qxe7+ Kxe7 


White is two pawns up. Remove the Rooks and the Knights and there is the tiniest chance for Black to move into a Bishops-of-opposite-colors endgame and hope for a draw...

Not going to happen. 

23.Be3 Nd5 24.Nxd5+ Bxd5 25.Rad1 Be4 26.c3 Rf6 27.Bc1 Rg6


 As a general rule, in a Bishops-of-opposite-colors middlegame, the side that can create an attack on the King has the advantage. Here, though, White has plenty of resources.

28.g4 Kf8 29.Rfe1 Bf3 30.Rd3 Bc6 31.b3 h5 32.Ba3+ Kg8 33.g5 b5 34.Rde3 Rd8 35.Bc1 Rf8 36.Re7 Rxf4

This slip ends the game.

37.Bxf4 Black resigned






Monday, April 10, 2023

Jerome Gambit: Win - Lose - Win



angelcamina, a longtime Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) player who specializes in 1-minute bullet games sent me the following game with the notation "Win - Lose - Win".

A player's fortune can vary widely across those 60 seconds of chess battle, although I think you will agree when you play over the following game that White had things under control, wherever it went.

The checkmate at the end was serious Win.


angelcamina - Maverikk

1 0 bullet, lichess.org, 2023


1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ g6 7.Qxe5 Bxf2+ 


Maverikk tries to surprise his opponent. Unfortunately for him, angelcamina has faced the line before.

From a post earlier this year, "Jerome Gambit: It's the Check that Counts"

I recently (in "Jerome Gambit: Gift Horses") mentioned the "Counter-Jerome Defense" or the "Counter-Jerome Gambit" as a strategy that defenders might use... 

The idea is to follow up 8.Kxf2 with 8...Qf6+ and exchange of Queens. In effect Black returns a sacrificed piece to stifle White's brutal attack and force him to play an (allegedly) painful pawn-up game.

8.Kxf2 Qf6+ 9.Qxf6+ Kxf6 


According to The Database this variation is less common than 9...Nxf6.

From a practical point of view, in 28 games 9...Kxf6 scores 66% for White, which is somewhat more favorable than White scoring 75% in the 235 games with 9...Nxf6

However, Stockfish 15.1 scoffs at this preference, evaluating the Knight capture as being about 2 pawns better than the King capture. Apparently the more active placement of His Majesty in the endgame / Queenless middlegame is outweighed by the danger he faces.

Two of angelcamina's games: 9...Nxf6 10.d3 (10.Nc3 Kg7 11.Rf1 d6 12.Kg1 Be6 13.d4 Bc4 14.Bh6+ Kxh6 15.Rxf6 Rhe8 16.b3 White won on time, angelcamina - jonas99, 1 0 bullet, lichess.org, 2021) 10...d6 11.Rf1 Kg7 12.b3 Bg4 13.Bb2 Rhf8 14.Kg3 Bd7 15.Nd2 Kg8 16.Rxf6 Rxf6 17.Bxf6 Rf8 18.Bd4 c5 19.Bc3 b5 20.Rf1 Re8 21.Ba5 b4 22.Bc7 Re6 23.Nc4 Bb5 24.Nxd6 a6 25.Nxb5 axb5 26.e5 c4 27.dxc4 bxc4 28.bxc4 Rc6 29.Bd6 Rxc4 30.Rf2 Rc3+ 31.Kf4 Rc4+ 32.Ke3 White won on time, angelcamina - ShadowKnight314, 1 0 bullet, lichess.org, 2020

10.Rf1 Kg7 

Loss of time for safety sake.

11.b3 d6 12.Bb2+ Nf6 


White is clearly better. His next move - bypassing the safer castling-by-hand - shows his comfort with the  position.

13.Ke3 Rf8 14.Nc3 Be6 15.d4 c6 16.Rae1 Rf7 17.e5 


The other center pawn was also ready to advance.

17...dxe5 18.dxe5 Nd5+ 

Missing the off ramp. Instead, 18...Ng4+ would have served him better - 19.Ke2 Rxf1 20.Rxf1 Nxh2 21.Rh1 Ng4 22.Ne4 and the game would have evened out. 

19.Nxd5 Bxd5 


Missing something.

20.e6+ Kg8 21.exf7+ Kf8 

Blockading the pawn - instead of capturing it - is a matter of choosing how to drop the full point.

22.Kd3 Rd8 23.Ba3+ Kg7 24.Re8 Bxf7+ 25.Kc3 Bxe8 26.Bf8+ Kg8 27.Bh6 Rd7 28.Rf8 checkmate


Ouch. Very Nice.



Sunday, April 9, 2023

Jerome Gambit: Opening Delayed, Ending Accelerated



In a 1-minute bullet game, a better knowledge of the opening is helpful, but what is more important is making good - not necessarily "great" - moves quickly. 

It also helps to have calm nerves, as the speed of the game accelerates and your clock's flag is hanging...

The following game is an exciting example.


angelcamina - Orla99

1 0 bullet, lichess.org, 2023


1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 

The Two Knights Defense.

4.d3 Bc5 

Now we have a quiet Italian Game.

5.Bxf7+ 

Jerome-ized.

5...Kxf7 6.Nxe5+ Nxe5 7.d4

Here we have a Jerome Gambit where it has taken White two moves to get his pawn to d4, giving Black the chance to develop his King's Knight in the meantime.

There are 44 examples in The Database.

Not suprisingly, Stockfish 15.1 (36 ply) sees the position as about 2 1/2 pawns better than the regular Jerome Gambit position, 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.d4, where it prefers Black.

The equalizing factor in this situation is that, according to The Database, angelcamina has played 269 1-minute Jerome Gambits, scoring 86%. He knows what he is doing.

7...Bb4+ 8.c3 Bxc3+ 


Black has to give back one of the two sacrificed pieces. This is his solution.

Simpler and stronger would have been 7...Bxd4.

9.bxc3 

White could have helped his development with 9.Nxc3, but he was thinking about pawn play against the enemy Knights.

9...Ng6 10.e5 Re8 11.O-O Nd5 12.Qb3 c6 13.c4 Ndf4 

14.g3  

The Knight is vulnerable, but the solution takes a bit longer: 14.Bxf4 Nxf4 15.Qf3 Qg5 16.g3 Rf8 17.Qxf4+ Qxf4 18.gxf4 Ke7 with about an even game.

14...Ne2+ 15.Kg2 Nxc1 

15...Nxd4

16.Rxc1 Kf8 17.Nc3 

17...Qg5 

The clock is ticking. As the time available shrinks, so does the depth of analysis. Things get wild.

White shows his skill and experience.

18.f4 Qg4 19.Rf1 c5 20.f5 Nh4+ 21.Kg1 Nf3+ 22.Kg2 Nd2 23.Qa3 Nxf1 24.Qxc5+ Kg8 25.Rxf1 b6 26.Qd5+ 


The enemy Rook hangs.

26...Kh8 27.Rf4 

Very nice. Instead, 27.Qxa8 Qxd4 would have given Black equalizing chances.

Now the pieces fly. Beat the clock.

27...Qh5 28.Qxa8 d5 29.Qxd5 Re7 30.Qd8+ Re8 31.Qd5 Bb7 32.Qxb7 g6 33.f6

33...Rf8 34.Qe7 Qh6 35.Qxf8+ Qxf8 36.e6 h6 37.e7 Qf7 38.Nd5 Kh7 39.Re4

White won on time

Whew.