Rybka Plays the Jerome Gambit

Here is the game from the Kasparovchess forum mentioned by Philidor1792 in the previous post, with Rybka matched against a player from the site.

Rybka - Mustitz
10 5 casual game, 2010

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

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

I have nicknamed this line "the annoying defense" as Black gives back a piece and drains the position of much of its dynamism.

It is ironic that White, a computer in this game, must face a variation very popular with computer defenders.

8.fxe5 dxe5 9.Qh3+ Kf7 10.Qh5+  Ke6 11.Qh3+ Kf7 12.Qh5+ g6

Rybka would be content with a draw by repetition. Mustitz would not.

13.Qxe5 Qe7 

This move is new to The Database.

14.Qxe7+ Nxe7 15.c3 Nc6 16.d4 Nxd4 

17.cxd4 Bxd4 18.Nc3 Re8 19.Bd2 Bxc3 20.O-O+ Kg8 21.Bxc3 Rxe4 

This is an interesting position for both sides. White can be "happy" it is only a pawn down in a Jerome Gambit, while Black can be pleased to still be a pawn ahead. Certainly the signs of a possible draw via Bishops-of-opposite colors are present.

22.Rae1 Bf5 23.Rxe4 Bxe4 24.Rd1 Bf5 25.h3 Re8 26.Kf2 Be6 27.b3 Kf7 28.g4 c6 29.Kg3 Bd5 30.Rf1+ Ke7 31.Re1+ Kd7 32.Rxe8 Kxe8 

33.Kf4 h5 34.Kg5 hxg4 35.hxg4 Be4 36.Be5 Kd7 37.b4 Ke6 38.Bb8 a6 39.Bc7 Kd5 40.Kf4 Bb1 41.a4 Kc4 42.Bd6 b6 43.a5 bxa5 44.bxa5 Kd4 45.Be5+ Kd3 46.Bd6 Kd4 47.Be5+ Kd5 48.Bc3 c5 49.Be1 Kd4 50.Bf2+ Kc4 51.Ke5 Kd3 52.Bxc5 Kc4 53.Bf2 Drawn

Play Against the Weak King

The Jerome Gambit is more than a tricky opening involving the sacrifice of a piece or two. The goal of all that giving is to get the enemy King, as Philidor 1792 shows in the game below.

Philidor 1792 - guest826
5 0 blitz, www.peshka.me, 2015

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

4...Kxf7 5.0-0 Nf6 6.c3 Nxe4 7.d4

Alternately: 7.b4 Bb6 8.b5 Na5 9.Nxe5+ Kg8 10.d3 Nf6 11.Re1 d6 12.Nf3 Bg4 13.h3 Bh5 14.Nbd2 h6 15.c4 Bd4 16.Rb1 Bc3 17.Qc2 Bxd2 18.Bxd2 b6 19.Nd4 Qd7 20.Bxa5 bxa5 21.Qa4 Kh7 22.Qxa5 Rhe8 23.Nc6 Bg6 24.Qc3 a6 25.a4 axb5 26.axb5 Qf5 27.d4 Ne4 28.Qc2 Rf8 29.f3 Ng3 30.Qf2 Nh5 31.Ne7 Qg5 32.Rbd1 Nf4 33.Kf1 Nd3 White resigned, Philidor 1792- guest826, ww.peshka.me, 2015. 

7...exd4 8.cxd4 Bb6 9.d5 Nb8 

Or 9...Na5 10.Qc2 Re8 11.Bg5 Nf6 12.b4 Kg8 13.bxa5 Bxa5 14.Nbd2 b5 15.a4 Bb7 16.axb5 Bb4 17.Qb3 Bd6 18.Rfe1 h6 19.Rxe8+ Qxe8 20.Bxf6 gxf6 21.Qd3 Qh5 22.Ne4 Kg7 23.Nxd6 cxd6 24.Qd4 Kf7 25.Qe4 Bxd5 26.Qh7+ Kf8 27.Re1 Qg4 28.Qxh6+ Qg7 29.Qh5 Be6 30.Nd4 Rc8 31.g3 Bg8 32.Qf3 Qf7 33.Qf4 Bh7 34.Qxd6+ Kg8 35.Re7 Qf8 36.Qxd7 Rc1+ 37.Kg2 Bg6 38.Ne6 Be4+ 39.f3 Rc2+ 40.Kg1 Rc1+ 41.Kf2 Bf5 42.Re8 Bxe6 43.Qxe6+ Kg7 44.Qg4+ Kf7 45.Qh5+ Kg7 46.Qg4+ Kf7 47.Qh5+ Kg7 48.Rxf8 Rc2+ 49.Ke3 Kxf8 50.Qh8+ Kf7 51.Qh7+ Ke6 52.Qxc2 Kd7 53.Qh7+ Kd6 54.Qxa7 f5 55.b6 Kd5 56.b7 Ke6 57.b8Q f4+ 58.Kxf4 Kd5 59.Qb5+ Kd6 60.Qad7 checkmate, Deep Shredder 10 UCI - Rybka 2.3.1 32-bit, jeromegambit computer tournament, 2008. 

10.Nbd2 Nf6 

Or 10...Nxd2 11.Bxd2 d6 12.Ng5+ Kg8 13.Qh5 Qf8 14.Rae1 Bd7 15.Re4 h6 16.Rf4 Qe7 17.Re4 Qf8 18.Rf4 Qe7 19.Re4 hxg5 20.Qxh8+ Kxh8 21.Rxe7 Bb5 22.Ra1 Na6 23.a4 Bc4 24.Bc3 Rg8 25.Ra3 Bxd5 26.a5 Bc5 27.b4 Nxb4 28.Bxb4 Bxb4 29.Rh3 checkmate, Philidor 1792 - guest826, www.peshka.me. 2015. 

11.d6 cxd6 12.Nc4 d5 13.Nd6+ Kg8 14.Bg5 Qf8 15.Bxf6 Qxd6 16.Be5 Qe6 17.Re1 Qf7 18.Ng5 Qg6 19.Qxd5+ Kf8 20.Bd6+ Qxd6 21.Qxd6+ Kg8 22.Re8 checkmate

BSJG: Move-by-Move

The following game - another recent one from GameKnot.com - allows a closer look at an interesting, more-positional variation of an otherwise wild, attacking opening.

mr_kill - syiedan86
Team match, GameKnot.com, 2014

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4

The Blackburne Shilling Gambit.

There are a number of ways to safely meet the BSG, including 4.Nxd4, 4.0-0, 4.d3 and 4.c3. Black wants to see 4.Nxe5?!, so he can respond with the thematic 4...Qg5!?

4.Bxf7+ 

The Blackburne Shilling Jerome Gambit. As a Jerome Gambit fan, I like the move. Also, it has been good to me - I have scored 91% in 41 games. (The Database* contains 4,452 BSJG games;White scores 56%.)

4...Kxf7 5.Nxe5+ Ke8 6.c3

This move shows up 398 times in The Database, scoring 55% for White. The major alternative is 6.Qh5+, which appears in 1,376 games, and with which White scores 61%.

Houdini 3 shows a slight preference for the text move, which doesn't surprise me, as 100% of my games contain the alternative.

6...Ne6

This move appears in 157 games, while the alternate retreat, 6...Nc6, appears 92 times. The challenging 6...Qg5 appears in 95 games. All score the same, plus or minus a percentage point or two.

7.d4

Here Houdini shows a slight preference for 7.Qh5+, leading to an even game.

7...d6 8.Nd3 Nf6 

This is a balanced position. (It must be remembered that Black cannot castle, as he has moved his King.)

A couple of ideas for White now include 9.Nd2, seen in the game GmCooper - Mazetov, lightning, FICS, 2001 (1-0, 28); and 9.d5 Nc5 10.Nxc5 dxc5 11.O-O, which is the choice of chess engines Houdini, Rybka and Stockfish.

9.e5 dxe5 10.Nxe5 Bd6 11.Qe2

Stronger might be 11.f4

11...Qe7 

In this tense position, White inexplicably dropped a piece, and the game was over.

12.Bg5 Nxg5 White resigned

[*A word about statistics. In any database devoted to a particular opening, the success of the line will be inflated, as partisans and publishers tend to show off successes, not abject failures. I have corrected for this somewhat, in that about 90% of the games in The Database are drawn from play at FICS, over a 12-year period - all the wins, losses and draws in each particular opening. Statistics in The Database, thus, largely reflect the results of "average" club players in an "average" online game environment.] 

Another RHP Jerome Gambit Tournament - Game 2

If you are playing an opening that is "off the beaten path" and your opponent takes you off of that path, is it a good thing or a bad thing? In the following game Black's opening creativity serves him much less well than following the stodgy main line of the Jerome Gambit.

jankrb (2055) - Red House (1588)
Giuoco Piano Jerome Gambit tournament, RedHotPawn, 2013

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Ke7

Of the almost 4,600 games in The Database which reach the position after White's 5th move, only 9 show Black's 5th move response, with the simple Knight recapture being the most popular alternative. See "An Odd Line In An Odd Line" and "Jerome Gambit, Paulsen Variation".

6.Nxc6+

Best play now appears to be 6.Qh5! which appeared in 3 games: guest1200 - satmonger, Internet Chess Club, 2001 (1-0, 26); stivb_99 - spymaster, GameKnot.com, 2008 (1-0, 7);  and UNPREDICTABLE - aladu, FICS, 2009 (0-1, 70).

Each time, Black missed the best defense, pointed out by Rybka, 6...Bxf2+!? 7.Kxf2 Qf8+ 8.Nf3 Nf6, although White would still have an edge.

6...bxc6

Stronger was 6...dxc6, preventing the White d-pawn from advancing 2 squares, as in perrypawnpusher - salla, blitz, FICS, 2010 (1-0, 31).

7.d4 Bb6

There is no need to repeat the offbeat 7...Bb4+, from CoachCrupp - zzzozzzo, FICS, 2008: 8.c3 Bd6 9.Bg5+ Nf6 10.e5 Bxe5 11.dxe5 Kf8 12.0-0 h6 13.exf6 gxf6 14.Be3 d5 15.Bc5+ Kg8 16.Nd2 h5 17.Nf3 Bg4 18.Qd3 Bxf3 19.Qg6 checkmate.

8.Bg5+ Nf6 9.e5 Ba6 

10.exf6+

An interesting try, according to Houdini, is 10.Qf3!? Kf7 (not 10...Bxd4 11.Bxf6+ gxf6 12.Qxf6+ Ke8 13.Qxh8+ Ke7 14.Qxh7+) 11.Nd2 h6 (a bit better than 11...Bxd4 12.exf6 Re8+ 13.Ne4 Rxe4+ 14.Qxe4 Bxb2 15.Rb1 Bxf6 16.Qf3 Qe7+ 17.Kd1 Re8 18.Bxf6 gxf6±) 12.Be3 Re8 13.0-0-0 and White has an advantage as he will recover his sacrificed piece.

10...gxf6 11.Be3 Qg8 12.Rg1 Re8 13.a4 c5 14.a5 cxd4 15.axb6 dxe3 16.Rxa6 axb6 17.fxe3 Qg5

Both Kings remain uneasy, but White has an extra piece which he can make use of as soon as he consolidates his position.

18.Qf3 Kf7 19.Kd2 Re7 20.Nc3 Rhe8 21.Nd5 Re4 22.Nf4 Qc5 23.Ra3 Qe5 24.Rd3 Qa5+ 25.c3 Qb5 26.Qh5+ Qxh5 27.Rxd7+ Black resigned

What Hole?

Ah! Another unsound attack with the Jerome Gambit by Yours Truly! At first glance, it looks powerful - but, there is a hole in it. (Or maybe more than one...) Lucky for me, my opponent missed his escape, and perished in rolling thunder.

perrypawnpusher - kenkenaitya
blitz, FICS, 2014

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

The Jerome Gambit.

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ng6

I like facing this defense, as it allows White to get rid of the annoying Black dark-squared Bishop and to get his "Jerome pawns" rolling.

7.Qd5+ Ke8 8.Qxc5 d6 9.Qe3 Nf6

10.0-0 Rf8 11.f4

Or 11.f3 as in mrjoker - PhlebasP, ICC, 2009 (1-0,35).

11...Kf7

Alternatives:

11...Qe7 was seen in perrypawnpusher - parlance, blitz, FICS, 2011 (1-0, 20),  perrypawnpusher - obturator, blitz, FICS, 2012 (1-0, 31) and perrypawnpusher - fogab, blitz, FICS, 2014 (1-0, 27);

11...Ng4 was seen in perrypawnpusher - lorecai, blitz, FICS, 2010 (1-0, 47), perrypawnpusher - jgknight, blitz, FICS, 2010 (1/2-1/2, 58) and perrypawnpusher - dkahnd, blitz, FICS, 2011 (24); and

11...Rf7 was seen in mrjoker - Mandragora, ICC, 2008 (1-0, 49) and perrypawnpusher - dkahnd, blitz, FICS, 2010 (1-0, 30).

12.f5 Ne5 13.d4 Neg4 14.Qd3 Kg8 

Black has castled-by-hand, but White already has a small edge.

15.h3 Nh6 16.Bxh6 gxh6 17.Nc3 a6 18.Qe3 Kg7

Rybka 3 gives a very complicated defense starting with 18...Ne8, and including ...Qe7 and ...Qf6 to exchange (or drive) off White's Queen.

19.Qg3+

Here, and for the next few moves, Rybka 3 prefers e4-e5. Then, when I finally play the move, it prefers something different. The reason? See the note to Black's 23nd move.

19...Kf7 20.Qh4 Rg8 21.Qxh6 Rg7

22.e5?

First White should play 22.Rae1.

22...dxe5 23.dxe5 Ng8

Instead, 23...Qd4+ 24.Rf2 Qxe5 gives Black the edge. White must always be wary of the Black Queen check and capture, if he plays the e4-e5 break.

24.e6+ Kf8 25.Qf4 b5

Here Rybka 3 suggests that Black try his luck in a Rooks and pawns endgame (where he is still worse): 25...Nf6 26.Rad1 Qe7 27.Nd5 Qc5+ 28.Ne3 Qe7 29.Ng4 Nxg4 30.hxg4 Bxe6 31.fxe6+ Kg8 32.Qc4 Rg6 33.Rfe1 b5 34.Rd7 bxc4 35.Rxe7 Rg7 36.Rd7 c3 37.bxc3 Re8 38.Rxg7+ Kxg7 39.e7 Kf6.

Now White finishes off the game in sparkling style.

26.Rad1 Qe8 27.f6 Rg6 28.f7 Qe7 29.fxg8Q+ Kxg8 30.Qf7+ Qxf7 31.exf7+ Kf8 32.Rd8+ Ke7 33.f8Q+ Ke6 
Black resigned as I played 34.Re1

(The list of top ten countries with visitors to this blog in the last 30 days is: Russia, Poland, United States, Germany, France, United Kingdom, China, Argentina, Turkey and Ukraine. Welcome - and please come back! - Rick)

Ask Houdini

A while back I bought Houdini 3, and have been enjoying its assessments, having placed it next to Rybka 3 as an analysis partner.

For fun, I set up the original Jerome Gambit sacrifice (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7) and asked Houdini the best follow-up for White. Not surprising, after 3 hours of "infinite analysis" it chose the classical 5.Nxe5+ as its top line, but only about a quarter of a pawn behind that move were 5.c3 and 5.d3.

Checking The Database, I noticed there were 1040 games with 5.c3, and White scored 44% with it. By comparison, there were 842 games with 5.d3, and White scored 31%. (There were 4165 games with 5.Nxe5+, with White scoring 54%)

I again challenged my silicon assistant, how to follow up after 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5, and after 3 hours of "infinite analysis" it chose 6.Qh5+ (heading for the "boring defense"), but only by a bit less than a half-pawn over 6.d4.

There were 2,886 games with 6.Qh5+, with White scoring 55%, versus 1,075 games with 6.d4, with White scoring 50%.

Allowing for all the cautions associated with statistics, and there are many, I think it's time for me to take a closer look at 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.c3, and maybe give a second glance at 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.d4.

Good Luck Is Better...

The old saying goes "Good luck is better than a license to steal."

As much as I would like to think that my success in this game was due to "superior skill", I have to attribute most of it to Dame Fortune.
perrypawnpusher - deceptionmaster
blitz, FICS, 2013

1.e4 e5 2.Nf3 Nc6 3.Bc4 h6 

The Semi-Italian Opening.

4.0-0 Bc5 5.Bxf7+ 

The Semi-Italian Jerome Gambit.

5...Kxf7 6.Nxe5+ Nxe5 7.d4 Ne7 

A reasonable decision by Black: since he is going to lose a piece anyhow, he simply develops another. I was surprised to see only one example in The Database, Jabbaman - HrDoctor, PlayChess.com, 2003, (1-0, 16)

8.dxc5

Also possible was 8.Qh5+ N5g6 9.Qxc5, but I was planning to use my pawns for a central clamp.

8...Rf8 9.f4 N5c6 10.Nc3 Kg8 11.e5 

After the game Rybka 3 recommended, instead, advancing the f-pawn, i.e., 11.f5 b6 12.f6 Ng6 13.fxg7 Rxf1+ 14.Qxf1 Kxg7 with the idea of exposing Black's King, leaving the defender only slightly better.

I still prefer my move.

11...b6 12.Be3 bxc5 13.Bxc5 Ba6 14.Rf2 Rf5 

Black has been developing well, but I thought that this move was a bit provocative. Set to "blunder check" at 5 minutes a move after the game, Rybka 3 had nothing to say, however.

15.g4 Rf7 16.Qe1 

Probably better was 16.Qd2.

16...Nc8 17.f5 Nb6 18.Qe4 Bb7 

19.Rd1

I was almost beginning to like my position. The problem is that my planned e5-e6 is not going to amount to much.

19...Qe8 20.e6 dxe6 21.fxe6 Rxf2 22.Kxf2 Rd8 

White's e-pawn is a weakie (as 22...Nd8 would have shown), but I still had a few ideas about how I could use it - if my opponent let me.

23.Rf1 Rd2+

I had expected this aggressive move, and appreciated the tempo it forced upon me.

24.Kg1

This works, but technically 24.Ke1 was better.

24...Qd8 

I am not sure what - or how much - my opponent overlooked here, but this is an unfortunate blitz error. Perhaps he was returning some material - a Queen for a Rook and a Bishop - and figured the remaining imbalance of 2 Knights, Bishop and Rook would outplay my Queen and Knight; but there is an oversight in that evaluation.

25.Rf8+ Qxf8 26.Bxf8 Kxf8 27.Qf4+ 

Black's Rook is hanging. Now all I had to do was avoid blundering and avoid running out of time.

27...Ke7 28.Qxd2

After the game Rybka 3 harped that I should have played 28.Qf7+ Kd6 29.Ne4+ Ke5 30.Nxd2. Whatever. 

28...Ne5 29.Qe3 Nf3+ 30.Kf2 Nxh2 31.Qc5+ Kxe6 32.Qxc7 Nxg4+ 33.Kg3 Bc8 34.Kxg4 Black resigned

Houdini 2 Too

As you may have noticed in recent posts, I have added Houdini 2 to my cast of analysis partners (including Rybka 3 and Fritz 10) for games reviewed on this site. It races through the plys in its analysis, while remaining quite strong. While I still refer to Rybka 3 for more positional questions, I have been finding a whole new perspective on the Jerome Gambit lines with Houdini 2.

One example: after a long think, its preference after 1.e4 e5 2.Nf3 Nc6 3.Bc4 Na5!? is neither the Jerome-ish 4.Bxf7+!? nor the straight-forward 4.Nxe5, but 4.Be2. Interesting...

The free version of Houdini 1.5 is available here.

Persistence

I met JoseSoza in the first round of our Chess.com tournament. I played the Jerome Gambit. He beat me.

I recently met JoseSoza in the second round of our Chess.com tournament. I played the Jerome Gambit again. He beat me again.

I am waiting to see if I will make it to the third round of our Chess.com tournament. I already know that JoseSoza will advance. If I meet him again, I will play the Jerome Gambit again.

perrypawnpusher - JoseSoza
"Italian Game" Thematic, Chess.com, 2012

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Kf8 

The same answer LeeBradbury offered in my other Jerome Gambit game in this round.

7.Qxe5 Qe7 8.Qf4+ Qf6 9.Qg3 d6 

I decided not to play the Queen exchange this time.

After 9...Bd6 I was able to outplay my opponent in  perrypawnpusher - molerat, blitz, FICS, 2010 (1-0, 23).

10.Nc3 

Instead, Bill Wall was successful with 10.c3, in Wall,B - Guest340293, Playchess.com, 2012 (1-0, 41). 

10...Ne7 11.d3 Be6 12.0-0 Qg6 13.Be3 Bxe3 14.Qxe3 a6 

JoseSoza has beaten my Jerome Gambit before, so he is not terribly concerned in this game. Still, he has allowed me to exchange off his dark-squared Bishop, keep my Queen, and prepare for f2-f4. Black's piece-for-two-pawns gives him an advantage, but the game is not unfamiliar to a Jerome Gambit player with white. 

15.f4 Bf7 16.Rae1 Qe6 17.f5 Qd7 18.e5 

The game is effectively over at this point.

That may surprise you.

What happened is that I spent hours and hours with my pocket chess set, looking at 18.f6!? I then studied the position further, and decided that I could play the text move first, as long as I followed it up properly.

So I played the text.

Then I realized that I had not written any of my analysis down, and I could not remember any of it at all! I played the rest of the game like a zombie.

For the record, after the game Rybka suggested that after 18.Rf2 Kg8 19.Ref1 Rf8 20.Qg3 Qe8 21.b3 h6 22.f6 Ng6 23.d4 Rh7 24.Qe3 White would have a tiny edge.

18...dxe5 19.Qxe5 Nc6 20.Qc5+ Qd6 21.Qxd6+ cxd6 

Exchanging Queens was not a good idea. Black now has a slight advantage, and he outplayed me the rest of the game.

22.f6 Re8 23.fxg7+ Kxg7 24.Ne4 Rd8 25.a3 Rhe8 26.Ng5 Rxe1 27.Rxe1 Rd7 28.Nxf7 Kxf7 

29.c3 Re7 30.Kf2 Rxe1 31.Kxe1 Ke6 32.Ke2 Ke5 33.Ke3 d5 34.d4+ Kf5 35.a4 Na5 36.b4 Nc4+ 37.Kf3 b6 White resigned