Recently Bill Wall sent me 16 of his Jerome Gambit games that were Human + Computer vs Computer encounters. Such teamwork is sometimes referred to as advanced chess, or cyborg chess, or centaur chess. Over the years, I have posted games from Human vs Computer matches (including the legendary 1993 Fisher-Kirshner - Knight Stalker battles, and the rolling 2006 RevvedUp - Fritz 8 - Crafty 19.19 - Hiarcs 8 - Shredder 8 - Yace Paderborn mayhem) as well as many Computer vs Computer games, but I think this is the first centaur chess I have presented. The results are interesting - even if it is difficult to assign the relative impact that the human had on the play. Also, the time controls, which affect the strength of computer programs, are not known. Over all, White scored 4 - 9 - 3 (34%), which would be unimpressive for a normal opening under normal circumstances, but which seems - as with all Jerome Gambit matches - a bit "high" for a many-times-refuted opening. A little more insight is available by breaking the games down into 4-game matches. Crafty vs Stockfish + Wall, for example, yielded 2 wins for Black when played by the team; and, likewise, 2 wins for White when played by the team. With all due respect to Dr. Robert Hyatt's computer engine, it appears it could have been simply outplayed by its stronger computer opponent. Who played what color did not seem to matter. On the other hand, the Komodo 5 vs Rybka + Wall match, which ended with a score of 2 - 2 - 0, was composed of 4 wins by Black. Neither engine, it appears, was able to ovecome the "handicap" of playing the Jerome Gambit. The Hiarcs 9 vs Critter + Wall match seemed a reflection of the comparative strengths of the computer programs, as Hiarcs 9 lost 2 games as White, and could only manage a draw as Black. Interesting, also, was the Fritz 12 vs Houdini + Wall match. The team was 1 - 0 - 1 as White, and 1 - 0 - 1 as Black, suggesting that Houdini was the brighter computer program. Looking at a couple of lines of play, 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.d4 scored 2 - 5 - 1, while 6.Qh5+ scored 2 - 4 - 2, not much of a difference. I will be sharing some of the games, taking a look at what "theoretical" enlightenment they bring.
Chessfriend Vlasta Fejfar of the Czech Republic has faced the "annoying defense" to the Jerome Gambit a number of times. In the following game, his most recent, he comes away with the whole point. Vlastous - Idalgit Internet, 2017 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
This is also called the "silicon defense" because it is the choice of many computer chess programs. Black returns a piece and takes a lot of the action out of the position. 8.fxe5 dxe5 9.Qh3+ Kf7 The text is about equal to 9...Ke7 which was seen in Fejfar,V -Goc,P, 2015(1/2 - 1/2, 70), Fejfar,V - Chvojka, correspondence, 2016(0-1, 32) and Vlastous - irinat, Chessmaniac, 2016 (0-1, 38). 10.Qh5+ Ke6 11.Qe2
A tactical slip that drops a piece. Black may have unconsciously decided that his opponent has finished moving his Queen. 14.Qc4+ Ke7 15.Nxc5 b6 16.Nd3 a5 17.Qc3 Ke6
18.Qb3+ Black resigned Perhaps a bit soon, but Black sees he will lose the b-pawn, and White's Queen will escape any danger, so the game may have lost its interest.
Here is another look at the 6...Qh4 defense in the Jerome Gambit, discussed in the previous blog post. Again, Bill Wall has the White pieces. This is pretty heavy going, and there is plenty to study in the notes, too. Or, you can just enjoy the main game, as, after a dozen moves, Bill decides enough is enough and goes after the enemy monarch. Wall, Bill - IraHaru lichess.org, 2016 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
It is time for White to start putting the pressure on. Curiously, Stockfish 8 prefers 13.Nd5 and recommends that Black sacrifice the exchange to keep an edge in the game: 13...Rg8!?14.Nxc7 Ne5 15.Nxa8 Bb7 16.f3 Qxc5+ 17.Kh1 Bxa8. Going after the King makes more sense to me. 13...Ke7 Now it's time for the Knight to step in. 14.Nd5+ Kd8 15.cxb6 axb6 16 Qf7 Qd4
Centralizing the Queen, stepping out of the possible exposed attacks by White's Queen (after the Knight moves), and protecting the f6 pawn. Alas, it leads to disaster. 17.Rad1 Qxb2 18.Rfe1 Ne5 19.Qe7 checkmate
I have been sharing games (starting with "Irrational") where Bill Wall has given various chess-playing computer programs "Jerome Gambit odds" - and won. There were some losses, however. Curiously, while I have been wordy and full of "insight" in presenting Bill's wins, I am at a loss for many words concerning the following game. As the "Talking Fritz" program might have said, "Q.E.D." Wall, Bill - Alfil engine Palm Bay, FL, 2015 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
Alonzo Wheeler Jerome developed and defended his gambit in both across-the-board and correspondence games in the late 1800s and early 1900s. He even arranged to play the Jerome Gambit against readers of the Literary Digest. In modern times, however, the Jerome is most likely to show up in internet games, often blitz; although the occasional face-to-face contest still can be found. Recently, I received an email from Vlastimil Fejfar, of the Czech Republic, who shared three of his Jerome Gambit correspondence games - a pleasant return to the days of AWJ. Fejfar,V - Pressl corr Czech Republic, 2015 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
White faces what I have called "the annoying defense", where Black calmly gives back a piece and avoids any risky misadventures, remaining up a piece for a pawn. The idea is at least as old as D'Aumiller, A.D. - A.P., Livorno, 1878(1-0, 19). It is the choice of many computer programs in games in The Database, including Fritz, Hiarcs, Junior, Rybka, Shredder and Spike; so I have also referred to it as "the silicon defense". Also, 7...d6 isthe move recommended by many authorities, including IM Gary Lane in his The Greatest Ever Chess Tricks and Traps. Vlasta proceeds calmly against it. 8.fxe5 dxe5 9.Qh3+ Kd6 10.Qd3+ Ke7 11.Qg3 Ke6 12.Qb3+ Ke7 13.Qg3 Ke6 14.Qb3+ Ke7 15.Qg3
Drawn It is not clear who came out "ahead" in this encounter, Black, who was able to split the point, or White, who was able to play a "refuted" opening and not lose. I am sympathetic. Thesecond roundof the Chess.com Italian Game Tournamenthas concluded for me, again (like in the first round) without being able to contest a single Jerome Gambit, which my opponents dodged. With White I scored two wins (one on time), four draws and no losses. Did more than half of my opponents "succeed" in "winning half a point" against me, or did they miss out on strolling to victory?
[This is blog post number 2,150, for those who might wonder. - Rick]
Immediately after blundering on move 11, I thought about resigning; but I decided to hold on for a move... then another... then another. Eventually, I held on for the win. Blitz games can be very strange, and those featuring the Jerome Gambit even more so. Some of the positions in the following game might make up for the questionable play of both me and my opponent. perrypawnpusher -Toscolano blitz, FICS, 2015 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
Surprisingly a TN according to The Database. Of note, however, is that after 8...d6 9.Qe3 a successful computer choice was 9...Qh4 in RevvedUp - Yace Paderborn, 2 12 blitz, 2006. In the same competition, 7...Kf8 8.Qxc5+ d6 9.Qe3 Qh4 was seen inRevvedUp - Shredder 8, 2 12 blitz, 2006 and Shredder 8 - RevvedUp, 2 12 blitz, 2006. 9.Nc3 d6 An indication of how weird things were going to get is that after the game Houdini recommended, instead, 9...Qg4 10.Rg1 Kd8. 10.Qe3 I probably could have taken the pawn with 10.Qxc7, but it looked complicated. After the game, Houdini agreed with me: 10...Qg4 11.Nb5 Qxe4+ 12.Kf1 Qe7 13.b3 Bf5 14.Ba3 Qxc7 15.Nxc7+ Kd7 16.Nxa8 Nf6 17.Nc7 Kxc7 18.d3 and things would be about balanced between White's extra Rook plus 2 pawns vs Black's two extra Knights.
analysis diagram 10...Nf4 11.g3 A mindless move, after which I immediately thought of resigning. Of course, White needed to be brave and castle. 11...Ng2+ 12.Ke2 Bg4+ Strong, but even stronger would be 12...Qh5+. 13.f3 Qh3
Black is still better after playing the text, but it was time for 13...Nxe3 14.gxh4 Nxc2 15.Rb1 Bh5, when Black would be up a piece for a pawn. 14.Qf2 Bh5 Black appears repelled by the tactical mess too, and continues to play "safe" and "normal" moves instead of doing concrete analysis - understandable in blitz. After the game Houdini suggested that Black wade in and keep his advantage with 14...Ne7 15.d3 Rf8 16.Bf4 Nxf4+ 17.gxf4 Rxf4 18.Qg3 Rxf3 19.Qxh3 Rxh3+ 20.Kf2 Kd7.
analysis diagram 15.d3 This is now good enough for equality, while 15.Rg1 would lead to a White advantage. 15...Nf6 This hands over the advantage - not that I was ready to take it! Black could maintain equality with 15...Ne7 16.g4 Bxg4 17.fxg4 Qxg4+ 18.Qf3 Qxf3+ 19.Kxf3 Nh4+. 16.Bg5 I was still floundering. Instead of this reasonable move White should have played 16. Rg1 with an small edge (2 pieces vs Rook) after 16...Rf8 17. Qxg2 Qxg2+ 18. Rxg2 Nxe4 19. dxe4 Bxf3+ 20. Kf1 Bxg2+ 21. Kxg2 Kd7.
analysis diagram 16...h6 Finally letting White off the hook. Instead, Black should have piled on with 16... Ng4, which wins, for example 17.fxg4 Qxg4+ 18.Kf1 (18.Kd2 Qxg5+ 19.Qf4 Nxf4; 18.Qf3 Qxf3+) 18...Rf8 19.Qxf8+ Kxf8 20.Kxg2 Qxg5. 17.Bxf6 gxf6 18.Rag1 f5 19.Qxg2 Qxg2+ 20.Rxg2 fxe4 21.dxe4 Rf822.Rf2 c6
The smoke has cleared, and White ("It is better to be lucky than good") is up a couple of pawns. 23.Rd1 Kd7 24.Rd3 Rae8 25.Kd2 An ending slip. 25...d5 26.exd5 cxd5 27.Rxd5+ Kc6 28.Rxh5 Rd8+ 29.Kc1 Rfe8 30. Rxh6+ Kc7 31.Re2 Rf8 32.Re7+ Kc8 33.Rhh7 Rxf3 34.Rc7+ Kb8 35.Rxb7+ Kc8 36.Rhc7 checkmate
Okay, say it with me, "Nobody ever won a game by resigning." Lesson learned.
9...Nf6 10.Qd8 Qxe4 Black's move allows White's Queen to escape and counter-attack. There are a number of alternative moves for Black, two of which bear attention.
Not 10...a6?, as in piratebopper - MiloBot, FICS, 2010, (1-0, 24).
Nor 10...a5? as in perrypawnpusher - MiloBot, blitz, FICS, 2010, (1-0, 24). Played, if in a stumbling manner, is 10...Bd7?. The core game continued 11.Qxc7 (instead, 11.Qxa8? Nxe4? [Black should finish Blackburne-style: 11...Ng4 12.h3 Bxf2+ 13.Kh1 Qg3 14.Rxf2+ Nxf2+ 15.Kg1 Nxh3+ 16.Kh1 Nf4 17.Qf8+ Kxf8 18.d3 Qxg2#] 12.d4 Bxd4 13.Be3 Bxb2 14.Qxb7 Bxa1 15.Qxc7 Nf6 16.Qxd6 Kg7 17.Qc5 Ne4 18.Bd4+ Bxd4 19.Qxd4+ Kh6 20.f3 Black resigned, UNPREDICTABLE - choron, FICS, 2009)11...Bb6? (instead, 11...Rc8 12.Qxb7 Ke7 is "gloriously obscure" according to Dr. Andrew Walker, University of Nottingham, personal communication, 2001. Probably 13.Qb3 with the idea of Qg3 is White's answer - not 13.Nc3? Ng4 when Black mates) 12.g3?! (12.Qxd6) 12...Qh3 13.Qxd6? Bc6 14.g4 Qxg4+ White resigned, Harris,S - Quayle,E, Los Angeles, CA, 1944.
Certainly solid, and Rybka's recommendation, is 10...Bb6. White should spring his Queen with 11.e5 dxe5 12.Qd3. The earliest reference that I have seen to this line is from The Art of the Checkmate by Renaud and Kahn (1953), which says White has the advantage. This proved true in Wall,B - Foo,N, Palm Bay, FL, 2010 (1-0, 33).
The most exciting move in the position is Chandler's 10...Bh3!?, when White has to temporarily forego the Rook at a8 and focus on checking Black: 11.Qxc7+ (11.g3? Qxe4 12.Qxc7+ Kf8 White resigned, Siggus -toe, FICS, 2007; 11.Qxa8? Qg4 12.Qe8+ Nxe8 13.g3 Qf3 14.Re1 Qg2#) 11...Kf8! (11...Kg8 lost in Chandler,G - Dimitrov, T/5 minute special game 2004; in Hiarcs 8 - RevvedUp, blitz 2 12, 2006 [1-0, 17]; and AsceticKingK9 - blackburne, ChessWorld JG6, 2011 [1-0, 22] ) and analysis by Geoff Chandler and, later, Dennis Monokroussos (see the twoposts on his blog) show that the outcome is a draw, as it is too dangerous for either player to look for more. 11.Qxc7+
This is an improvement over 11.Nc3 Qe7 (11...Qxc2 12.Qxc7+ Nd7 13.Nd5 Qe4 14.Nc3 Qe5 15.d3 Qd4 16.Re1 Qxf2+ White resigned, ionman - GriffyJr, FICS, 2005 (0-1,16); Interested readers should check out "Ionman vs The Bots" ) 12.Qxe7+ Kxe7 as in RevvedUp - Shredder 8, blitz 2 12, 2006 (0-1, 25). 11...Kg8
radicalmove played over a dozen games against LuigiBot on FICS in 2012, testing the somewhat weaker 11...Bd7 (see "Poor, Poor Computer" for a couple of examples). The games are in The Database.
Likewise, ionman took on the computer GriffyJr a couple of times at FICS in 2005, over the somewhat weaker 11...Qe7 (the games are also in The Database). White triumphed in Shredder 8 - RevvedUp, blitz 2 12, 2006 (1-0, 28). 12.Nc3
Maybe a little better was 12.Qd8+ Kf7 13.d3 Qe5 14.Bh6 Qe8 15.Qxe8+ Nxe8. 12...Qxc2
The pawn is poison. 13.Qd8+ After the game Houdini preferred 13.d4!? Bb6 (13...Bxd4 14.Bh6 Bd7 15.Qc4+ d5 16.Qxd4) 14.Qe7 when White steals the Knight in broad daylight. 13...Kg7 14.Qe7+ Kh6 15.Qxf6 Black resigned
We continue (see the earlier "Opening Report" parts 1, 2, and 3) to uses ChessBase's Opening Report function to take a look at the games in The Database. As a reminder, the Report gathers what lines are being played, and what lines are being used by the top players in the database. It gives a good "real life" view of the play of the Jerome Gambit, even if it does not always identify the objectively best move or line. Opening Report 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.Qh5+
It's time again (see the earlier "Opening Report" parts 1, 2, and 3) to use ChessBase's Opening Report function to take a look at the games in The Database. The Report gathers what lines are being played, and what lines are being used by the top players in the database. It gives a good "real life" view of the play of the Jerome Gambit, even if it does not always identify the objectively best move or line. (That would require a process called backsolving, which is not available in my ChessBase8.)
The Opening Report again also highlights a number of games from an interesting 2008 30-game human vs computers match (starting with "Jerome Gambit: Drilling Down (1)" This post starts an extended series (which may be interrupted from time-to-time for news, games or analysis) wherein the intrepid "RevvedUp" (a good chess player) and his trusted companions Hiarcs 8, Shredder 8, Yace Paderborn, Crafty 19.19 and Fritz 8 explore the Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) by taking turns playing the White and Black pieces.
The human moves first, and takes notice of the defense the computer plays. In the next game, where he moves second, the human plays that defense against a new computer – and sees how it attacks. In the third game, the human plays the recent attack against his new computer foe. Collectively, the players drill deeper and deeper into the Jerome Gambit.
