While it is nice to surprise an opponent with the Jerome Gambit, more and more often defenders are showing that they have an idea about how to defend - if not outright refute - the opening. What's an attacker to do?
Well, play on, of course!
Wall,B - Guest1839713
PlayChess.com, 2013
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6 8.fxe5 dxe5
The so-called "annoying defense" (see "Philidor1792 vs the Annoying Defense" Parts 1, 2 and 3; "Further Exploration" and "Theory from the Thematic Tournament" Parts 3 and 4), the refutation choice of most computers, which has even given Bill trouble from time to time.
It is interesting to put this in perspective, however, in light of the previous post on "refutation".
The Database has over 10,000 games starting with the Jerome Gambit sequence 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+, with White scoring 45%. Interestingly enough, against the "annoying defense" White has scored 58%.
Bill has scored 70% against the "annoying defense" which looks good until you realize that his over-all Jerome Gambit score is 96%.
It is as if a sage has warned players: After the refutation, the gods have placed the rest of the game...
9.Qh3+ Kf7 10.Qh5+ Kf8 11.Qxe5
11...Bd6
This certainly looks aggressive enough, but Bill recommends, instead,11...Qh4+ 12.g3 Qe7.
12.0-0+
Wise. As Bill points out, there is more danger in 12.Qd5? Qh4+ 13.Ke2 Qg4+ 14.Ke1 Qxg2 15.Rf1+ Nf6.
12...Nf6 13.Qg5 Kg8
Instead, 13...h6 14.Qh4 g5 was seen in Colossus - Spike1.2, Jerome Forced Computer Chess Match, 2006 (0-1, 43); while 13...Be6 was seen in several games in the legendary Fisher-Kirshner,M - Knight Stalker, Mission San Jose, Fremont, CA, 1993 match.
14.e5
White needs to push on, as Bill demonstrates. Not 14.Nc3? Bxh2+ 15.Kxh2 Ng4+
14...Bc5+ 15.Kh1 Be7 16.exf6 Bxf6
What can I say?
Bill Watterson said it best when he titled one of his Calvin and Hobbes books:
To wit, Black has "scientifically" returned the sacrificed material to calm the storm. True, he is a pawn down, but he hopes that the "two Bishops" will compensate in an open position.
17.Qc5 h6 18.Nc3 c6 19.Ne4 Bd4 20.Qc4+ Kh7
Black has castled by hand.
21.c3 Be5?
Instead, Bill recommends 21...Re8 22.cxd4 Rxe4 with about an equal game, although 22.Ng3!? might be a way for White to maintain an edge.
22.d4 Bc7 23.Qd3
This is a strong move, but Bill points out 23.Nf6+! gxf6 (23...Kg6 24.Qd3+ Kf7 25.Ne8+ kills) 24.Qf7#
23...Rf8 24.Nf6+
Better late than never: a nice discovered double check.
24...Kh8 25.Qh7 checkmate
Clever!
Here is another Jerome Gambit exploration by Philidor1792, tackling an annoying defensive variation (for a game, and a discussion by MrJoker, see "Theory From the Thematic Tourney" Part 3 and Part 4), battling with his opponent and prevailing.
Philidor1792 - NN
Casual game, 2012
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
8.fxe5 dxe5 9.Qh3+ Ke7
9...Kd6 was seen in Yace Paderborn - RevvedUp, 2006 (1-0, 48) and Morin,L-Dumont,F, Montreal, 2011 (0-1, 57).
10.Qg3 Kf7 11.Qxe5 Qh4+
Other choices by Black (all in The Database):
11...Qd6 in Wall,B, - Qi,H, Chess.com, 2011 (1-0, 14);
11...b6 in Wall,B - Redom,T, Chess.com, 2010 (½-½,59);
11...Bd4 in Fritz 5.32 - Junior 7, The Jeroen Experience, 2003 (0-1, 32)
12.g3 Qe7 13.Qxe7+
Or 13.Rf1+ in Colossus - Rybka v1.0 Beta.w32, USA, 2006 (0-1,42).
13...Nxe7 14.c3 Bb6 15.d4 Ke8
Here we have a typical Jerome Gambit Queenless middlegame/endgame, where Philidor1792 wants to make the most of his "Jerome pawns".
16.Be3 Nc6 17.Nd2 Ne5 18.Kd1 Ng4 19.Ke2 Be6 20.Bf4 Kd7 21.Raf1 Raf8 22.h3 Nf6 23.Kd3 Kc8 24.Re1 Nh5 25.Be5 Rf7 26.Rhf1 Nf6
White patiently maneuvers and improves the placement of his pieces.
27.g4 Nd7 28.Bg3 Rxf1 29.Rxf1 a6 30.Nc4 Ba7 31.b3 h5 32.Ne3 hxg4 33.hxg4 Bb6 34.Rf3 Nf6 35.g5 Nh5 36.Be5 Kd7
37.Nf5 Rf8 38.c4 Bxf5 39.exf5 c6 40.Ke4 Rf7
A slip, but the game remains drawish..
41.Rh3 c5 42.Rxh5 cxd4 43.Bxd4 Bxd4 44.Kxd4 Rxf5
45.Ke4 Ke6 46.b4 Re5+ 47.Kd4 b6 48.Rh8 Rxg5 49.Rb8 Rg4+ 50.Kc3 Rg3+ 51.Kd4 Rg4+ 52.Kc3 Rg3+
Has White reached his goal of an endgame draw with the Jerome Gambit?
53.Kb2
No!
White wants more.
This seems to upset Black, who soon settles for less.
53...b5 54.Rb6+ Ke5 55.cxb5 axb5 56.Rxb5+ Kd4 57.Rc5 g5 58.a4 g4 59.a5 Rg2+ 60.Kb3 Rg3+ 61.Ka4 Rg1 62.Kb5 g3 63.a6 Ra1 64.Rg5 Ra3 65.Kb6 Black resigned
In the last game in this "human vs computers" Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) match, Yace Paderborn seems headed for the draw that Crafty 19.19 and Hiarcs 8 had found before it, when RevvedUp decides to vary – after all, his position is better.
Unfortunately, that gives the silicon beast the opportunity to lengthen the game and eventually outplay its human opponent.
Yace Paderborn - RevvedUp
blitz 2 12, 2006
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6
7.f4 d6 8.fxe5 dxe5 9.Qh3+
Kd6 10.Qd3+
10...Ke7 11.Qg3 Kf7
The draw comes with 11...Kd6, when Yace Paderborn likely would repeat with 12.Qd3+, etc. Perhaps RevvedUp was aware of a couple of previous examples of this line of play – where Black won.
12.Qxe5 Bd4
Colossus - Rybka v1.0 Beta.w32, Jerome Forced Computer Chess Match USA, 2006, transposed to this line, but saw ...Qh4+ here instead: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6 8.fxe5 dxe5 9.Qh3+ Ke7 10.Qg3 Kf7 11.Qxe5 Qh4+ 12.g3 Qe7 13.Rf1+ Kg6 14.Qxe7 Nxe7 15.c3 Bh3 16.Rf4 Bd6 17.Rh4 Bd7 18.d4 Rae8 19.e5 Nd5 20.a3 Be7 21.Re4 Bf5 22.Re2 Bd3 23.Rg2 Rhf8 24.Bf4 c5 25.Nd2 Kh5 26.Rc1 Nxf4 27.gxf4 Rxf4 28.Kd1 Rg4 29.Rxg4 Kxg4 30.b4 Rf8 31.bxc5 Rf2 32.h3+ Kxh3 33.Rb1 Bxb1 34.Nxb1 Rf1+ 35.Kc2 Rxb1 36.Kxb1 h5 37.d5 Bxc5 38.d6 Kg4 39.Kc2 h4 40.Kd3 h3 41.Kc4 b6 42.d7 Be7 White resigns
13.Rf1+ Nf6 14.Qh5+ Ke7
Junior 7 preferred ...Kg8: Fritz 5.32 - Junior 7, Jerome Gambit The Jeroen Experience (1.1), 2003: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6 8.fxe5 dxe5 9.Qh3+ Ke7 10.Qg3 Kf7 11.Qxe5 Bd4 12.Rf1+ Nf6 13.Qh5+ Kg8 14.Qg5 h6 15.Qf4 Qe7 16.d3 Bg4 17.Nd2 Rf8 18.h3 Bh5 19.c3 g5 20.Qg3 Be5 21.Qe3 Bf4 22.Rxf4 gxf4 23.Qxa7 Qg7 24.Qf2 Nd7 25.d4 Kh7 26.e5 Rhg8 27.Kf1 Qg6 28.Kg1 f3 29.g3 Qxg3+ 30.Qxg3 Rxg3+ 31.Kf2 Rg2+ 32.Kf1 Bg6 White resigned
15.Qg5 Rg8
"Objectively" stronger (i.e. a computer thought of it) was 15...Re8 16.Qxg7+ Kd6, but it is not hard to see why RevvedUp preferred the text, even at the cost of an even game.
16.e5 h6 17.Qf4 Re8 18.exf6+ Kf7+
Black's development and relative King safety easily compensates for the lost pawn. But, give White a few more moves to get his position together, and...
19.Kd1 Bxf6 20.Nc3 Kg8
Black has castled-by-hand and retains an edge.
21.Ne4 Rf8
22.Rf3 Bd7 23.Nc5 Bc6 24.Ne6 Bxf3+ 25.gxf3 Qd6 26.Qxd6 cxd6 27.Nxf8 Rxf8
28.f4 g6 29.d3 h5 30.a4 a6 31.Ra3 Rf7 32.c4 Rd7 33.b3 d5 34.Kc2 Kf7
Again, the dynamic aspects of Black's position compensate for the pawn down. Yace Paderborn, however, nibbles away, move-by move.
35.Be3 dxc4
This gives Yace Paderborn a passer, and may not be best.
36.bxc4 Ke6 37.Rb3 Bd4
This exchange opens the floodgates, and White's game improves rapidly. Perhaps something neutral like 37...Bg7 was better.
38.Bxd4 Rxd4 39.Rb6+ Kf5 40.Rxb7 Rd6 41.Rf7+ Kg4 42.Kc3 Kh3 43.f5 gxf5 44.Rxf5 Rh6 45.Rf2 h4 46.d4 Kg4 47.c5 h3 48.d5
Black resigns
