Critical
Lines
in the Halloween Gambit
By Paul Keiser
The article below has been updated by the author in October 2005.
Read the latest analysis in this pdf
file.
Part 1
Reprinted with the kind permission of the author. This article was
originally published in Issue 8 of the Unorthodox Openings Newsletter,
edited by
Davide Rozzoni.
Berlaar, January 2003
Introduction
The Halloween Gambit
was already known by Oskar Cordel in 1888 who pointed to the
formidable attacking possibilities in his « Fuerher der
Eroeffnungstheorie « but concluded that this finally was not worth the
sacrifice.
In that time the gambit
was called the Mueller-Schultze gambit, but Rainer Schlenker,
well known by all Unorthodox chess lovers, re-baptized
it as the Halloween
Gambit in his magazine « Randspringer » in December 1993. The
name is founded on the fact that players who are for the first time
confronted with the surprising and unexpected fourth move of white become so
shocked as if they were suddenly confronted with the horror of some scary
Halloween mask.
Indeed, this fourth
move is shocking. After : 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 when
Black is
expecting a normal Four Knights Game move as 4.Bb5, White surprisingly takes
the pawn on e5 (diagram).
       
 
 
    
A certain Steffen A.
Jakob, German chessplayer and computer programmer was heavily touched by
the Randspringer article and created Brause. Brause was a
clone of the chess program Crafty loaded with a Halloween gambit opening
book and played more than 3000 internet games in the
period from 1996 to 1998 in which it scored 72 %. Jakob constantly
changed and improved the opening book of Brause up
to more than 500 sub-variations.
Brause showed in a lot
of short games how deadly this gambit could be. One example : 1.e4 e5 2.Nf3 Nc6 3.Nc3
Nf6 4.Nxe5 Nxe5 5.d4 Nc6 (this is one main line, the other one is 5 ..Ng6),
6.d5 Nb8 7.e5 Ng8 8.d6 c6 9.Bc4 f6 10.Qh5+ g6 11.exf6 Qxf6 12.Qe2+ Kd8
13.Ne4 and Black resigned Brause (2355)-Betrueger
(2315),ICC,1997,1-0(13).
In these days the
theory of the Halloween gambit knows a second revival due to new analyses of
Maurits Wind and to the extended power of
chess programs as Fritz and Hiarcs.
This article is in large part based on an extended analysis
by FM Maurits Wind which is going to appear in the German magazine Kaissiber
nr 20, an article which will be very interesting for all Halloween fans.
In this article
I want to summarize the lines I have met most frequently while playing this
gambit, and how to handle so-called refutations and critical lines.
Model game
1.Nc3
The preparation part : the
Queens pawn opening is the best way trying to reach the Halloween. Via e2-e4
it’s more difficult : black has more ways to deviate (Sicilian, Caro-Kann,
etc.)
1...e5
2.e4 Nf6
3.Nf3 Nc6
4.Nxe5?!
Here the fun starts!
4...Nxe5
What does White get for
the piece? A pawn and the center. Not enough? Let's try it out.
5.d4
Now we are in the second
and most amusing part : the hunt for the Knights. The following White moves
are not hard to find.
5...Nc6
Ng6 is the other main line.
6.d5 Ne5
7.f4 Ng6
8.e5 Ng8
The third part of the game
: searching for combinations. White has the center, and his bishops are free
to run. If there was not the loss of the piece, White
would already have a superior position. The strategy is : try to hamper
Black's development while attacking, and
look for a way to get definitive advantage.
9.d6
Now after cxd6 exd6 White
is menacing Qe2 to win back the piece with the better pawnstructure.
9...a6
Black is afraid of the
manoeuvre Nc3-Nb5-Nc7.
10.h4
The queenside defended by
a6 ? No problem, the point of attack will be switched to the kingside.
There still is a knight waiting for us.
10...cxd6
11.exd6 Qf6
12.Qe2+ Kd8
13.Nd5
What follows is typical
play in this kind of position.
13...Qxd6
14.Be3
14...Qxd5 being impossible
because of 15. Bb6+ and the game is over
14...Nxf4
Black decides to give the
piece back, but he still is vulnerable : pawn d7 is isolated and the King
has to stay in the middle.
15.Nxf4
Fourth part of the game :
trying to win a won position.
15...Ne7
16.Nd3 Nd5
17.0-0-0 Nxe3
18.Qxe3 Qe6
19.Qd4 Qxa2
20.Qb6+ Ke8
21.Re1+ Be7
22.Nb4
With Nd5 in the air.
22...Qa1+
23.Kd2 Qxe1+
24.Kxe1 h5
25.Nd5 Rh6
26.Nc7+ Kf8
27.Nxa8 Rxb6
28.Nxb6
and Black resigned, as he
will lose the bishop too. keiserpaul
(1918)-Thedrum (1748),gameknot,2002,1-0
Index of
Variations
1.e4 e5 2.Nf3 Nc6 3.Nc3
Nf6 4.Nxe5 Nxe4 !?
1.e4 e5 2.Nf3 Nc6 3.Nc3
Nf6 4.Nxe5 Nxe5 5.d4
The Nc6 Section
5. ..
Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 0.exd6
Qf6 11.Nb5 Kd8 12.Be3
5. .. Nc6
6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qf6
11.Nb5 Rb8
5.
.. Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qb6
5.
.. Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qa5
5. .. Nc6 6.d5 Ne5 7.f4
Ng6 8.e5 Bb4
5. .. Nc6 6.d5 Ne5 7.f4 Bd6
5. .. Nc6 6.d5 Ne5 7.f4
Neg4
5. .. Nc6 6.d5 Ne7 7.e5
Nfxd5
5. .. Nc6 6.d5 Bb4
5. .. Nc6 6.d5 Na5
5. .. Nc6 6.d5
Nb8
The following sections are covered in part 2 of this
article:
The Ng6 Section
5. .. Ng6 6.e5 Ng8 7.Bc4 c6
5. .. Ng6 6.e5 Ng8 7.Bc4 d6
8.Qf3 Qd7 9.0-0
5. .. Ng6 6.e5 Ng8 7.Bc4
d6 8.Qf3 others
5. .. Ng6 6.e5 Ng8 7.Bc4 d5
5. .. Ng6 6.e5 Ng8 7.Bc4 Bb4
8.Qf3 Bxc3+
5. .. Ng6 6.e5 Ng8 7.Bc4
Bb4 8.Qf3 others
5. .. Ng6 6.e5 Qe7
5. .. Ng6 6.e5 Bb4
Bb4 and other fifth moves
5. .. Bb4 6.dxe5
5. .. Nxe4
5. .. Bd6
5. .. Neg4
5. .. Qe7
Similar ideas
The Chicago gambit
The Halloween gambit
reversed
The
Gambit
Declined
1.e4 e5 2.Nf3 Nc6 3.Nc3
Nf6 4.Nxe5 Nxe4 !?
Black is not obliged to
accept the gambit, but this is very rarely. Why should he ? After all, a
knight is a knight !
4.Nxe5 Nxe4 5.Nxc6 Nxc3 6.dxc3 bxc6 7.Bd3
Be7 leads to equality Brause(2590)-scorpiocy 2410),ICS,1997,1-0(29)
The
Gambit
Accepted
1.e4 e5 2.Nf3 Nc6 3.Nc3
Nf6 4.Nxe5 Nxe5 5.d4
Nc6 Section
Main Line
The key position of the
Nc6 section occurs after following main line:
1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.Nxe5 Nxe5 5.d4
Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qf6 11.Nb5 Kd8 12.Be3
(diagram)
White has created a
strong outpost on d6 . Black has tried several moves here, but most of them
give no satisfaction.
Of course 12. .. a6 is
not possible because of 13.Bb6+ Ke8 14.Nc7+ Kd8 15.Nxa8+ Ke8 16.Nc7+ Kd8
17.Nxa6+ Ke8 18.Nc7+ Kd8 19.Nd5+. Very nice !
12. .. Nh6 !?
This is a very
strong reply and hard to find an answer to it.
13.Qd2 b6 and now
14.0-0-0 has proved to be bad :
14. .. Ng4 15.Bg1 Qxf4
16.Qxf4 Nxf4 17.Nxa7 Ne6 18.Be2 Ne5 19.Bxb6+ Ke8
keiserpaul(2006)-1962(1878),gameknot,2002,0-1(52)
14. .. Nf5 15.Kb1 Nxe3
keiserpaul (2014)-chessnake
(2033), gameknot,
2002, 0-1 (31)
But what else could
have been played ? Moving the Queen a second time 14.Qd5!?(idea : 14. ..
Rb8 15.0-0-0 Nxf4 16.Qc4) failed on 14. .. Qe6 and White is obliged to
exchange Queens. keiserpaul (2006)-lexherman
(2197), gameknot, 2003,
0-1(38) went
15.Qxe6 dxe6 16.Nxa7 bxd6 17.Bxb6 with two pawns for the piece, but this was
not enough compensation.
Another interesting idea
is the suggestion of M.Wind 14.f5!? forcing the queen
capture 14. .. Qxf5 15.Nxa7 Qe6 16.0-0-0 . A game
Keiserpaul(2014)-wakigawa(2137), gameknot, 2003, 0-1 (43) went 15. ..
Rb8 16.Nxc8 Rxc8 17.Bxb6+ Ke8 18.Bc7 Be7! 19.c3 Bg5 20.Qd3. Also in this
game white has not enough compensation with 2 good pawns for the piece.
In this line the move
15.Bd3! leaded to a drawn by threefold repetition in Lexherman(2276)-keiserpaul(1947),gameknot,2003,(43)
after 15. .. Qh5 16.Be2 Qe5 17.Bd4 Qf4 18.Be3 Qe5 (18. .. Qh4+ 19.g3 Qa4
does not promise much).
12. .. Qxb2 ?!
The logical idea of
making room at f6 and taking a pawn at the same moment fails on 13.Bd4 Qb4+
14.c3 Qa5 (14. .. Qb2? 15.g3 traps the Queen) 15.f5 Ne5 (15. .. a6 16.Nc7
Rb8 17.fxg6 hxg6 18.Nd5 += analyse Wind) 16.Bxe5 a6 and now not
17.Qd5 17...Nh6 18.Qd4 axb5 19.Bf6+ gxf6 20.Qxf6+ Ke8 21.Qe5+ Kd8 22.Qxh8
Ng8! as in keiserpaul(2004)-lanesra(1936),gameknot,2002,0-1(54) but
Antonio Torrecillas winning suggestion of 17.Qh5! axb5 18.Qxf7 Nh6
19.Bf6+ gxf6 20.Qxf6+ Ke8 21.Be2
12. .. b6 ?!
The idea of
fianchetting the bishop gives White the opportunity to execute his
aggressive plan by taking the a7 pawn 13.Qd5 Rb8 14.0-0-0 Bb7 15.Qd2 (
Wind opts for 15.Qc4 Rc8 16.Qa4) Qe6 16.Nxa7 Ba8 17.Qc3 Bxd6 18.Bc4 Qe7
19.Rhe1 27.Nb5 keiserpaul (2022)-beardsleyx
(1906), gameknot,
2002,1-0(27)
12. .. Nxf4 ?!
Doesn’t help Black's
development either 13.Qd2 Ne6 14.0-0-0 Qg6 (14. .. b6 15.Bc4 Wind)
15.Bd3 Qg4 16.Qa5+ Ke8 17.Bf5 Qxg2 18.Bxe6 fxe6 19.Nc7+ Kf7 20.Qh5+ Qg6
21.Qe5 Nf6 22.Rdf1 b6 23.Rhg1 Rb8 24.Rxg6 Harms,F-Moeckel,J,TGT
ICCF,1998,1-0(24)
12. .. Qe6 ?
13.Qd4 Nf6 14.0-0-0 Qe4
15.Qc3 Qc6 16.Qb4 Ne4 17.f5 with advantage for White (analysis M.Wind)
Deviations on the Main Line
Deviation A)
11. .. Rb8
The Plasma Variation
1.e4 e5 2.Nf3 Nc6 3.Nc3
Nf6 4.Nxe5 Nxe5 5.d4 Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qf6
11.Nb5 Rb8
Brause inconsistently
plays 12.Be3 here but Schiller’s Qe2 looks better.
12.Qe2+ Kd8 13.Be3 b6
14.0-0-0 Qe6 p.e. 15.Nc7 Qxa2 16.Bxb6 Nf6 17.Ne6+ Ke8 18.Nxg7#
Deviation B) 10. .. Qb6
1.e4 e5 2.Nc3 Nf6 3.Nf3
Nc6 4.Nxe5 Nxe5 5.d4 Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qb6
B1) 11. Qe2+ ?!
11.
.. Kd8 12.f5
(diagram)
Black has played :
12. .. Bxd6 !
Following the Brause DB
where White scores 100% , White has not to be afraid for a Black attack and
can simply continue his development. But this is not true if we take a
closer look. Brause had switched from 14.Nb5 to 14.h3, but both moves are
inadequate.
13.fxg6 hxg6 14.Nb5
Bg3+ 15.Kd1 a6 (15. .. Rxh2! 16.Rxh2 Bxh2 17.Be3 Qa6 18.Qf3 qf6 and Black is
a pawn up) 16.Qf3 Nf6 17.Be3 Qe6 18.Qxg3 axb5 19.Bxb5 Rxa2 20.Rxa2 Qxa2
21.Bb6+ Ke8 22.Qc7 Qa8 23.Qd8
Brause(2375)-sadatsugu(1895),ICC,1997,1-0(23)
13.fxg6 hxg6 14.h3 Nf6
15.Qf3 Re8+ ( 15. .. Be5! 16.Be2 d6 and Black was better in paulkeiser
(2014)-harryman (1974),
gameknot,0-1(45)),
16.Be2 Bc5 17.Rf1 Be3 18.Kd1 Qd6+ 19.Bd3 Bxc1 20.Kxc1
Brause
(2505)-SirJames (2060),
ICC,1998,1-0(43)
12. .. Nh4 ?!
Black cannot save the
knight 13.Qg4 Nxg2+ 14.Bxg2 Nf6 15.Qg3 Bxd6 16.Qxg7 Re8+ 17.Ne2 Nh5 18.Qg5+
f6 19.Qxh5 Brause(2555)-OLTS (1995),
ICC,1998,1-0(19)
12. .. Nf6 ?!
This development move
cannot stop white’s aggression
13.fxg6 hxg6 14.Qc4
Brause(2700)-Drk(2355),ICC,1998,1-0(21)
B2) 11. Nb5
Results in a balanced
position according to M. Wind. He gives following analysis : 11.
..Kd8 12.f5 Ne5 13.Bf4 f6 (13. .. Nc6 !?) 14.Qd2 Qc6 (the point of white’s
play is that 14. .. a6? fails to 15.Be3 +-) 15.0-0-0 A) 15. .. a6 16.Na3 b5
17.Bxe5 fxe5 18.Nc4 bxc4 19.Qa5+ Ke8 20.Qxe5+ Kd8 = B) 15. .. b6 16.Nc7 Bb7
18.Nxa8 Bxa8 18.h4 idea Rh3
Deviation C)
10. .. Qa5
1.e4 e5 2.Nc3 Nf6 3.Nf3
Nc6 4.Nxe5 Nxe5 5.d4 Nc6 6.d5 Ne5 7.f4 Ng6 8.e5 Ng8 9.d6 cxd6 10.exd6 Qa5
This move is not very
popular, so I can only give the comments and analysis of M. Wind :
Following Wind
this prevents Nb5 and brings Qb4 in the position to capture pawn d6. White
has two replies, one solid and one very sharp. He gives 11.Qd4! (prevents
Qb4 and invites an exchange of queens) Qb6 12.Qxb6 axb6 13.Nb5 Ra5 14.Be3
(14.f5!? Ne5 15.Be3) 14...Rxb5 15.Bxb5 Bxd6 16.Bd4 f6 17.g3 Bc5 18.Bc3
roughly equal 11.Bc4!? Nh6 [11. .. Qb4 12.Qe2+ Kd8 13.Bxf7 ] 12.h4! Nf5
13.h5 Ngh4 14.Qe2+ Kd8 15.Bd2 Bxd6 (15...Nd4 16.Qd3 Qxh5) 16.0-0-0 nice
compensation, but very unclear.
Deviation D)
6. .. Ne5 7.f4 Ng6 8.e5 Bb4 !
9.exf6 Bxc3+ 10.bxc3
Qxf6 Black has attacking possibilities and the better pawn structure
Brause-crafty,ICS,1997,0-1(127)
Deviation E)
6. .. Ne5 7.f4 Bd6 !?
8.fxe5 Bxe5 now that
the advance d5 is played and the f4 pawn has
disappeared is the bishop well placed on e5 evbad
(2390)-agrino (2420), ICC,
1997,0-1(65)
Deviation
F) 6. .. Ne5 7.f4 Neg4 !?
This is a suggestion of
Bradley Zang : “If you are afraid of the Halloween Gambit just give back the
knight. Black will have the better endgame.” 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6
4.Nxe5 Nxe5 5.d4 Nc6 6.d5 Ne5 7.f4 Neg4 8.e5 Bc5 9.exf6 Nxf6 10.Qe2+ Qe7
11.Qxe7+ Kxe7
Deviation
G) 6. .. Ne7 7.e5 Nfxd5
8.Nxd5 c6 followed by
Nf6. Black has obtained equality.
Deviation H)
6. .. Bb4 !?
White can take the
Knight and if black exchange Bishop for Knight , White has the better game
due to his bishop pair. 7.dxc6 Bxc3+ 8.bxc3 dxc6 9.Qxd8+ Kxd8
Brause(2570)-Axl(2755),ICC,1997,1-0(30)
But Black can do better
with 7. .. Nxe4 and now 8.Qd4 Qe7 [ Brause plays 9.Be3?! but was never
confronted with 9. .. f5! Which lead to a dubious position according to
M. Wind ] 9.Qxg7 Nxc3+ 10.Be3 Nd5+ 11.c3 Rf8 12.cxb4 Nxe3 13.fxe3 Qxb4+
14.Kf2 dxc6 15.Be2 roughly equal.
Deviation I)
6. .. Na5 !?
Brause had a score of
only 42 % after this move but could have played better.
Dorobanov(2425)-Tremere(2740),ICC,0-1(25) went
7.e5 Qe7 8.Qe2 Ng8 9.Nb5 Kd8 10.d6 cxd6 11.exd6 with an
early Queen exchange, but with 10.Nxc7!? Rb8 11.d6 Qh4 12.g3 White could
have keep on the pressure.
Deviation J)
6. .. Nb8 !?
7.e5 Qe7 [An analysis
of M. Wind goes 7...Ng8 8.d6 cxd6 9.exd6 with sufficient compensation
after 9. .. Qa5! (9...Qf6 10.Nb5 Na6 11.Bc4; 9...Qb6 10.Nb5 Na6 11.Bc4)
10.Bf4 Nc6 (10...Qb4? 11.Qd2; 10...Nf6 11.Bc4) 11.Bc4 (11.Qe2+ !?) Qf5
12.Bg3 Nd8 13.0-0 Ne6 14.Nb5 Kd8 15.Re1] 8.Qe2 Ng8 9.d6 Qe6
8.Qe2
Ng8 9.d6 Qe6 10.Nb5 Na6 and Black has a good position.
Also Read: Part 2
Covering the Ng6 Section,
Bb4 and other fifth
moves,
The
Chicago Gambit,
and
The Halloween
Gambit
Reversed,
PLUS game downloads.
|
