Halloween Gambit: Early History

For more than a century, the line known today as the Halloween Gambit was an entertaining option in casual play, but received minimal attention. Wikipedia notes that it appeared in a brief note in an 1888 publication as Gambit Müller und Schulze (not the names of specific people, but rather anyone or everyone). The article relies for historical details upon a article by Tim Krabbe for Chess Cafe (March 2000) and now hosted elsewhere. Krabbe's article is entertaining and the central facts are verifiable. He was prompted to write it after losing to a Crafty clone on ICC named Brause. Brause was the work of Steffen Jacob who coined the name The Halloween-Attack in the Four Knight's Game.

Although Krabbe goes back as far as Oskar Cordel, Führer durch die Schachtheorie: Ausführliche Tabelle der Spiel-Eröffnungen auf Grundlage neuester Forschung (1888), there were at least two games published in magazines prior to Cordel's text. Cordel notes that the opening had been played in Leipzig and a game played there in 1874 is the oldest in ChessBase Mega 2024.

The annotation in Deutsche Schachzeitung (July 1874) of the game between an unnamed player who played the gambit against Johannes Minckwitz, editor of the publication, offers what remains today an apt assessment of the gambit: it is unsound, but in practical play White gets a dangerous attack.

I offer the original annotations, which were in German, translations, and some of my own comments.

NN -- Minckwitz,Johannes [C47]
Leipzig, 02.05.1874

1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.Nxe5

Dieses Opfer ist zwar incorrect, gewährt jedoch einen im practischen Spiele sehr gefährlichen Angriff. This sacrifice is indeed incorrect, but grants a very dangerous attack in the practical game.

4...Nxe5 5.d4

Black to move

5...Ng6

5...Nc6 is the principal alternative.

6.e5 Ng8 7.Bc4 Bb4 8.0–0 Bxc3 9.Qf3 Qe7 10.bxc3 Nh6 11.a4 Nh4 12.Qe4

Black to move

12... d5

Schwarz ist hierzu genöthigt, da er sonst auf keine Weise zur Entwickelung gelangen kann.
Black is compelled to do this, because otherwise he cannot achieve development in any way.

13.Bxd5 c6 14.Bxc6+!

Diese Fortsetzung fällt zu Gunsten von Schwarz aus. Die Combinationen sind sehr interessant.

This sequel is in Black's favor. The combinations are very interesting.

14...bxc6 15.Qxc6+ Qd7 16.Qxa8 0–0

White to move

17.Bxh6

This move gives Black opportunities. Better would be 17.Ba3 or 17.Rb1.

17...Bb7 18.Qxa7 Ra8 19.Qb6

19.Qc5 Qg4

19...Ra6?

White to move

 20.Qb5??

20.e6!! Qd5 21.exf7+ Kf8 22.Qb4+ Kxf7 23.f3 and White should prevail. The move played walks into a mate in four, which Minckwitz finds.

20...Nf3+ 0-1 

21.gxf3 Rg6+
21.Kh1 Rxh6 22.h3 Rxh3+

Ten years later, the second game in the database was played and published. It was part of a blindfold exhibition in St. Petersburg. Ernst Falkbeer provided annotations.

The White player is not well-known. Sarah Cohen (aka batgirl) described him as a strong amateur in "The Prince of Mingrelia, Part 1" on chess.com. The game was published in a Vienna newspaper: Wiener Hausfrauen-Zeitung (1885), which I was able to find online.

De Marcovan,M - NN [C47]
Blindfold Exhibition M de Marcovan St Petersburg, 1884
[Falkbeer,Ernst Karl]

1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.Nxe5

Ein ganz neues und geniales Opfer des Blindspielers, um die Partie interessanter zu machen.

A completely new and ingenious sacrifice of the blind player to make the game more interesting.

4...Nxe5 5.d4 Ng6 6.e5 Ng8 7.Bd3 d6 8.f4 dxe5 9.fxe5

Black to move

9...Bb4

Augenscheinlich darf die Dame den Bauer auf d4 nicht schlagen, weil sie sonst durch 9...Qxd4 10.Bb5+ verloren ginge.

Apparently, the queen must not capture the pawn on d4, otherwise she will pass through (discovered attack) 9...Qxd4 10.Bb5+ was lost.

10.0–0! Qxd4+ 11.Kh1 Bxc3 12.bxc3 

Black to move

12...Qxc3

Schwarz hätte den Bauer auf c3 nicht schlagen, sondern die Dame nach 12...Qd8 zurückziehen sollen.

Black should not have captured the pawn on c3, but should have withdraw the queen with 12...Qd8.

13.Qf3

Ausgezeichnet gespielt. Weiß droht jetzt zu gleicher Zeit das direkte Schach auf f7 und das Abzugsschach, respektive den Gewinn der Dame durch 14.Lb5+.

Excellently played. White is now threatening direct check on f7, or winning the queen by 14.Bb5+.

13...Qxe5

Es war schon gar nichts mehr zu machen.

There was nothing more to be done.

Wenn die Dame den Turm nahm, so folgte 13...Qxa1 14.Qxf7+ Kd8 und dann 15.Bg5+

If the queen took the rook, then 13...Qxa1 14.Qxf7+ Kd8 and then Bg5+.

14.Qxf7+ Kd8 15.Bxg6 hxg6

White to move

16.Bg5+

Ein schönes Endspiel dieser reizenden kleinen Partie.

A beautiful endgame of this charming little game.

16...Ne7 17.Rad1+ Bd7

Weiß setzt in zwei Zügen matt:

White mates in two moves:

18.Qf8+ Rxf8 19.Rxf8# 1-0

Falkbeer Counter-gambit

Ernst Karl Falkbeer (1819-1885) founded Austria's first chess magazine, Wiener Schachzeitung, but is remembered today mostly for the response to the King's Gambit that bears his name. He published an article analyzing Black's third move and also played several games employing the gambit.

The main line appears to be 1.e4 e5 2.f4 d5 (the signature move of the Falkbeer Counter-gambit) 3.exd5 e4 (the topic of Falkbeer's 1850 article) 4.d3 Nf6 and then White has four main options.

a) Nc3
b) Qe2
c) Nd2
d) dxe4

According to John Shaw, The King's Gambit (2013), "White has excellent chances of an edge in the traditional main lines" (560). The Falkbeer "has become something of a museum piece at the highest levels," according to Neal McDonald, The King's Gambit: A Modern View of a Swashbuckling Opening (1998). Even so, Dmitrij Jakovenko has played it as recently as the 2014 Russian Championship.

My Round Four Opponent

When I was facing the King's gambit yesterday, I had recollections of one of the Polgar sisters playing the Falkbeer, and also remembered two games that I have studied in some depth, Schulten -- Morphy 1857 and Rosanes -- Anderssen 1862. Both historic games were won quickly by Black. A check of my database this morning shows that Susan Polgar lost to Boris Spassky in 1988 when she played the Falkbeer. She opted for the 3...c6 line that was popular for awhile.

The oldest game in the ChessBase database with Black's 3...e4

Anderssen,Adolf -- Falkbeer,Ernst Karl [C32]
Berlin m3 Berlin, 1851

1.e4 e5 2.f4 d5 3.exd5 e4 4.Bb5+ Bd7 5.Qe2 Nf6 6.Nc3 Bc5 7.Nxe4 0–0 8.Bxd7 Nbxd7 9.d3 Nxd5 10.Nf3 Re8 11.f5 Bb4+ 12.Kf2 N7f6 13.g3 Qd7 14.c4 Nxe4+ 15.dxe4 Nf6 16.e5 Qxf5 17.Kg2 Rad8 18.a3 Bd6 19.Rd1 Qh5 20.c5 Rxe5 21.Qxe5 Qg4 22.cxd6 Re8 23.Qxe8+ Nxe8 24.d7 Qe4 25.d8Q Qc2+ 26.Bd2 1–0

Two memorable historic games.

Schulten,John William -- Morphy,Paul [C32]
New York blindfold m New York, 1857

1.e4 e5 2.f4 d5 3.exd5 e4 4.Nc3 Nf6 5.d3 Bb4 6.Bd2 e3 7.Bxe3 0–0 8.Bd2 Bxc3 9.bxc3 Re8+ 10.Be2 Bg4 11.c4 c6 12.dxc6 Nxc6 13.Kf1 Rxe2 14.Nxe2 Nd4 15.Qb1 Bxe2+ 16.Kf2 Ng4+ 17.Kg1 Nf3+ 18.gxf3 Qd4+ 19.Kg2 Qf2+ 20.Kh3 Qxf3+ 0–1

Rosanes,Jacob -- Anderssen,Adolf [C32]
Breslau m Breslau, 1862

1.e4 e5 2.f4 d5 3.exd5 e4 4.Bb5+ c6 5.dxc6 Nxc6 6.Nc3 Nf6 7.Qe2 Bc5 8.Nxe4 0–0 9.Bxc6 bxc6 10.d3 Re8 11.Bd2 Nxe4 12.dxe4 Bf5 13.e5 Qb6 14.0–0–0 Bd4 15.c3 Rab8 16.b3 Red8 17.Nf3 Qxb3 18.axb3 Rxb3 19.Be1 Be3+ 0–1

My fourth round opponent started with the Bird, which I met with the From, then we transposed into the King's Gambit and the Falkbeer. That is exactly how one of my worst tournament games ever began two years ago, except that I had White (see "Knowing Better").

Tito Tinajero (1614) -- James Stripes (1845) [C32]
25th Collyer Memorial Spokane Valley (4), 26.02.2017

1.f4 e5 2.e4 d5 3.exd5 e4 4.Nc3

4.d3 is considered best. 4...Nf6 5.dxe4 Nxe4 6.Nf3 Bc5 7.Qe2 "theory and practice have demonstrated with a high degree of certainty that White will obtain an advantage" (Shaw, 585).

4...Nf6 5.Bb5+

I expected 5.d3, which is the main line. Shaw mentions 5.Bb5 as a means of avoiding established theory. It's certainly a worthy move at the club level.

5...c6

5...Nbd7 or 5...Bd7 were options. Because I needed to win, I was happier accepting weaknesses in pawn structure than offering minor piece exchanges. Later, however, concrete analysis forced me to consider some exchanges.

6.dxc6 bxc6 7.Bc4

Black to move

7...Bg4?!

7...Bc5 seems better and was played in the only game in PowerBook 2016 with 5.Bb5+. That game continued 8.d4 Qxd4 9.Qxd4 Bxd4 10.Nge2 Bb6 11.Na4 Ba6 12.Nxb6 axb6 13.Bb3 0–0 14.h3 c5 15.a3 e3 16.Bxe3 Re8 17.Kf2 Ne4+ 18.Kf3 Bb7 19.Rhe1 Nd7 20.Ng3 Ndf6 21.Nxe4 Nxe4 22.Ke2 Ba6+ 23.Kf3 Bb7 24.Ke2 Ba6+ 25.Kf3 Bb7 26.Ke2 ½–½ Metz,H (2275) -- Baburin,A (2530) Liechtenstein 1993

8.Be2

8.Nge2 was worth considering.

8...Be6 

8...Bf5 9.d3 Qb6 10.dxe4 Bxe4 was played in Richter,E -- McAloon,J, Ca'n Picafort 1992, which White won in 36 moves.

9.d3 Bb4 10.Bd2

Black to move

10...exd3

I first thought of 10...e3, as Morphy played against Schulten, but my c-pawn makes a critical element of Morphy's attack impossible.

I missed an opportunity: 10...Qd4! 11.Qc1 (11.dxe4? Bc5 12.Qc1 Qf2+-+) 11...Nbd7 and Black's pieces are more active.

11.Bxd3 Qb6?

I should have castled, but failed to anticipate White's next move.

12.f5 Bd5

I spent twenty minutes on this move.

13.Nxd5 Nxd5

During that twenty minutes, I thought that I would play 13...Bxd2+, but now spent another five minutes considering whether that was best. Although several of my moves in this game were not best, the time that I invested and the care taken are indicators that I was taking the game seriously. I was playing the board, not my opponent. I was not making the sort of hasty moves that cost me a draw on Saturday (see "Stronger King").

14.Qe2+

Black to move

14...Kf8

I spent another thirteen minutes on this move. I considered 14...Kd8, as well as several other options that seemed to fail tactically. I wanted to castle, but blocking the check and the castling seemed to drop a piece after White drove the knight away. Of course, White would need to castle first or face a skewer along the e-file. Blocking the check with the knight seemed to be going backwards.

My chess engine prefers 14...Kd8, which I rejected because it seemed too easy for White to move the bishops, producing a discovered attack against my king. If I could calculate as well as a computer, I might have been able to assess these dangers. 15.0–0–0 Re8 16.Qf1 Nd7 and Black is equal, according to Stockfish.

14...Kf8 is the computer's second choice.

15.0–0–0 Nd7

Stockfish prefers 15...Be7, which would have been consistent with my earlier plan to avoid exchanges. But, now, I sensed the need to catch up in development. Despite my sacrifice of a pawn for activity, my opponent's pieces seem more active.

16.Bxb4+ Qxb4 17.Bc4?

17.Qd2 and White retains the edge.

17...Re8

The game is equal.

White to move

18.Qd3??

18.c3 Nxc3 19.bxc3 Qxc3+ 20.Qc2 Qa1+= and Black forces a draw. I likely would have played something else and been worse.

After 17 moves, my opponent had 1:15 remaining to my 45 minutes. Being behind thirty minutes on the clock might have motivated me to cut my losses and bail. Happily, he spent four minutes finding a horrendous move that I quickly exploited.

18...Ne5 19.Qf1 0-1

After making this move, my opponent tipped his king before I could play 19...Nxc4.

It may be worth my time to find another line against the King's Gambit, or to meet the Bird with something other than the From. Against 1.e4, I usually play the French.

The Lesson

Teaching chess to children is a constant search for lessons that will challenge without frustrating. Sometimes the lesson is too easy; sometimes it is too hard. This week's lesson for my advanced students seemed difficult for the young students in my Thursday club. Perhaps the questions were too abstract and they needed more guidance going through the game. Perhaps the questions were clear, but the answers were too complex.

They did answer the first question rather quickly, which did not surprise me as several weeks ago they saw the games in "Patterns".

I presented them with a worksheet that consisted of a game score with one diagram and some text that included questions at the beginning. This game came to my notice while searching a position from The Art of the Checkmate (1953) by Georges Renaud and Victor Kahn. A game won by Ernest Falkbeer via Legall's checkmate is presented there. A search of the ChessBase database revealed three games that reached the diagram position. The best move, which Falkbeer executed, was not played in any of the other three. I chose the game with the highest rated players for this exercise.

Bautista Ballester,Jordi (1885) -- Mejia Fernandez,Josep (1734) [C44]
Roncana Tancats-chT 1st Santa Eulalia de Roncana (1), 23.04.2009

From the diagram position, White has a crushing attack. However, he did not find it. Rather, his move offered prospects of a slight advantage, which he later squandered. Hence, we have some questions:

1) What was the best move 8 for White?

2) How did White let the advantage slip away?

3) Where did Black err, giving White once again the upper hand?

1.e4 e5 2.d4 exd4 3.c3 dxc3 4.Bc4 Nc6 5.Nf3 d6 6.0–0 Bg4 7.Nxc3 Ne5

White to move

8.Qa4+ Qd7 9.Nxe5 Qxa4 10.Bxf7+ Ke7 11.Nxa4 dxe5 12.f3 Kxf7 13.fxg4+ Kg6 14.h4 h6 15.g5 Be7 16.Rf5 hxg5 17.Bxg5 Bf6 18.Raf1 Ne7 19.Rxf6+ gxf6 20.Rxf6+ Kg7 21.Nc5 Nc6 22.Ne6+ Kg8 23.Nxc7 Rd8 24.Nd5 Rf8 25.Rd6 Rf7 26.g4 Kh7 27.h5 Rhf8 28.Rh6+ Kg7 29.Rg6+ Kh7 30.Nf6+ Kh8 31.Rh6+ 1–0

My beginning students worked on recognizing simple forks via a worksheet that had such positions as the following.

White to move

White to move