|Moves||1.e4 e5 2.f4|
|Origin||No later than 16th century|
The King's Gambit is a chess opening that begins with the moves:
White offers a pawn to divert the black e-pawn. If Black accepts the gambit, White has two main plans. The first is to play d4 and Bxf4, regaining the gambit pawn with central domination. The alternative plan is to play Nf3 and Bc4 followed by 0-0, when the semi-open f-file allows White to barrel down onto the weakest point in Black's position, the pawn on f7. Theory has shown that in order for Black to maintain the gambit pawn, he may well be forced to weaken his , with moves such as ...g5 or odd piece placement (e.g. ...Nf6–h5). A downside to the King's Gambit is that White weakens his own king's position, exposing it to the latent threat of ...Qh4+ (or ...Be7–h4+). With a black pawn on f4, White cannot usually respond to the check with g3, but if the king is forced to move then it also loses the right to castle.
The King's Gambit is one of the oldest documented openings, as it was examined by the 17th-century Italian chess player Giulio Cesare Polerio. It is also in an older book by Luis Ramírez de Lucena.
The King's Gambit was one of the most popular openings in the 19th century, but is infrequently seen at master level today, as Black can obtain a reasonable position by returning the extra pawn to . There are two main branches, depending on whether or not Black plays 2...exf4: the King's Gambit Accepted (KGA) and the King's Gambit Declined (KGD).
- 1 History
- 2 Variations
- 2.1 King's Gambit Accepted: 2...exf4
- 2.1.1 King's Knight's Gambit: 3.Nf3
- 220.127.116.11 Classical Variation: 3...g5
- 18.104.22.168 Becker Defense: 3...h6
- 22.214.171.124 Bonch–Osmolovsky Defense: 3...Ne7
- 126.96.36.199 Cunningham Defense: 3...Be7
- 188.8.131.52 Schallopp Defense: 3...Nf6
- 184.108.40.206 Modern Defense: 3...d5
- 220.127.116.11 Fischer Defense: 3...d6
- 18.104.22.168 MacLeod Defense: 3...Nc6
- 22.214.171.124 Wagenbach Defense: 3...h5
- 2.1.2 Bishop's Gambit 3.Bc4
- 2.1.3 Other 3rd moves for White
- 2.1.1 King's Knight's Gambit: 3.Nf3
- 2.2 King's Gambit Declined
- 2.1 King's Gambit Accepted: 2...exf4
- 3 Related lines
- 4 ECO
- 5 References
- 6 Further reading
- 7 External links
|This article uses algebraic notation to describe chess moves.|
The King's Gambit was one of the most popular openings for over 300 years, and has been played by many of the strongest players in many of the greatest , including the Immortal Game. Nevertheless, players have held widely divergent views on it. François-André Danican Philidor (1726–1795), the greatest player and theorist of his day, wrote that the King's Gambit should end in a draw with best play by both sides, stating that "a gambit equally well attacked and defended is never a decisive [game], either on one side or the other." Writing over 150 years later, Siegbert Tarrasch, one of the world's strongest players in the late 19th and early 20th centuries, pronounced the opening "a decisive mistake" and wrote that "it is almost madness to play the King's Gambit." Similarly, future World Champion Bobby Fischer wrote a famous article, "A Bust to the King's Gambit", in which he stated, "In my opinion the King's Gambit is busted. It loses by force" and offered his Fischer Defense (3...d6) as a refutation. FM Graham Burgess, in his book The Mammoth Book of Chess, noted the discrepancy between the King's Gambit and Wilhelm Steinitz's accumulation theory. Steinitz had argued that an attack is only justified when a player has an advantage, and an advantage is only obtainable after the opponent makes a mistake. Since 1...e5 does not look like a blunder, White should therefore not be launching an attack.
None of these pronouncements, however, have been proved to be actual refutations of the King's Gambit. In 2012, an April Fool prank by Chessbase in association with Vasik Rajlich—inventor of chess engine Rybka—claimed to have proven to a 99.99999999% certainty that the King's Gambit is at best a draw for White. In a later post, owning up to the prank, Rajlich estimated that "we're still probably a good 25 or so orders of magnitude away from being able to solve something like the King's Gambit. If processing power doubles every 18 months for the next century, we'll have the resources to do this around the year 2120, plus or minus a few decades."
While the King's Gambit Accepted was a staple of Romantic era chess, the opening began to decline with the rise of positional play in the 1870s, although King's Gambit Declined variants remained popular among high-level players until World War I. By the 1920s, 1.e4 openings declined in popularity with the rise of the hypermodern school, with many players switching to 1.d4 and 1.c4 openings and positional play.
After World War II, 1.e4 openings became more popular again, with David Bronstein being the first grandmaster in decades to use the King's Gambit in serious play. He inspired Boris Spassky to also take up the King's Gambit, although Spassky was not willing to risk using the opening in any of his World Championship matches. However, Spassky did beat many strong players with it, including Bobby Fischer, Zsuzsa Polgar, and a famous brilliancy against Bronstein himself.
The King's Gambit is rare in modern grandmaster play, and even rarer at the top level. A handful of grandmasters have continued to use it, including Joseph Gallagher, Hikaru Nakamura, Nigel Short, and Alexei Fedorov.
Although Black usually accepts the gambit pawn, two methods of declining the gambit—the classical variation (2...Bc5) and the Falkbeer Counter Gambit (2...d5)—are also popular.
King's Gambit Accepted: 2...exf4 
If Black accepts with 2...exf4, the two main continuations for White are:
- 3.Nf3 (King Knight's Gambit)
- 3.Bc4 (Bishop's Gambit)
3.Nf3 is the most common as it develops the knight and prevents 3...Qh4+.
King's Knight's Gambit: 3.Nf3 
Classical Variation: 3...g5 
The Classical Variation arises after 3.Nf3 g5. Black defends his extra pawn, and threatens to kick the f3-knight with ...g4. The main continuations traditionally have been 4.h4 (the Paris Attack), and 4.Bc4. However, recently 4.Nc3 (the Quaade Gambit or Quaade Attack) has been recommended by Scottish grandmaster John Shaw as a less explored alternative to 4.h4 (he considers 4.Bc4 inferior).
4.h4: Kieseritzky Gambit and Allgaier Gambit 
After 4.h4 g4 the main line is 5.Ne5, the Kieseritzky Gambit. It was popularized by Lionel Kieseritzky in the 1840s and used successfully by Wilhelm Steinitz. Boris Spassky used it to beat Bobby Fischer in a famous game at Mar del Plata in 1960. This motivated Fischer to develop his own defense to the King's Gambit (see Fischer Defense).
4.Bc4 g4: Muzio Gambit and others 
The extremely Muzio Gambit arises after 4.Bc4 g4 5.0-0 gxf3 6.Qxf3, where White has gambited a knight but has three pieces bearing down on f7. Such wild play is rare in modern chess, but Black must exercise care in consolidating his position. Perhaps the sharpest continuation is the Double Muzio after 6...Qf6 7.e5 Qxe5 8.Bxf7+!?, leaving White two pieces down in eight moves, but with a position that some masters consider to be equal.
Similar lines are the Ghulam Kassim Gambit, 4.Bc4 g4 5.d4, and the McDonnell Gambit, 4.Bc4 g4 5.Nc3. These are generally considered inferior to the Muzio, which has the advantage of reinforcing White's attack along the f-file. Also inferior is the Lolli Gambit 4.Bc4 g4 5.Bxf7+?!, which leaves White with insufficient compensation for the piece after 5...Kxf7 6.Ne5+ Ke8 7.Qxg4 Nf6 8.Qxf4 d6.
The Salvio Gambit, 4.Bc4 g4 5.Ne5 Qh4+ 6.Kf1, is considered better for Black due to the insecurity of White's king. Black may play safely with 6...Nh6, or counter-sacrifice with 6...f3 or 6...Nc6.
4.Bc4 Bg7: Hanstein Gambit and Philidor Gambit 
A safer alternative to 4...g4 is 4...Bg7, which usually leads to the Hanstein Gambit after 5.d4 d6 6.0-0 h6 or the Philidor Gambit after 5.h4 h6 6.d4 d6 (other move orders are possible in both cases).
4.Nc3: Quaade Gambit 
The Quaade Gambit (3.Nf3 g5 4.Nc3) is named after a Danish amateur who discussed it in correspondence with the Deutsche Schachzeitung in the 1880s. The move has received renewed attention following its recommendation by John Shaw in his 2013 book on the King's Gambit. A well-known trap here is 4...g4 5.Ne5 Qh4+ 6.g3 fxg3 7.Qxg4 g2+? (7...Qxg4 8.Nxg4 d5 is about equal) 8.Qxh4 gxh1=Q 9.Qh5! and White is close to winning. (Black's best defense is considered to be 9...Nh6 10.d4 d6 11.Bxh6 dxe5 12.Qxe5+ Be6 13.Qxh8 Nd7 14.Bxf8 0-0-0 and White will emerge a clear pawn ahead.) Instead, 4...Bg7 has been recommended. 4...d6 and 4...h6 transpose to Fischer's Defense and Becker's Defense, respectively. Also possible is 4...Nc6, recommended by Konstantin Sakaev.
4.d4: Rosentreter Gambit 
This is likely to lead to similar positions to the Quaade Gambit, however 4...g4 5.Ne5 Qh4+ 6.g3 fxg3 7.Qxg4 g2+!? (7...Qxg4=) is now viable due to the threat against the pawn on e4. After 8.Qxh4 gxh1=Q Shaw recommends 9.Nc3 for White, with a complicated position.
Becker Defense: 3...h6 
The Becker Defense (3.Nf3 h6), has the idea of creating a on h6, g5, f4 to defend the f4 pawn while avoiding the Kieseritzky Gambit, so Black will not be forced to play ...g4 when White plays to undermine the chain with h4. White has the option of 4.b3, although the main line continues with 4.d4 g5 (ECO C37) and usually transposes to lines of the Classical Variation after 5.Bc4 Bg7 6.0-0 (ECO C38).
Bonch–Osmolovsky Defense: 3...Ne7 
The rarely seen Bonch–Osmolovsky Defense (3.Nf3 Ne7) was played by Mark Bluvshtein to defeat former world title finalist Nigel Short at Montreal 2007, even though it has never been highly regarded by theory.
Cunningham Defense: 3...Be7 
The Cunningham Defense (3.Nf3 Be7) threatens a check on h4 that can permanently prevent White from castling; furthermore, if White does not develop his King's Bishop immediately, he would be forced to play Ke2, which hems the Bishop in. A sample line is 4. Nc3 Bh4+ 5. Ke2 d5 6. Nxd5 Nf6 7. Nxf6+ Qxf6 8. d4 Bg4 9. Qd2 (diagram). White has strong central control with pawns on d4 and e4, while Black is relying on the white king's discomfort to compensate.
To avoid having to play Ke2, 4. Bc4 is White's most popular response. Black can play ...Bh4+ anyway, forcing 5.Kf1 (or else the wild Bertin Gambit or Three Pawns' Gambit, 5.g3 fxg3 6.0-0 gxh2+ 7.Kh1, played in the nineteenth century). In modern practice, it is more common for Black to simply develop instead with 4...Nf6 5.e5 Ng4, known as the Modern Cunningham.
Schallopp Defense: 3...Nf6 
The Schallopp Defense (3.Nf3 Nf6) – intending 4.e5 Nh5, holding onto the pawn – is considered somewhat inferior and is rarely played today. In one of the lines, White can usually obtain a crushing attack via a rook sacrifice, 4.e5 Nh5 5.d4 g5 6.h4 g4 7.Ng5 Ng3 8.Bc4! Nxh1 9.Bxf7+ Ke7 10.Nc3 (looking for immediate mate at d5, or later via queen at f6) and Black appears doomed.
Modern Defense: 3...d5 
The Modern Defense, or Abbazia Defense, (3.Nf3 d5) has much the same idea as the Falkbeer Counter-Gambit, and can in fact be reached by transposition, e.g. 2.f4 d5 3.exd5 exf4. Black concentrates on gaining piece play and fighting for the initiative rather than keeping the extra pawn. It has been recommended by several publications as an easy way to equalize, although White keeps a slight advantage due to his extra central pawn and piece activity. If White captures (4.exd5) then Black may play 4...Nf6 or recapture with 4...Qxd5, at which point it becomes the Scandinavian Variation of KGA.
Fischer Defense: 3...d6 
"The refutation of any gambit begins with accepting it. In my opinion the King's Gambit is busted. It loses by force." – R. Fischer, "A Bust to the King's Gambit"
The Fischer Defense (3.Nf3 d6), although previously known, was advocated by Bobby Fischer after he was defeated by Boris Spassky in a Kieseritzky Gambit at the 1960 Mar del Plata tournament. Fischer then decided to refute the King's Gambit, and the next year the American Chess Quarterly published Fischer's analysis of 3...d6, which he called "a high-class waiting move".
The point is that after 4.d4 g5 5.h4 g4 White cannot continue with 6.Ne5 as in the Kieseritzky Gambit, 6.Ng5 is unsound because of 6...f6! trapping the knight, and 6.Nfd2 blocks the bishop on c1. This leaves the move 6.Ng1 as the only option, when after six moves neither side has developed a piece. The resulting slightly odd position (diagram) offers White good attacking chances. A typical continuation is 6.Ng1 Bh6 7.Ne2 Qf6 8.Nbc3 c6 9.g3 f3 10.Nf4 Qe7 with an unclear position. (Korchnoi/Zak)
The main alternative to 4.d4 is 4.Bc4. Play usually continues 4...h6 5.d4 g5 6.0-0 Bg7, transposing into the Hanstein Gambit, which can also be reached via 3...g5 or 3...h6.
MacLeod Defense: 3...Nc6 
Joe Gallagher writes that 3.Nf3 Nc6 "has never really caught on, probably because it does nothing to address Black's immediate problems." Like Fischer's Defense, it is a waiting move. An obvious drawback is that the Nc6 may prove a target for the d-pawn later in the opening.
Wagenbach Defense: 3...h5 
An invention of the Hungarian/English player, János Wagenbach. John Shaw writes: "If given the time, Black intends to seal up the kingside with ...h4 followed by ...g5, securing the extra pawn on f4 without allowing an undermining h2–h4. The drawback is of course the amount of time required".
Bishop's Gambit 3.Bc4 
Of the alternatives to 3.Nf3, the most important is the Bishop's Gambit, 3. Bc4.
White allows 3...Qh4+ 4.Kf1, losing the right to castle, but this loses time for Black after the inevitable Nf3 and White will develop rapidly.
Korchnoi and Zak recommend as best for Black 3...Nf6 4.Nc3 c6, or the alternative move order 3...c6 4.Nc3 Nf6. After 5.Bb3 d5 6.exd5 cxd5 7.d4 Bd6 8.Nge2 0-0 9.0-0 g5 10.Nxd5 Nc6, Black was somewhat better in Spielmann–Bogoljubow, Märisch Ostrau 1923.
Black's other main option is 3...d5, returning the pawn immediately. Play might continue 3...d5 4.Bxd5 Nf6 5.Nc3 Bb4 6.Nf3 Bxc3 7.dxc3 c6 8.Bc4 Qxd1+ 9.Kxd1 0-0 10.Bxf4 Nxe4 with an equal position (Bilguer Handbuch, Korchnoi/Zak).
3...Nc6!? is relatively untested, but if White plays 4.Nf3 Black can transpose into the Hanstein Gambit after 4...g5 5.d4 Bg7 6.c3 d6 7.0-0 h6 (Neil McDonald, 1998).
Other 3rd moves for White
Other 3rd moves for White are rarely played. Some of these are:
- 3.Nc3 (Mason Gambit or Keres Gambit)
- 3.d4 (Villemson Gambit or Steinitz Gambit)
- 3.Be2 (Lesser Bishop's Gambit or Tartakower Gambit)
- 3.Qf3 (Breyer Gambit or Hungarian Gambit)
- 3.h4 (Stamma Gambit)
- 3.Nh3 (Eisenberg Gambit)
King's Gambit Declined 
Falkbeer Countergambit: 2...d5 
The Falkbeer Countergambit is named after the 19th-century Austrian master Ernst Falkbeer. It runs 1.e4 e5 2.f4 d5 3.exd5 e4, in which Black sacrifices a pawn in return for quick and easy . It was once considered good for Black and scored well, but White obtains some advantage with the response 4.d3!, and the line fell out of favour after the 1930s.
A more modern interpretation of the Falkbeer is 2...d5 3.exd5 c6!?, as advocated by Aron Nimzowitsch. Black is not concerned about pawns and aims for early piece activity. White has a better pawn structure and prospects of a better endgame. The main line continues 4.Nc3 exf4 5.Nf3 Bd6 6.d4 Ne7 7.dxc6 Nbxc6, giving positions analogous to the Modern Variation of the gambit accepted.
Classical Defense: 2...Bc5 
A common way to decline the gambit is with 2...Bc5, the "classical" KGD. The bishop prevents White from castling and is such a nuisance that White often expends two tempi to eliminate it by means of Nc3–a4, to exchange on c5 or b6, after which White may castle without worry. It also contains an opening trap for novices: if White continues with 3.fxe5?? Black continues 3...Qh4+, in which either the rook is lost (4.g3 Qxe4+, forking the rook and king) or White is checkmated (4.Ke2 Qxe4#). This line often comes about by transposition from lines of the Vienna Game or Bishop's Opening, when White plays f2–f4 before Nf3.
Other 2nd moves for Black
Other options in the KGD are possible, though unusual, such as the Adelaide Countergambit 2...Nc6 3.Nf3 f5, advocated by Tony Miles; 2...d6, when after 3.Nf3, best is 3...exf4 transposing to the Fischer Defense (though 2...d6 invites White to play 3.d4 instead); and 2...Nf6 3.fxe5 Nxe4 4.Nf3 Ng5! 5.d4 Nxf3+ 6.Qxf3 Qh4+ 7.Qf2 Qxf2+ 8.Kxf2 with a small endgame advantage, as played in the 1968 game between Bobby Fischer and Robert Wade in Vinkovci. The greedy 2...Qf6 (known as the Norwalde Variation), intending 3...Qxf4, is considered dubious. Also dubious are the Keene Defense: 2...Qh4+ 3.g3 Qe7 and the Mafia Defense: 1.e4 e5 2.f4 c5.
2...f5?! is among the oldest countergambits in KGD, known from a game published in 1625 by Gioachino Greco. Vincenz Hruby also played it against Mikhail Chigorin in 1882. It is nonetheless considered dubious because 3.exf5 with the threat of Qh5+ gives White a good game. The variation is sometimes named the Pantelidakis Countergambit because Grandmaster Larry Evans answered a question from Peter Pantelidakis of Chicago about it in one of his columns in Chess Life and Review.
In several lines of the Vienna Game White offers a sort of delayed King's Gambit. In the Vienna Gambit (1.e4 e5 2.Nc3 Nf6 3.f4), Black should reply 3...d5, since 3....exf4?! 4.e5 forces the knight to retreat. 1.e4 e5 2.Nc3 Nc6 3.f4 exf4 may lead to the Hamppe-Muzio Gambit after 4.Nf3 g5 5.Bc4 g4 6.O-O gxf3 7.Qxf3, or to the Steinitz Gambit after 4.d4 Qh4+ 5.Ke2. Both of these lines may be reached via the King's Gambit proper, but the Vienna move order is more common.
White may also offer the gambit in the Bishop's Opening, e.g. 1.e4 e5 2.Bc4 Nc6 3.f4, though this is uncommon.
The Encyclopaedia of Chess Openings (ECO) has ten codes for the King's Gambit, C30 through C39.
- C30: 1.e4 e5 2.f4 (King's Gambit)
- C31: 1.e4 e5 2.f4 d5 (Falkbeer Countergambit)
- C32: 1.e4 e5 2.f4 d5 3.exd5 e4 4.d3 Nf6 (Morphy, Charousek, etc.)
- C33: 1.e4 e5 2.f4 exf4 (King's Gambit Accepted)
- C34: 1.e4 e5 2.f4 exf4 3.Nf3 (King's Knight's Gambit)
- C35: 1.e4 e5 2.f4 exf4 3.Nf3 Be7 (Cunningham Defense)
- C36: 1.e4 e5 2.f4 exf4 3.Nf3 d5 (Abbazia Defense)
- C37: 1.e4 e5 2.f4 exf4 3.Nf3 g5 4.Nc3 /4.Bc4 g4 5.0-0 (Muzio Gambit)
- C38: 1.e4 e5 2.f4 exf4 3.Nf3 g5 4.Bc4 Bg7 (Philidor, Hanstein, etc.)
- C39: 1.e4 e5 2.f4 exf4 3.Nf3 g5 4.h4 (Allgaier, Kieseritzky, etc.)
- Ristoja, Thomas; Aulikki Ristoja (1995). Perusteet. Shakki (in Finnish). WSOY. p. 58. ISBN 951-0-20505-2.
- Hooper, David; Kenneth, Whyld (1992), The Oxford Companion to Chess (2nd ed.), Oxford University Press, ISBN 0-19-866164-9
- Philidor, François-André Danican (2005), Analysis of the Game of Chess (1777) (2nd ed.), Harding Simple Ltd., p. 67, ISBN 1-84382-161-3
- Tarrasch, Siegbert (1938), The Game of Chess, David McKay, p. 309
- Bobby Fischer, "A Bust to the King's Gambit", American Chess Quarterly, Summer 1961, pp. 3–9.
- Fischer, Bobby (1961). "A Bust to the King's Gambit" (PDF). ChessCafe.com. Retrieved 2009-11-08.
- Burgess, Graham (2010), The Mammoth Book of Chess, Running Press
- "Rajlich: Busting the King's Gambit, this time for sure".
- "The ChessBase April Fools revisited".
- "The ChessBase April Fools revisited".
- Spassky vs. Fischer, Mar del Plata 1969
- Spassky vs Polgar, Plaza 1988
- Spassky vs. Bronstein, USSR Championship 1960
- "Medias R4: Carlsen plays the King's Gambit in the King's Tournament!". Chessbase. Retrieved 31 May 2016.
- For the origin of the term "Quaade Attack" or "Quaade Gambit" see "A Chess Gamelet" by Edward Winter, 2014
- John Shaw, The King's Gambit, Quality Chess, 2013, p. 137. ISBN 978-1-906552-71-8.
- Spassky vs. Fischer, Mar del Plata 1960 Chessgames.com
- Kasparov, Gary; Keene, Raymond (1982). Batsford Chess Openings. American Chess Promotions. pp. 288–89. ISBN 0-7134-2112-6.
- For the origins of the name "Muzio" and how the eponymous variation came to be labeled, see Polerio Gambit
- Nakamura vs. Andreikin
- Peter Millican 1989
- Shirov vs. J Lapinski
- Edward Winter, A Chess Gamelet, 5 March 2014
- Shaw, p. 141
- Korchnoi & Zak, p. 38-39.
- Shaw, p. 186-196
- Named after Soviet national master Mikhail Aleksandrovich Bonch-Osmolovsky (1919-1975), also chess theorist and arbiter. See Mikhail Aleksandrovich Bonch-Osmolovsky at ChessGames.com
- Short vs. Bluvshtein
- "Chess Opening Explorer". Chessgames.com. Retrieved 7 June 2016.
- The name comes from a tournament, played in Abbazia in 1912, in which all the games had to be a King's Gambit Accepted. The town, at the time in the Austria-Hungary empire, is now in modern Croatia
- Joe Gallagher, Winning with the King's Gambit, Henry Holt, 1993, p. 105. ISBN 0-8050-2631-2.
- John Shaw, The King's Gambit, Quality Chess, 2013, p. 431. ISBN 978-1-906552-71-8.
- Named after Martin Villemson (1897-1933) of Pärnu, Estonia, editor of the chess magazine Eesti Maleilm. See Oxford Companion to Chess, Oxford University Press, Oxford, UK, 1984
- Fischer vs. Wade
- King's Gambit: Declined, Mafia Defense, ChessGames.com
- anonymous vs. Greco, ChessGames.com
- Chigorin vs. Hruby, Vienna 1882, 365chess.com
- Korchnoi, Victor; Zak, V. G. (1974). King's Gambit. Batsford. ISBN 9780713429145.
- Estrin, Yakov; Glazkov, I. B. (1982). Play the King's Gambit. ISBN 978-0080268736.
- Schiller, Eric (1989). Who's Afraid of the King's Gambit Accepted?. Thinkers Pr Inc / Chessco. ISBN 978-0931462900.
- Gallagher, Joe (1993). Winning With the King's Gambit. Henry Holt. ISBN 978-0805026313.
- McDonald, Neil (1998). The King's Gambit. Batsford. ISBN 978-0-7134-8451-9.
- Johansson, Thomas; Wallin, Maria (Illustrator) (2005). The Fascinating King's Gambit. Trafford on Demand. ISBN 9781412046473.
- Shaw, John (2013). The King's Gambit. Quality Chess. ISBN 978-1-906552-71-8.
|The Wikibook Chess Opening Theory has a page on the topic of: King's Gambit|