Chess Opening Theory/1. d4/1...Nf6/2. c4/2...e5/3. dxe5/3...Ng4/4. Nf3

< Chess Opening Theory‎ | 1. d4‎ | 1...Nf6‎ | 2. c4‎ | 2...e5‎ | 3. dxe5‎ | 3...Ng4
Budapest Gambit
a b c d e f g h
8 a8 b8 c8 d8 e8 f8 g8 h8 8
7 a7 b7 c7 d7 e7 f7 g7 h7 7
6 a6 b6 c6 d6 e6 f6 g6 h6 6
5 a5 b5 c5 d5 e5 f5 g5 h5 5
4 a4 b4 c4 d4 e4 f4 g4 h4 4
3 a3 b3 c3 d3 e3 f3 g3 h3 3
2 a2 b2 c2 d2 e2 f2 g2 h2 2
1 a1 b1 c1 d1 e1 f1 g1 h1 1
a b c d e f g h
Position in Forsyth-Edwards Notation(FEN)


Moves: 1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3
ECO code: A51-A52
Parent: Indian Defence


Adler variationEdit

The Adler variation is named after the game Adler – Maróczy which was played at the 1896 Budapest tournament.[1][2] White is ready to give up the e5-pawn in order to develop all his pieces on their best squares, i.e. the d5-square for the Nb1, the f3-square for the Ng1 and the a1–h8 diagonal for the Bc1.

In the main line 4...Bc5 the f2-pawn is attacked, forcing 5.e3 that blocks the way for the Bc1. Then after 5...Nc6 White has not enough pieces to protect his e5-pawn on the long run, e.g.:

  • 6.Qd5?! is a doomed attempt, exposing the queen and occupying the d5-square that should belong to the Nb1. This move has a historical interest as this was the line played in the first game where the Budapest Gambit occurred.[3] Black continues calmly with Qe7/Ngxe5/d6/Be6.
  • 6.Bd2 O-O 7.Nc3 Qe7 does not prevent Black to regain his pawn either, and it obstructs the way to the d5-square for the Nb1. Once he has the pawn back Black has equality.[4]
  • 6.b3 immediately allows Black to play 6...d6! when White cannot capture in d6 because of Qd8–f6 winning the Ra1. Play will likely transpose into the 6.Be2 variation.

An important theoretical decision for White is to choose whether he wants to make a2–a3 part of his plan or not. While this move avoids any possible Nc6–b4 and creates the possibility b2–b4, it may also be seen as a possible waste of time in some lines. As Lalic puts it:

It was not so long ago that 8.a3, with the obvious intention of expanding with b2–b4, was the standard move. However, after Black responds with the logical a7–a5, it became apparent in tournament practice that the inclusion of these moves is in fact in Black's favour, as it gives his queen's rook access into play via the a6-square.

Black can also try the minor line 4...Nc6 that allows Black to delay the development of the f8-diagonal depending on the circumstances, and allows White to transpose into the 4.Bf4 variation if he wishes to do so.

The 4...Bc5 line with a2–a3Edit

a b c d e f g h
8 8
7 7
6 6
5 5
4 4
3 3
2 2
1 1
a b c d e f g h
After 12.O-O, the middleplay begins

1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3 Bc5 5.e3 Nc6 6.a3

After the standard moves 6...a5 7.b3 O-O 8.Bb2 Re8 9.Nc3 Ngxe5 10.Nxe5 Nxe5 11.Be2 d6 12.O-O both kings are in safety and Black has regained the invested pawn in the process. Both players can deviate at various points but the positions reached are similar, e.g. in one game White played an early Ra1–b1 in order to push b2–b4 in one move, but Black continued with the same plan as explained hereunder.[5] In another game White developed his Bf1 in d3 instead of e2, but this gave the opportunity for Black to sacrifice the Ng4 on the f2-pawn before the white castle.[6]

White has a space advantage in the center and can initiate pressure here or on the queenside by some pawn pushes like b3–b4 and c4–c5 (possibly supported by a knight on the d5-square). Meanwhile, the White king lacks some defenders so Black can start a pieces-driven attack with the standard "rook lift". As Tseitlin puts it, "the point is that 6...a5 fits into the plan of attacking White's kingside (!), whereas 6.a3 does little in the way of defending it".[7] Thus if White does not find a clear way to make good use of his move a2–a3, it may turn out to be a waste of tempo.[8]

After 12...Re6 White has to chose a way to react to the oncoming assault:

  • 13.Nd5 Rh6 14.g3 (to avoid the crushing 14...Qh4) 14...Bh3 15.Re1 when the Nd5 seems excellently placed, supporting the b3–b4 push on the queenside and one jump away from the f4-square where it can cover the weak light squares h3 and g2 if needed. Nevertheless, after 15...c6!? 16.Nf4 Bf5 it may be difficult for White to realise the b3–b4 push and the weakness of the d6-pawn is not of great significance as long as White cannot attack it with his minor pieces, so chances are level.[9] Thus placing the Nc3 in d5 may be premature.
  • 13.g3 Rh6 14.Ne4 seems crushing as White threatens to win both the Bc5 and the Ne5 (the d6-pawn is pinned). Moreover the Ne4 is well placed to support a later c4–c5 push. Black must react with 14...Qd7 in order to get out of the pin and continue the initiative. Now White has to lose a tempo and weaken his kingside further with 15.h4 in order to avoid the terrific threat Qh3–Qxh2#. After 15...Ba7 Black has dynamic equality.[9]
  • 13.Na4 is another try for White, hindering Re6–h6 because of Bxe5. Black must switch to defence on the queenside with 13...b6!? 14.Nxc5 bxc5 15.f4 (necessary to improve the prospects of the Be2) Nd7 16.Bf3 Rb8 17.Qd3 a4! when Black is even better.[10] In this plan it is important that Black does not fear to move his imprisoned Bc5, e.g. 14.Bc3 (threatening 15.b4) 14...Bd7 15.b4? Bxa4 16.Qxa4 axb4.

The 4...Bc5 line without a2–a3Edit

1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3 Bc5 5.e3 Nc6

In this variation White reasons that the advance a2–a3 is only a loss of time, because Black will play a7–a5 anyway. By refraining from the advance a2-a3 White tries to gain a tempo on the lines of the previous section, to make it more difficult for Black to initiate the Re8–e6–h6 lift. For example after the typical 6.Be2 O-O 7.O-O Re8 8.Nc3 Ngxe5 9.Nxe5 Nxe5 10.b3 d6 11.Bb2 Re6 12.g3 Rh6?! 13.Ne4 Qd7 14.h4 Black does not have the a7-square for his Bc5 because he had no time to play a7–a5.[11] As we shall see, however, a rook lift is still one of the best options for Black.

White generally continues with 6.Be2, but can instead try the tricky 6.Nc3. The immediate recapture 6...Ngxe5?! 7.Nxe5 Nxe5 allows 8.f4! Nc6 9.Bd3 when the white bishop is much more actively placed than on e2.[12] However, simply 6...O-O! 7.Be2 Ngxe5 transposes to the 6.Be2 move-order.

Note that Black should recapture the e5-pawn first with the Ng4, not the Nc6. Lalic analyzes 6.Be2 O-O 7.O-O Ncxe5?! 8.Nd4! (attacking the Ng4), when 8...Nf6 leaves Black without a clear plan and "White can expect an opening advantage".[13] Alternatively, 8...Qh4 gives White a positional advantage with 9.h3 Nh6 10.Nc3 d6 11.Nd5 ("!" Lalic) to bring the knight back to f4 in case of a Black sacrifice in h3.[14]

a b c d e f g h
8 8
7 7
6 6
5 5
4 4
3 3
2 2
1 1
a b c d e f g h
White attacks: centre or kingside?

After the standard moves 6.Be2 O-O 7.O-O Re8 8.Nc3 Ngxe5 9.Nxe5 Nxe5 White has tried two different plans:

  • The older fianchettoes the Bc1 and then plays in the centre with moves like Qd1–d5, Nc3–e4 and c4–c5. White attains a huge central space advantage, but Black can attack the kingside with the usual Ra8–a6–h6 rook-lift.
  • More recently, White has tried leaving the bishop on c1 for a time, and starting a kingside blitzkrieg with moves like f2–f4, Be2–d3 and Qd1–h5. In this line, Black tries to exploit White's weak pawn on e3.

Instead of the usual 6.Be2 White has also tried developing the queenside first, but it is dangerous to leave White's king in the centre too long. After 6.b3 O-O 7.Bb2 Re8 8.Bd3 ("?!" Lalic) d6 ("!" Lalic) 9.exd6? Nxf2! 10.Kxf2 Rxe3 11.Kf1 Bg4 Black had a crushing attack.[15]

White attacks in the centreEdit

1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3 Bc5 5.e3 Nc6 6.Be2 O-O 7.O-O Re8 8.Nc3 Ngxe5 9.Nxe5 Nxe5 10.b3

After 10.b3 a5 White can try to capture the Bc5 with 11.Na4 or 11.Ne4, one point being that the retreat 11...Ba7 would lock the Ra8 because Black has not played Ra8–a6 already. Nevertheless, Lalic still thinks 11...Ba7 is the right move after 11.Ne4 due to the importance of the a7–g1 diagonal.[16] However, Black can also reroute the bishop with 11...Bf8 and "White has no obvious path to even a minute advantage".[16] After 11.Na4 Black can also simply react by 11...b6 when the loss of the bishop pair is compensated by the semi-open B-file and the improved control on the central squares.[17] Tseitlin considers that after the exchange in c5 Black has the better position.[18] That is why the most common White move is 11.Bb2, keeping the knight jumps for later.

Then the most common plan for Black is a rook lift: the plan Ra8–a6–h6 was tried in the much-commented game Åkesson – Tagnon (Berlin Open 1984). Black duly won, but after the game continuation 11...Ra6 12.Qd5! Qe7 13.Ne4 Ba7 14.c5 Rg6 15.Rac1 Bb8 16.f4 authors do not agree on which side had the advantage. Borik considers that White had a positional advantage,[19] Tseitlin agrees and recommends 15...Nc6! instead of 15...Bb8, with dangerous threats.[20] However about 15...Bb8 Lalic says "it is true that the bishop pair look a bit pathetic lined up on the back rank just now, but there is no way to stop them breaking out later".[21]

In fact, after 11...Ra6 much depends if White goes for the moves Qd1–d5 and Nc3–e4, or deviates with Nc3–a4:

  • After 12.Qd5 several games saw 12...Ba7. A game from Kramnik (playing as Black) saw 13.Ne4 Rae6 14.Ng3 d6 15.Qxa5 Bb6 16.Qc3 Rh6 and "Black has obtained a powerful attacking position in terurn for the pawn sacricice" according to Tseitlin.[22] In another game, White varied with 13.Ne4 Rae6 14.c5!?, when Tseitlin recommends the tactical 14...d6 15.cxd6 c6 16.Qd1 f5 17.Bxe5 Rxe5 "with good prospects for Black".[23] Another attempt for White is 13.Ne4 Rae6 14.Qxa5 Bb6 15.Qc3 Qh4, when Tseitlin writes that 16.f4 is "the only move that enables White to organise resistance",[24] while Lalic considers the move dubious, saying "16.Ng3 offers more resistance".[25] Kasparov presented the line 13.c5 Rh6 14.f4 Qh4 15.h3 as being worth consideration.[26] Black had had trouble finding a good square for his attacked knight in a 1988 Yugoslav game, where 14...Nc6 lost quite quickly; hence the try 14...Qh4!?, attacking h2. Also unclear is acceptance of the piece sacrifice with 15.fxe5 Qxh2+ 16.Kf2, when Black obtains definite compensation in a very sharp position.
  • Even if 12.Qd5 Ba7 is fully satisfactory for Black, Lalic still considers 12...Qe7 best, and on 15...Bb8 he says "it is true that the bishop pair look a bit pathetic lined up on the back rank just now, but there is no way to stop them breaking out later".[27] When White tries a different move-order with 12.Ne4 Ba7 13.Qd5, Black usually opts for 13...Rae6, a promising pawn sacrifice. After the possible 14.Qxa5 Bb6 15.Qc3 Qh4 only 16.f4 enables White to organise resistance.[28]

White attacks on the kingsideEdit

1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3 Bc5 5.e3 Nc6 6.Be2 O-O 7.O-O Re8 8.Nc3 Ngxe5 9.Nxe5 Nxe5 10.Kh1

The plan of attacking on the kingside was unveiled by Spassky in a game against Illescas (Linares 1990). Black did not understand immediately White's idea, so that after 10...a5?! 11.f4 Nc6 12.Bd3 d6 13.Qh5 (awarded a "!" by Lalic) 13...h6 14.Rf3 Black's pieces were ill-placed to counter White's attack.[29]

A more principled plan for Black is to react by an action on the centre, specifically against the backward e3-pawn. After 10...d6 11.f4 Nd7 ("!" Lalic) 12.Bd3 Nf6 13.Qf3 Ng4 14.Nd1 f5 ("!" Lalic) and Black has succeeded in avoiding White's expansion e3–e4.[30]

The 4...Nc6 lineEdit

1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3 Nc6

Black plays this when he wants to postpone the placement of its dark-squared bishop. Now White has a wide choice:

  • 5.Bf4 transposes in the 4.Bf4 variation explained hereafter.
  • 5.Qd5 transposes in the minor line 4.Qd5 explained hereafter.
  • 5.e3 and now 5...Bb4+ is not that good because White can react with the simple 5.Bd2. Better for Black is 5...Ngxe5 when Black can go into a kind of King's Indian Defence setup with g7–g6 and Bf8–g7.[31] Then the pressure along the a1–h8 diagonal can be enhanced via the quick advance a7–a5–a4–a3. For example after 5.e3 Ngxe5 6.Be2 g6 ("!?" Lalic) 7.O-O Bg7 8.Nc3 O-O 9.Qd2 d6 10.h3 ("?" Lalic) 10...a5 ("!" Lalic) 11.b3 a4 and now 12.Bb2 would have been followed by 12.a3! 13.Bc1 Nxf3+ 14.Bxf3 Qf6 winning the Nc3.[32]
  • 5.Nc3 will transpose into the 4...Bc5 line if Black plays 5...Bc5, but Black can also wait a bit to see what White is up to, e.g. 5...Ngxe5 6.Nxe5 Nxe5 7.Qc2 Bb4 when both players are still hesitating to castle long or short.[33]
  • 5.Bg5 Be7 6.Bxe7 (6.Bf4 Bb4+ transposes in the 4.Bf4 variation) 6...Qxe7 7.Nc3 with the dangerous positional threat Nc3–d5. Here Borik advocates 7...Qc5 8.e3 Ngxe5, when he can react to Qd1–d5 with Qc5–e7 (and the d5-square is no more available to the Nc3), and to Nc3–d5 with Nc6–e7 (to exchange the annoying knight).[31] Black can also delay the recapture of the e5-pawn with 7...O-O 8.Nd5 Qd8 9.e3 Ngxe5 10.Nxe5 Nxe5.[34] Meanwhile, the natural 7...Ngxe5 falls into White's positional trap and after 8.Nd5 ("!?" Lalic) 8...Qd8 9.Nxe5 Nxe5 10.Qd4 f6 11.f4 Ng6 12.Qe4+ Kf7 White got an edge.[35]

Theory tableEdit

For explanation of theory tables see theory table and for notation see algebraic notation

1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Nf3

4 5 6 7 8


When contributing to this Wikibook, please follow the Conventions for organization.


  1. Oleinikov chapter 5
  2. "ECO classification of WCCF". Retrieved 2008-05-25. 
  3. Adler – Maroczy, Budapest 1896, won by Black
  4. Borik 1986, p.12
  5. Tseitlin 1992, p.134
    Gavrilov – Berdichevsky, Moscow 1989
  6. Tseitlin 1992, p.135
    Yrjola – Liew, Dubai Olympiad 1986
  7. Tseitlin 1992, p.83
  8. Tseitlin 1992, p.87
  9. a b Borik 1986, p.15
  10. Borik 1986, p.16
  11. Borik 1986, p.18
  12. Lalic 1998, p.74
    Razuvaev – Bardel, Geneva 1995
  13. Lalic 1998, p.83
    Lovass – Husari, Kecskemet 1991
  14. Lalic 1998, p.84
    F. Portisch – Ivan, Zalakaros 1994
  15. Lalic 1998, p.84
    Alexandria – Schnepp, Biel open 1994
  16. a b Lalic 1998, p.79
  17. Borik 1986, p.19
    Osnos – Yermolinsky, Leningrad 1977
  18. Tseitlin 1992, p.78
  19. Borik 1986, p.17
  20. Tseitlin 1992, p.80
  21. Lalic 1998, p.76
  22. Tseitlin 1992, p.133
    Odessky – Kramnik, USSR 1987
  23. Tseitlin p.134
    Gurevich – Korchnoi, Madrid 1988 (rapid)
  24. Tseitlin, p.134
    Vainerman – Legky, USSR 1986
  25. Lalic 1998, p.79
    Whiteley – Agnos, London 1994
  26. Kasparov 1989
  27. Lalic 1998, p.76
  28. Tseitlin 1992, p.82
  29. Tseitlin 1992, p.132
    Spassky – Illescas, Linares 1990
  30. Lalic 1998, p.81
    Alekseev – Bliumberg, Minsk 1993
  31. a b Borik 1986, p.11
  32. Lalic 1998, p.90
    Maurer – Nurkic, Imperia 1990
  33. Lalic 1998, p.91
    Hebden – Hodgson, Guernsey 1985
  34. Lalic 1998, p.92
    Polugaevsky – Nunn
  35. Lalic 1998, p.94
    Laketic – Gavric, Yugoslavian team championship 1994
  • Nunn's Chess Openings. 1999. John Nunn (Editor), Graham Burgess, John Emms, Joe Gallagher. ISBN 1-8574-4221-0.
  • Batsford Chess Openings 2 (1989, 1994). Garry Kasparov, Raymond Keene. ISBN 0-8050-3409-9.