The Danish Gambit is essentially a Göring Gambit with the moves Nf3 and ...Nc6 omitted. In some lines play often transposes into the Göring Gambit, especially the lines with 3...d3, 3...d5, and 3...dxc3 4.Nxc3, as Nf3 and ...Nc6 often follow, but there are some important differences. I am currently preparing to upload some updated games and analysis to supplement the Danish Gambit part of my Gambiteer's Guild site and have been looking over the declined lines. If White wishes to meet 3...dxc3 with 4.Nxc3 then the Danish Gambit move-order is probably no better or worse than the Göring Gambit version, but if White wants to offer the second pawn with 4.Bc4 then I strongly recommend that White plays Nf3 before c3, since the "pure" Danish Gambit with 4.Bc4 has numerous additional drawbacks. In the next blog article I expect to be able to explore these after having looked at some recent games in the line.
Perhaps the most important difference in the declined lines concerns the declined variation with 3...Ne7, known as the Svenonious Defence, although this line continues to be surprisingly neglected at high levels of play, and there are few practical examples. In the Göring Gambit version with 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.c3 Nge7, White can get quite a good position with 5.Bc4 d5 6.exd5 Nxd5 7.0-0, but in the Danish Gambit version, White has no way to transpose into that line. For example, 4.Bc4 d5 5.exd5 Nxd5 6.Nf3, hoping for 6...Nc6, is thwarted by 6...Nb6, which gives Black a comfortable equality, while 4.Nf3 d5 5.exd5 (5.Qxd4 Nbc6 6.Bb5 Bd7) 5...Nxd5 6.Bc4 Nb6 transposes.
Thus White's best option against 3...Ne7 is 4.cxd4 d5 and now 5.e5 isn't too bad, since if 5...Nbc6, inviting 6.Nf3 Bg4, White side-steps the pin with 6.Nc3. White will aim to generate play on the kingside while Black will try to undermine White's centre. In the game Voigt-Hector, Hamburg 2000, following 5...Nf5 6.Nc3 Be7 7.Nf3 0-0 8.Bd3 Nc6
9.Ne2 f6 10.a3?!, White got into trouble following 10...fxe5 11.dxe5 Nh4, but could have improved with 9.Bc2, or 10.0-0, both of which promise White some attacking chances although objectively the chances are approximately equal.
5.Nc3!? is also worth investigating: 5...dxe4 6.Bc4 Nf5 7.Nge2 "with compensation" was suggested in Danish Dynamite, and play can then continue with 7...Nd6 8.Bb3.
With best play White will generally regain the pawn on e4 and emerge with a typical isolated queen's pawn position with equal chances for both sides, since it is risky for Black to try to hold onto the extra pawn. For example, pushing with ...f7-f5 runs into Ne2-f4 and Black's light squares around the king end up very weak.
I don't think White needs to worry too much about 3...d3 4.Bxd3, with the idea 5.Nf3 and following one of the two favourable plans in the analogous Göring Gambit variation (either 0-0 with the idea Nd4 and f4, as in D.Velimirovic-A.Muratovic, Serbia 2005, or a plan with Bf4, Nbd2, Qc2, and castling to either side.) Black ends up with a passive position.
3...Nf6 4.e5 is not much of an issue for White either, though following 4...Nd5 White's best is probably 5.Qxd4 since the attempt to transpose into Göring Gambit lines with 5.Nf3 runs into 5...d6, challenging the pawn on e5.
The most important of Black's declining lines, though, is 3...d5 which will be encountered very frequently in over-the-board play.
Here White has no good alternative to 4.exd5, since after 4.Bd3?! dxe4 5.Bxe4 Nf6, there is no knight on c6 for White to trade the e4-bishop for, and 6.Bg5 Be7 7.Bxf6 Bxf6 doesn't really help White. Black keeps an extra pawn for limited compensation.
After 4.exd5 Black can try 4...Nf6!? after which White's best way to mix things is probably 5.Bb5+, when one of the most important continuations is 5...Bd7 6.Bc4 b5!? 7.Bb3 dxc3 8.Nxc3 b4 9.Nce2 Bd6. Chances are roughly level in this position.
In the main line with 4...Qxd5 5.cxd4 Nc6, White normally transposes into the Göring Gambit with 6.Nf3, whereupon White can avoid Capablanca's Defence with Mark Nieuweboer's suggestion 6...Bg4 7.Nc3 Bb4 8.Be3, and if 8...Qa5 then 9.Qb3, though this allows the deviation 7...Bxf3 which leads to a sharp endgame which offers equal chances. White has also experimented with 6.Be3, which will lead to a very similar situation after 6...Bb4+ 7.Nc3 Nf6 8.Nf3 (8.Nge2 is less convincing because of 8...Bg4), and again 8...Qa5 is dealt with via 9.Qb3. This avoids the aforementioned 7...Bxf3 line, but allows Black a few alternatives to transposing with 9...Bg4, such as 9...0-0 (F.Nepustil-F.Cottegnie, email 2006), 9...Ne4 (which is best met by 10.Rc1) or 9...Be6. I don't see a way for Black to prevent White from achieving dynamic equality, though, so 6.Be3 appears to be a reasonable deviation.
Finally, there is 3...Qe7, which counterattacks against e4 and is more effective than it is against the
Göring Gambit. White should seek compensation for a pawn with 4.cxd4 Qxe4+ 5.Be3 (although the dubious second pawn sacrifice 5.Be2 Qxg2 might work in blitz games). Danish Gambit aficionado Martin Voigt won quite an attractive attacking game as White in this line, although Black overlooked some important defensive resources (most notably 14...g6!) My feeling is that White gets sufficient compensation for the pawn in this line, but no more, which means that the rather cheeky 3...Qe7 is a reasonable alternative to 3...dxc3 if Black wishes to hold onto an extra pawn. Note that 4.Qxd4?! Nc6 5.Qe3 avoids the loss of a pawn but transposes into a poor line of the Centre Game (White would normally play Nb1-c3 rather than c2-c3).