Danish Gambit

Strategic Summary & Main Idea

The Danish Gambit gives up a pawn on move three and, if Black accepts, gives up a second: 3.c3 dxc3 4.Bc4 cxb2 5.Bxb2 leaves White with two bishops aimed at f7 and g7 in exchange for two pawns. Measured, the deal does not close. The position after 3.c3 measures −0.26 and the double acceptance −0.10 — Black stays better in every version. The gambit remains a practical weapon at fast time controls, because the correct defence requires giving material back at the right moment, and giving material back is the least practised skill there is.

The pawn structure this opening produces

What the Danish Gambit buys is not structure but diagonals. After 5.Bxb2, White has no pawns beyond e4 and a kingside majority, and all the value of the position sits in two bishops: one on c4 eyeing f7, the other on b2 eyeing g7 and the long diagonal. While Black's king is in the centre, those two diagonals are worth more than two pawns; once he castles and gives material back with ...d5, they stop being worth it. That is why the whole theory of the defence reduces to one rule: return the b2 or c3 pawn at the moment the return gains a tempo of development. White's alternative, transposing into the Göring Gambit with 4.Nxc3, swaps both diagonals for normal development and measures −0.20 — much the same. And declining with 3...d3, returning the pawn immediately, is the one black choice that comes out worse: White recovers the pawn effortlessly and keeps the centre, and the position measures +0.87. The gambit is bad; declining it is worse.

Main Variations of the Danish Gambit

Evaluations measured by us with Stockfish 19, depth 22, on 14/09/2026.

Frequently asked questions

Is the Danish Gambit sound?

Objectively no: the position after 3.c3 measures −0.26 and the double acceptance −0.10. The two bishops compensate for almost everything, but almost is not everything, and the 'almost' disappears as soon as Black returns material and castles. As a fast-chess weapon it still works, because it asks the opponent to decide to give material back — the opposite of the natural reflex.

Should Black take both pawns?

He may, provided he gives them back later. The line where Black takes everything and trades off the c4 bishop with ...Be6 measures −1.00, the best in this entry for Black. What does not work is taking and trying to hold: every move spent defending material is a move the two white diagonals charge for.

Is it worth declining with 3...d3?

No. The move returns the pawn immediately and for nothing: White recaptures with the bishop, keeps the centre, and the position measures +0.87 — the only line in this entry where White is genuinely better. If the aim is to avoid complications, Capablanca's 3...d5 costs far less: +0.21.

Guides on this opening

• The Smith-Morra Gambit: a practical guide to facing the Sicilian a pawn down
• The Scotch Gambit: a complete guide to 4.Bc4, the open centre and the attack on f7

Common Mistakes & Where Most Amateurs Go Wrong

• 01 Retreating the knight to g8 after e5. The two lost tempi pay for exactly the two won pawns, and White keeps development, centre and both diagonals: +1.49. The line: 1.e4 e5 2.d4 exd4 3.c3 dxc3 4.Bc4 Nf6 5.e5 Ng8 6.Nxc3 d6 7.Nf3 Nc6 8.O-O Be6 9.Bxe6 fxe6 +1.49
• 02 Giving the pawn back with ...d5 before getting the king out of the centre. After 6.Bxf7+ Kxf7 7.Qxd8 Black loses both the castling rights and the queen, and the position measures +1.81. The line: 1.e4 e5 2.d4 exd4 3.c3 dxc3 4.Bc4 d5 5.Bxd5 Nf6 6.Nf3 Bb4+ 7.Nc3 O-O +1.81