Chessy
The Blackburne Shilling Gambit: O-O – How to Play It (as Black)
You've just played 3...Nd4, hoping your opponent takes the bait. Instead of grabbing your knight, White ignores the threat and castles: 4.O-O. Now you have a choice to make — and the engine says your position is already under pressure. The Blackburne Shilling Gambit is a sharp, tricksy line, but White's 4.O-O is a cool-headed response. Let's see what the statistics say, what your best move is, and how to keep the game alive when your opponent doesn't fall for the trap. The interactive drill below lets you practise the key moment right now.
Play the Blackburne Shilling Gambit: O-O against the engine
Free, no signup — you play black, the engine adapts to your level.
Play through the first moves: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 4.O-O Nxf3+. Now practise both White's options below. If White plays Qxf3, try 5...Qf6. If White plays
Create a free account →The Main Idea: What Are You Fighting For?
In the Blackburne Shilling Gambit, you (as Black) sacrifice a knight on d4 for a pawn and attack. Your dream scenario is 4.Nxd4? exd4, when you regain the piece with interest and White's king is stuck in the centre. But 4.O-O neutralises that trap. You're not just giving away material — after 4...Nxf3+, you're hoping for the recapture gxf3, which opens the g-file against White's king. That recapture is a known mistake, but it's how you keep the attack alive. If White recaptures with the queen instead, your plan shifts to solid development and quick castling. The fight is about activity versus soundness.
The Engine Verdict: White Is Better
Stockfish rates this position at +1.03, a clear edge for White. That means you are clearly worse here. The gambit hasn't refuted anything — White has a comfortable advantage with best play. Across 351,100 games on Lichess, White wins 53.9%, draws 4.3%, and Black wins 41.8%. Those are not terrible odds for Black at amateur level, but against strong opposition you're playing for practical chances, not a forced win. Don't expect cheap wins; expect a fight where you're slightly behind from the start.
Your Best Move: 4...Nxf3+ and What Follows
After 4.O-O, you must play 4...Nxf3+ — there's no other sensible move (5...Qf6? loses more material and abandons the attack). Now White's best is Qxf3 (played in 349,120 games, White scores 54.0%). From there, the engine suggests: 5.Qxf3 Qf6 6.Qg3 d6. Black prepares ...Be6 or ...O-O-O and aims to develop smoothly. If White recaptures with gxf3 instead (1,869 games), White scores only 30.1% — a big drop. That's because Black gets attacking chances on the g-file (...Qg5, ...h5, ...Bg7, ...O-O soon). So your job is to make gxf3 tempting for White. If you get it, you're the one pressing. If you face Qxf3, you're in a standard but slightly worse position — play solidly and outplay your opponent later.
Known White Mistakes You Should Exploit
Two replies from White are bad news for them and good news for you. gxf3 is a mistake (loses about 2.3 pawns by engine reckoning; better was Qxf3). After gxf3, Black has a strong initiative. Your plan: ...Qf6 (attacking f2), ...h5, ...Bg7, ...O-O, and pile up on the kingside. Kh1 is even worse — a blunder losing about 3.8 pawns. After 4...Nxf3+ 5.Kh1??, Black has a crushing attack: ...Qh4 with ideas of ...Ng4 and ...Qxh2#, or simply ...Nd4 and ...Nf2+. If you ever see Kh1, you're winning. In real play, most opponents will find Qxf3, so don't bank on a blunder every game.
Your Strategy Against Qxf3
In the main line (5.Qxf3), Black's plan is straightforward: develop your queen to f6 (attacking f2 and preparing to castle), then retreat the queen to g3 with the idea of ...d6, ...Be6, ...O-O-O, and a solid middlegame. The engine's suggestion is 5...Qf6 6.Qg3 d6. White will likely play d3, Nc3, or even h3 to stop ...Bg4. You should aim to castle queenside and challenge the centre with ...d5 later if possible. It's a quiet line with no immediate tactics — you're playing for slow pressure. The gambit hasn't won outright, but you've avoided the worst (gxf3 for White actually helps you). Keep your king safe, develop, and trust your activity to compensate for the small material deficit.
Results across 351,100 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| Qxf3 | 349,120 | 54.0% |
| gxf3 | 1,869 | 30.1% |
| Kh1 | 111 | 14.4% |
Frequently asked questions
Is the Blackburne Shilling Gambit sound after 4.O-O?
No, it's not sound. Stockfish gives White +1.03, meaning White is clearly better with best play. However, the gambit can score well at club level if White makes one of the known mistakes (gxf3 or Kh1). In the main line after Qxf3, Black has a slightly worse but playable position. Don't expect a forced win — expect a fight.
What is the difference between Qxf3 and gxf3 for White?
Qxf3 is White's best move, leaving them with a +1.03 advantage and no immediate attack for Black. gxf3 is a mistake that loses about 2.3 pawns of advantage, giving Black a strong initiative on the g-file. After gxf3, Black can play ...Qf6 (threatening f2), ...h5, and ...Bg7 with excellent attacking chances. White scores only 30.1% after gxf3 across 1,869 games.
What should Black play if White recaptures with gxf3?
Celebrate — you got what you wanted. Play ...Qf6 (attacking f2), then ...h5, ...Bg7, ...O-O, and build up on the kingside. Your immediate threats include ...Qxf2+ (if allowed) and later ...Qg6 with ideas like ...h4 and ...Bh6. White's king is unsafe on g1. The engine says this is a mistake by White, so you have a clear edge.
What is the most common mistake White makes here?
By far the most common mistake (and the one you hope for) is gxf3. It happens in over 1,800 games out of 351,000 — not super common, but it's a real trap. The blunder Kh1 is rarer (111 games) but even more catastrophic for White. In the vast majority of games, White plays Qxf3 and Black must defend carefully.
How many games feature the Blackburne Shilling Gambit: O-O?
Over 351K Lichess games have reached the Blackburne Shilling Gambit: O-O position. White wins 53.9%, Black wins 41.8%, with 4.3% draws — based on real rated games.