Van Geet Opening: Sleipnir Gambit – How to Punish White as Black

After 1.Nc3 d5 2.e3 e5 3.d4 Bb4, White has a crucial decision. The engine's best move is dxe5, capturing the pawn and accepting the gambit. From there, the continuation dxe5 Ne7 Bd2 O-O leads to a position where White's extra pawn comes with development pressure — you'll need to fight for activity. But here's the key: most White players in the database do NOT play dxe5. Only 769 out of 4,861 games saw that move. The vast majority choose something else, and many of those alternatives are mistakes that hand you the advantage. Your job as Black is to recognise those subpar moves and punish them immediately.

The database labels three common responses as errors. Know them by heart so you can pounce when your opponent stumbles. Bd2 (played in 2,667 games) is a mistake that loses about 1.1 pawns in evaluation. White blocks the pin on the knight, but it costs valuable time — you can capture on c3 or simply continue developing with Ne7, keeping the pressure. Ne2 (267 games) is an even bigger mistake, losing approximately 1.2 pawns. White retreats the knight voluntarily, giving up the centre and letting your bishop on b4 breathe. a3 (583 games) is classified as an inaccuracy, losing roughly 0.8 pawns. This push attacks the bishop but gives you an easy decision: take on c3 or retreat to a5, keeping the pin alive. Against all three, your plan is straightforward — maintain the initiative and don't rush to give back the pawn you've already won.

With the engine showing +1.07 in White's favour, you might expect Black's winning chances to be dismal. But the Lichess database of 4,861 games tells a different story: White wins 48.3%, draws happen only 4.2% of the time, and Black wins 47.5%. Those numbers are almost dead even. Why? Because human players struggle to convert the computer's advantage. Most White players don't find the precise follow-up after dxe5, and the alternative moves (Bd2, a3, Ne2) are outright mistakes that flip the evaluation in your favour. If you study the key replies and learn to punish White's common errors, your practical results will far outpace what Stockfish predicts. The Sleipnir Gambit is a psychological weapon — White thinks they're outplaying you, but one inaccuracy on their part and you're the one pressing.

When White does find dxe5 — the engine's top choice — you need a solid response. The most-played continuation from the database is Ne7, developing the knight toward the centre while preparing to recapture on d5 or castle quickly. After White replies with Bd2 (developing and breaking the pin), you answer with O-O, getting your king to safety and connecting your rooks. At this point you have a pawn deficit but active pieces, a lead in development, and the bishop pair. Your plan: finish development (Bf5 or Bg4, Nbc6, Re8), put pressure on the e5 pawn, and look to open lines in the centre once your pieces are coordinated. Don't chase material — your compensation comes from activity and the fact that White's extra pawn is a weak target, not a strength.