Chessy
Van Geet Opening: Nowokunski Gambit
You've pushed your f-pawn, Black captured it, and now you've answered with 3.e4 — the Van Geet Opening: Nowokunski Gambit. You've gambitted a pawn to seize the centre and gain fast development, but Stockfish evaluates this position at -0.98, a clear edge for White's opponent. That means you are clearly worse in engine terms, so you'll need accurate play to make Black's life difficult. The interactive drill below will sharpen your responses to Black's most popular tries, especially the tricky Qh4+ and the quiet fxe3. Let's dig into what the statistics and the engine recommend.
Play the Van Geet Opening: Nowokunski Gambit against the engine
Free, no signup — you play white, the engine adapts to your level.
Practice these lines now in the interactive drill below — focus on punishing 3...Qh4+ and 3...Nf6, and learn the best setup against the critical 3...fxe3.
Create a free account →What the Nowokunski Gambit Is Fighting For
After 1.Nc3 e5 2.f4 exf4 3.e4, White has staked a bold claim in the centre. Your pawn on e4 controls d5 and f5, your knight on c3 eyes the centre, and you're ready to develop quickly with d4, Nf3, and Bc4 or Bb5. The catch? You're down a pawn on f4, and Black has the extra tempo of having already captured. Over 524 games from this position, White wins 45.4%, draws 2.5%, and Black wins 52.1% — so while the engine favours Black, the practical chances are closer than you might think. Your goal is to turn your lead in development and central space into an attack before Black can consolidate the extra pawn.
Black's Best Move: fxe3 (The Critical Line)
The engine's top choice for Black is 3...fxe3 — capturing the other pawn. This continuation has been played 221 times in the database, and White scores 47.5% from here — the best White result against any common reply. After 3...fxe3, the engine line runs: 4.fxe3 d4 5.Nf6 Bd3. White opens the f-file for the rook, and Black immediately pushes d4 to challenge the knight. You develop the bishop to d3, eyeing the kingside. White's compensation is a lead in development and a semi-open f-file — you want to castle quickly and start putting pressure on f7. This is the main test of the gambit.
The Tricky Check: 3...Qh4+ and Why It's an Inaccuracy
A natural reaction is 3...Qh4+, checking the king and threatening ...Qxe4. This has been played 110 times, but it's a trap for Black: Stockfish calls it an inaccuracy, losing roughly 0.6 pawns compared to 3...fxe3. White scores only 35.5% after this move, which means many White players mishandle it. The key is to block with g3! — Black's queen has to move again on a wasted tempo, and White's king finds a safe square on g2 after ...fxg3 (which isn't forced). Meanwhile, White develops with d4, Nf3, and Bg2, using the queen's misplaced position to gain time. If you've never practised this reply, the drill will help you punish 3...Qh4+ with confidence.
Common Mistakes Black Makes (and How to Exploit Them)
The statistics flag three suboptimal replies from Black, and knowing them will boost your results. Besides 3...Qh4+ (inaccuracy), 3...Nc6 is also an inaccuracy, losing roughly 0.6 pawns. Played 35 times, White scores 45.7% against it — a decent result. You can respond with d4, Nf3, and develop naturally, keeping the centre strong. More punishing is 3...Nf6, classified as a full mistake (losing roughly 1.6 pawns). After 3...Nf6, White scores 48.6%. Developing the knight to f6 lets White push e5 with tempo (gaining space and kicking the knight) or simply play d4 and Bc4 with a big edge. Interestingly, 3...Bc5 yields White's best score at 63.3% (over 30 games) — the engine might not flag it as badly as the others, but in practice White does very well. The pattern is clear: when Black doesn't grab the second pawn on e3, White's central presence becomes hard to handle.
Results across 524 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| fxe3 | 221 | 47.5% |
| Qh4+ | 110 | 35.5% |
| Nc6 | 35 | 45.7% |
| Nf6 | 35 | 48.6% |
| Bc5 | 30 | 63.3% |
| Bb4 | 24 | 37.5% |
Frequently asked questions
Is the Nowokunski Gambit sound for White?
Stockfish evaluates the position after 3.e4 at -0.98, a clear edge for Black, so it's not objectively sound at high levels. That said, Black has to know what they're doing — in practice White scores 45.4% in the Lichess database, and several common Black moves are inaccuracies or mistakes that White can punish.
How do I respond to 3...Qh4+ in the Nowokunski Gambit?
Meet the check with g3! Black's queen then has to move again, losing a tempo. White follows up with d4 and Nf3, chasing the queen further. The statistics show 3...Qh4+ is an inaccuracy that loses roughly 0.6 pawns compared to the best move fxe3, so you should be happy to see it.
What is Black's best move after 1.Nc3 e5 2.f4 exf4 3.e4?
The engine recommends 3...fxe3, capturing the second pawn. The main line continues 4.fxe3 d4 Nf6 Bd3. From there White has development and the open f-file as compensation for the pawn deficit. In the database White scores 47.5% after fxe3 — the highest White win percentage among common replies.
Why do I score better against 3...Bc5 than against 3...fxe3?
White scores a remarkable 63.3% against 3...Bc5, but only 47.5% against 3...fxe3. That's likely because Black is not fighting for the centre immediately. After 3...Bc5, White can play d4 with tempo and develop rapidly, while after fxe3 Black takes firm control of the extra pawn. The engine still slightly prefers fxe3 for Black, but at club level Bc5 gives White excellent practical chances.