Chessy
French Defense: King's Indian Attack with 2...d5 — Playing as White
You've chosen a cunning setup against the French. By opening 1.e4 e6 2.d3 d5 3.Nf3, you side-step the main lines and invite Black into a patient, strategic struggle. Stockfish evaluates the resulting position at +0.11 — dead level, as close to equal as chess gets. That means neither side has a tangible advantage yet, and everything depends on what happens next. The statistics, however, tell a sobering story: across nearly 850,000 games, Black actually wins 54.6% of the time, while White scores just 40.8%. Your task is to understand why those numbers look that way — and how to tip the scales back in your favour starting with the critical first decision. Jump into the drill below to play the position yourself.
Play the French Defense: King's Indian Attack: d5 against the engine
Free, no signup — you play white, the engine adapts to your level.
results: [{ id: 0, overall_0to100: 85 }, { id: 1, overall_0to100: 70 }, { id: 2, overall_0to100: 55 }, { id: 3, overall_0to100: 65 }, { id: 4, overall_0to100: 0
Create a free account →The Critical Moment After 3.Nf3
The position after 1.e4 e6 2.d3 d5 3.Nf3 is a tabiya for the King's Indian Attack. Black has several reasonable responses, but the engine's clear favourite is dxe4 — the move that immediately challenges your pawn centre. After 4.dxe4 Qxd1+ 5.Kxd1, White's king has lost castling rights but the position stays roughly equal at +0.11. This exchange line is by far the most popular, appearing in over 404,000 games. Your most important job here is to stay calm after the queens come off; the resulting endgame is playable but demands careful piece play. Many White players feel uneasy without the king safely castled, but the engine considers this line completely fine — you just need to develop actively and not let Black's extra central pawn mass become a problem.
What the Numbers Reveal About Each Black Reply
Here's how White scores against Black's five most common continuations across real Lichess games, along with the evaluation where available: - dxe4 (404,357 games): White scores 38.8%. The most challenging response. - c5 (196,106 games): White scores 41.9%. A flexible Sicilian-style setup. - Nf6 (102,559 games): White scores 40.3%. Natural development, inviting 4.e5. - d4 (38,669 games): White scores 45.5%. An inaccuracy that loses about half a pawn — Black should have captured with dxe4 instead. - Nc6 (35,356 games): White scores 43.6%. Also an inaccuracy; again dxe4 was better. - c6 (19,498 games): White scores 45.1%. A solid, slow approach. The pattern is clear: the higher the score for White, the less punishing the Black move is. When Black pushes d4 prematurely or develops the knight to c6, you get a small edge — those are the moments to seize.
Punishing Black's Known Mistakes
According to the data, two Black moves stand out as concrete inaccuracies: 3...d4 and 3...Nc6. Both lose roughly half a pawn compared to the correct move dxe4. If your opponent plays 3...d4, they've blocked the centre prematurely, giving you a spatial advantage and a clear plan: aim for c3 to challenge their pawn, or prepare e5 followed by a kingside attack. If they play 3...Nc6, they develop a piece that doesn't immediately pressure your centre. In both cases, the engine says you emerge slightly better — your task is to convert that small plus into a lasting edge. Keep your development flowing, don't rush, and look to exploit the extra space you gain from Black's passive choices.
A Practical Plan Against the Main Line (Black Captures on e4)
When Black plays 3...dxe4 — the engine's best move and by far the most common reply — you should recapture with 4.dxe4. Black will almost always trade queens with 4...Qxd1+ 5.Kxd1. Now you're in an endgame a pawn up in the centre but without castling rights. Here are the key ideas: - Develop quickly. Bring your bishop to e3 or g5, knight to c3 or d2, and aim to connect your rooks. - King safety matters. Your king is already on d1 after the queen trade, but you can often move it to e2 or even castle-by-hand later. - Don't fear the endgame. White's central pawn on e4 controls key squares, and Black's pawn on e6 can become a target. The statistics show White scores only 38.8% here, but that includes players at all levels. With a bit of endgame know-how, you can outperform those numbers.
Results across 849,638 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| dxe4 | 404,357 | 38.8% |
| c5 | 196,106 | 41.9% |
| Nf6 | 102,559 | 40.3% |
| d4 | 38,669 | 45.5% |
| Nc6 | 35,356 | 43.6% |
| c6 | 19,498 | 45.1% |
Frequently asked questions
Is the French Defense King's Indian Attack d5 good for White?
The position after 3.Nf3 is dead equal according to Stockfish (+0.11). However, the practical results across 849,638 games favour Black, who wins 54.6% against White's 40.8%. The opening is perfectly sound, but you need to know the key plans — especially in the queenless endgame after 3...dxe4 — to score well.
What should White do after 3...dxe4?
Recapture with 4.dxe4. Black will likely play 4...Qxd1+ 5.Kxd1, leading to an endgame where White has a central pawn on e4 and the king on d1. Develop your pieces actively (Be3, Nc3, Rd1) and don't worry about losing castling rights — the position is manageable with accurate play.
Which Black moves are mistakes in this position?
Both 3...d4 and 3...Nc6 are inaccuracies that lose about half a pawn, according to the engine. The best move is 3...dxe4. If Black pushes d4, you gain space and a plan to attack the centre. If they play Nc6, they develop without pressuring your pawn structure.
How can White improve their winning chances in this line?
Focus on the endgame after the queen trade — that's where most games are decided. Develop efficiently, target Black's e6 pawn, and avoid passive play. The statistics show White wins only 38.8% after 3...dxe4, but many players mishandle the resulting positions. Study simple king-and-pawn endgame principles to gain an edge.