Queen's Gambit Declined: Normal Defense Nf3 — Playing Black

By playing 4...c5 before developing your dark-squared bishop, you challenge White's centre on your own terms. The pawn on d4 comes under fire, and White must decide how to handle the tension. This line avoids the more passive setups Black sometimes adopts in the Queen's Gambit Declined, and it often leads to an IQP (Isolated Queen's Pawn) position if White captures on d5 — a middlegame structure that gives Black clear counterplay on the c-file and active piece play. The statistics back this up: across nearly 1.4 million games, Black scores a very respectable 46.4%, with White winning 48.9% and draws at 4.7%. For a sharp opening, that's excellent practical odds.

Stockfish's favourite move for White in this position is 5.cxd5, leading to the forcing sequence 5...cxd4 6.Qxd4 exd5. The engine evaluates this as +0.26, a small edge for White. That means you, as Black, are slightly worse but very much in the game. The resulting pawn structure is symmetrical in the centre, and Black's development is smooth — the queen on d4 can be a target for moves like Nc6 or Be6. This line has been played 289,417 times in the database, and White's winning percentage is actually lower here (48.9%) than in several other continuations. So if White chooses the engine's best move, you are still fine.

White's most common choice is 5.Bg5, appearing in 474,424 games. It pins the knight on f6 and threatens doubled pawns, but White only scores 49.4% — essentially a dead-even result. The pin is annoying but not crushing. The second-most popular is 5.e3 (353,316 games), also scoring 49.4% for White. This more solid approach leads to quieter play. The rarest major option is 5.g3, played in 19,799 games, and White's score suddenly jumps to 52.6% — but here's the catch: the engine calls 5.g3 a mistake, losing about 1.2 pawns in evaluation compared to the best move. That 52.6% score may come from weaker players not knowing how to punish it. If you see 5.g3, you can seize the advantage. The worst-performing White move statistically is 5.dxc5 (46.6% for White), so capturing on c5 is actually favourable for Black in practice.

When White plays 5.g3, they are preparing to fianchetto the bishop while the centre is still unresolved. But this wastes a tempo and leaves the d4-pawn vulnerable. The engine says the correct reply was 5.cxd5, and that 5.g3 is about 1.2 pawns worse. Your task is to pile pressure on d4: after 5.g3, you can consider 5...cxd4 (opening the centre) followed by developing with tempo, or simply 5...Nc6 increasing the pressure on d4. White's king is still in the middle and their pieces aren't coordinated yet. In the drill below, you'll face 5.g3 and get to practise the best response — turn White's mistake into your advantage.