Chessy
The English Opening: Two Knights e3 Variation — Black’s Guide
When White plays 1.c4 e5 2.Nc3 Nf6 3.e3, you might wonder: isn't e3 a little passive? It blocks in White's light-squared bishop. But the position is trickier than it looks. After your reply 3...Bb4, you pin the knight and fight for the centre. With over 295,000 games in the database, this line is well tested. What does the engine say? Stockfish rates the position +0.02 — dead level. That means you are neither better nor worse out of the opening. If you know what to do next, you can steer this into a comfortable middlegame.
Play the English Opening: King's English Variation, Two Knights Variation: e3 against the engine
Free, no signup — you play black, the engine adapts to your level.
Now that you know the key ideas, test yourself against the engine. Play the position as Black and see if you can hold the balance or punish White's inaccuracies
Create a free account →What Black Is Fighting For
Your move 3...Bb4 is a pin with purpose. By putting the bishop on b4, you stop White from playing d4 without kicking the knight first. White’s e3 move has weakened their control of d4, and you are punishing that immediately. The pin also prepares ...e4 ideas in some lines, and it makes it harder for White to develop their pieces fluidly. You are not trying to win the game in one move — you are competing for the dark squares and making sure White’s extra central pawn does not give them a free ride. The statistics back you up: Black wins 46.4% of games from here (with 3.7% draws), which is very respectable for the second player. White's 49.9% win rate is barely above yours.
The Engine’s Top Move and the Most Popular Replies
The engine's best continuation for White is 4.Nge2, with the follow-up ...0-0 a3 Bxc3. That line leads to a symmetrical-looking structure where Black has traded the bishop pair for fast development and a solid position. But Nge2 is not the only move you will face. Here are the most common White replies and how they score, so you know what to be ready for: - 4.Nge2 (58,229 games) — White scores 50.9%. You castle and prepare to meet a3 by taking on c3. - 4.d4 (51,869 games) — White scores 50.0%. This is the most direct approach; Black can take or push. - 4.a3 (48,989 games) — White scores 48.8%. This move is actually an inaccuracy (see below). - 4.Nf3 (33,737 games) — White scores 48.7%. Also an inaccuracy. - 4.Qc2 (28,384 games) — White scores 51.2%. Prepares to break the pin. - 4.Nd5 (17,805 games) — White scores 53.6%, the highest winning percentage for White. This aggressive jump can be annoying, but Black has good resources.
Punishing White’s Inaccuracies
Two of White's most popular replies are actually mistakes that give you a real advantage if you know how to handle them. According to the engine's analysis: - 4.a3 loses about 0.5 pawns (the engine prefers Nge2). After a3, you can retreat the bishop to a5 or e7 — or sometimes take on c3 and trust your pawn structure. Black scores 51.2% from this position, which means you win more often than White does. - 4.Nf3 loses about 0.9 pawns — nearly a full pawn! Black scores 51.3% here. After Nf3, you can push ...e4, gaining time and space. The knight has to move again, and suddenly White's setup looks clumsy. Keep an eye out for these two moves; they are your best chances to take over the game early.
The Typical Plan: Castle and Trade
In the main line after 4.Nge2, Black's most natural plan is to castle kingside quickly (0-0) and wait for White to play a3. When a3 comes, you take the knight on c3: ...Bxc3. This trade gives up the bishop pair, but it serves several purposes. First, it removes a key defender of White's centre. Second, it gives Black a half-open b-file after ...bxa3 or ...bxc3 in some lines. Third, it simplifies the position, which is often what Black wants in a symmetrical English. After the trade, White recaptures with the knight from e2 (Nxc3), and the game takes on a quiet, manoeuvring character. You can aim for ...d5, ...Re8, or even ...b6 and ...Bb7. The engine says this is completely equal — your job is to play accurate chess from here, not to force a win.
Results across 295,264 Lichess games
| Most-played continuation | Games | White wins |
|---|---|---|
| Nge2 | 58,229 | 50.9% |
| d4 | 51,869 | 50.0% |
| a3 | 48,989 | 48.8% |
| Nf3 | 33,737 | 48.7% |
| Qc2 | 28,384 | 51.2% |
| Nd5 | 17,805 | 53.6% |
Frequently asked questions
Why is 3...Bb4 a good move for Black in this line?
3...Bb4 pins the knight on c3, which indirectly attacks White's control of the centre. It also prevents White from playing d4 comfortably. Black is fighting for the dark squares and prepares quick castling. The statistics bear this out: Black wins 46.4% of games from this position, which is excellent for the second player.
What should Black do if White plays 4.a3?
4.a3 is actually an inaccuracy that loses roughly half a pawn. Black can retreat the bishop to a5 or e7, or sometimes capture on c3 straight away. Black scores 51.2% in this position, so you are already slightly favourite. Just make sure you do not drop the bishop for nothing — a3 is a waste of time for White.
Is the English Opening Two Knights Variation good for beginners?
Yes, it is a fine choice. The positions are generally quiet and strategic, with less immediate tactical fire than many 1.e4 openings. Black's plan is clear: develop, castle, and trade the bishop on c3 when White pushes a3. The position is dead equal (+0.02), so you are never worse out of the opening just by following the main line.
What is the most dangerous White move Black needs to be ready for?
4.Nd5 is the highest-scoring White move (53.6% for White). It attacks the knight on f6 and can be annoying. Black can respond with ...Nxd5, ...Bd6, or even ...0-0. Since it is rare (only 17,805 games), just know it exists and have a plan — do not panic.
How many games feature the English Opening: King's English Variation, Two Knights Variation: e3?
Over 295K Lichess games have reached the English Opening: King's English Variation, Two Knights Variation: e3 position. White wins 49.9%, Black wins 46.4%, with 3.7% draws — based on real rated games.