The Accelerated Dragon: Be3 – Why Black Already Has the Upper Hand

ECO B34 20,552 games Stockfish +0.43

Welcome to the Accelerated Dragon: Be3. After the moves 1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6 5.Be3 Nf6, you've reached the tabiya. Stockfish evaluates the position at +0.43, a small edge for White — meaning you are slightly worse according to the engine. But don't let that number fool you. Across 20,552 real games from this exact position, Black scores an impressive 53.4%, with White winning only 42.2% and draws at 4.4%. The practical chances are firmly in your favour. The drill below puts you in Black's shoes against a live engine — let's see why humans struggle to convert White's theoretical advantage.

Play the Accelerated Dragon: Be3 against the engine

Free, no signup — you play black, the engine adapts to your level.

The best way to learn this opening is to play it. Start the interactive drill below and practise the Accelerated Dragon: Be3 as Black against the engine — try 6

Create a free account →

What You're Fighting For: The Dark-Squared Bishop

In the Accelerated Dragon: Be3, your main trumps are piece activity and a comfortable kingside. Unlike the standard Dragon, you've kept ...d7-d6 flexible, meaning your dark-squared bishop on g7 has a clear diagonal to the centre and kingside. White's Be3 move is a natural developing move, but it also signals that White may be aiming for the dreaded Yugoslav Attack — though without an early Bc4 or f3, the pressure is less immediate. Your plan is straightforward: fianchetto your bishop (already done), castle short, and prepare ...d7-d6 (or sometimes ...d5 in one go if White allows it). The engine's best line shows White continuing with Nc3, developing naturally, followed by Bg7, Bc4, and Qa5 — and that queen sortie to a5 is your key counterplay, hitting the c3-knight and eyeing the queenside. You're fighting for active piece play, not passivity.

The Best Reply: Meet Nc3 with Confidence

White's most popular move in this position — and the engine's top choice — is Nc3, appearing in 13,004 of the 20,552 games in the database. It's a natural developing move, supporting the d4-knight and preparing to bring the bishop to c4 or e2. Your response according to the engine's best continuation is 6...Bg7, which completes your fianchetto. After White's likely 7.Bc4, you reply with 7...Qa5 — a powerful move that attacks the c3-knight, threatens ...Nxe4 if White isn't careful, and prepares to castle. White scores only 43.1% from this line, meaning you win more than half the time. The key idea: don't rush to play ...d6 — keeping the tension in the centre lets your queen on a5 create immediate threats. If White tries to chase your queen with Bd2, you can retreat to c7 or b6 with a solid position.

The Statistics: Where White Goes Wrong

The database reveals that White frequently mishandles this position. The most common deviation from Nc3 is f3 (3,748 games), which scores a mediocre 44.4% for White — slightly better than the main line but still favourable to you. After f3, Black can continue with ...Bg7 and ...0-0 normally, but the early f3 stops White from using that square for a knight and slightly weakens the kingside. The real gifts come when White plays one of the known inaccuracies. Bd3 (1,199 games, White scores only 37.5%) is an inaccuracy costing about 0.8 pawns — the bishop is passively placed and blocks the d2-pawn. Nxc6 (934 games, White scores 36.4%) is also an inaccuracy of about 0.8 pawns — trading your strong knight on c6 only helps you recapture with ...bxc6, opening lines on the queenside and giving you a solid centre. Nd2 (419 games, White scores 42.7%) loses about 0.6 pawns and misplaces the knight. Whenever White plays any of these suboptimal moves, your winning chances jump significantly.

The Critical Moment: Punishing White's Mistakes

When White plays an inaccuracy like Bd3 or Nxc6, you need to know how to capitalise. After 6.Bd3, the engine prefers ...Bg7 followed by ...0-0 — the bishop on d3 is less threatening than on c4, and you can consider an early ...d5 break. The statistics back this up: White wins only 37.5% of the time, so you're already the favourite. After 6.Nxc6, recapture with 6...bxc6 (not ...dxc6, which would open the d-file for White's queen). Your pawn duo on c6 and c7 controls the centre, your bishop on g7 becomes even more powerful, and you have an open b-file for your rook. These positions are comfortable and well-known to Accelerated Dragon players — the engine may call them equal-to-slightly-worse, but in practice, you'll outplay most opponents. Your task in the drill below is to spot these moments and seize the initiative.

Results across 20,552 Lichess games

42.2%
4.4%
53.4%
■ White 42.2% ■ Draw 4.4% ■ Black 53.4%
Most-played continuationGamesWhite wins
Nc313,00443.1%
f33,74844.4%
Bd31,19937.5%
Nxc693436.4%
Nd241942.7%
Bc425437.8%

Frequently asked questions

Is the Accelerated Dragon: Be3 a good opening for beginners?

Yes. The Accelerated Dragon: Be3 is a solid choice for club players. You develop naturally with ...g6, ...Bg7, and ...Nf6, and your plans are clear. The statistics show Black scores 53.4% in practice, and White's most common mistakes give you excellent winning chances without needing deep theory.

What is the difference between the Accelerated Dragon and the regular Dragon?

The key difference is move order. In the Accelerated Dragon, Black plays ...g6 before ...d7-d6, keeping the option of an early ...d5 break. This avoids some of White's most dangerous attacking setups in the Yugoslav Attack. The Be3 variation is a flexible approach for White, but you can still reach comfortable positions.

How should Black respond to 6.f3 in this line?

After 6.f3, simply continue with 6...Bg7, developing your bishop. White's f3 move is solid but not threatening — it takes away the f3-square from the knight and slightly weakens the kingside dark squares. You can castle, play ...d6, and proceed with your standard queenside counterplay. White scores only 44.4% from this line.

Why does Stockfish give +0.43 if Black wins more often?

Stockfish evaluates the position objectively, and +0.43 means White has a small theoretical edge. However, practical play is different — human players misplay these positions more often than engines. The 53.4% Black win rate in the database shows that White's edge is hard to convert, and Black's active piece play scores well in practice.