Sicilian Defense: O'Kelly Variation, Venice System, Ljubojevic Line

The O'Kelly Variation (2...a6) is a flexible Sicilian. You stop White's pieces from using b5 (no Bb5 pin, no Nb5) and prepare ...b5 to gain space on the queenside. With 3.c3, White signals a slower, Alapin-style setup — they'll try to build a strong centre with d4. Your 3...b5 grabs queenside space and eyes the c4 square. The resulting structure is asymmetrical: White targets the centre, you claim the b-file and prepare ...Bb7, putting pressure on e4. This is not a passive defence — you're ready to fight for the initiative. The price is that White's centre can become very powerful if you don't keep them honest. That's the core tension in this line.

White's best move here is d4 — the engine's top choice and by far the most common at 39,064 games. After 4.d4, the typical continuation is ...e6, then Bd3 and ...Bb7. White builds a classical centre: pawns on d4 and e4, bishop on d3 aimed at your kingside. Your job is to challenge that centre with moves like ...cxd4, ...d5, or prepare ...Nc6 and ...Qc7. Notice that your b5-pawn supports a later ...c4 push, gaining space and kicking the Bd3. The engine line shows that Black's bishop on b7 opposes White's centre — that's your long-term pressure. Play the drill below to feel out the right response to 4.d4.

One of the most useful things about this position is that several of White's natural alternatives to d4 are actually inaccuracies that give you a better position. Here are the ones to watch for, based on the Lichess database of over 46,000 games: - d3 (1,048 games): An inaccuracy that loses White about 0.5 pawns. White's centre is less threatening, and you can play ...d5 or ...Nc6 with comfort. White scores only 50.5% here — basically nothing. - Bd3 (1,000 games): An inaccuracy losing ~0.6 pawns. This looks natural but White should have played d4 first. Wait for them to waste a tempo and pounce. - a3 (800 games): The worst of the bunch, losing ~0.9 pawns. White stops ...b4 but neglects the centre completely. Black scores a strong 51.4% here. If you see a3, think about ...d5 immediately — you're already better. - a4 (1,287 games): Not listed as a mistake, but White scores only 53.9% — lower than with 4.d4. It's playable but not scary. The takeaway: if White doesn't play 4.d4, you've likely already outplayed them in the opening.

From the full database of 46,904 games at this exact position, the overall results tell a clear story. White wins 51.5% of games, Black wins 44.7%, and draws are rare at just 3.7%. That 44.7% is a healthy number for Black in an opening where the engine says you're worse by +0.88. Why the difference? Because club-level White players often choose one of the inaccurate moves (d3, Bd3, a3) instead of the best move d4. When they do, Black's winning chances jump significantly. The Ljubojevic Line rewards patience: survive White's centre push, and you'll often outplay them in the middlegame. Your goal in the drill is to become comfortable against 4.d4 so that when your opponent chooses anything else, you're ready to take over.