Grob Opening: Alessi Gambit – How to Play as Black

By playing 1...f5, you immediately put White's g-pawn under attack and stake a claim in the centre. If White captures with 2.gxf5 (the engine's best and most popular move), you'll continue with 2...d5, preparing to develop naturally with e6 and Bxf5, or even push e5 later. The pawn on f5 is gone, but Black gets a solid central presence and easy development. The key idea: you're not trying to refute 1.g4 — you're making White prove their compensation for the weird opening. If White plays anything other than gxf5, you can often seize an advantage.

White has several options here, and the engine is clear: only gxf5 keeps the balance. The most common alternatives are all punished by the statistics. Bg2, the second-most popular move, is actually an inaccuracy — it loses about 1.0 pawns of advantage. Even worse, g5 and h3 are both mistakes (losing ~1.0 and ~1.4 pawns respectively). If you see White play Bg2, g5, or h3, you can be confident that you're already doing well. The engine says White's best is gxf5, which still leaves you with a clear, lasting advantage in White's favour — meaning you are clearly worse but have a fully playable position with plenty of counterplay.

Across 23,279 games that reached this position, the results are surprisingly close for an opening that starts with 1.g4. White wins 49.8%, Black wins 43.3%, with 6.9% draws. Black's 43.3% is respectable — compare that to Black's normal share in many mainstream openings. When White plays gxf5 (11,603 games), they score 51.1%, which is solid but not crushing. The mistakes really stand out: after g5 (3,593 games), White scores 53.2% but that move is a mistake per the engine. And after f4 (274 games), White scores a miserable 26.3% — a clear sign that pushing the f-pawn here is terrible. The lesson: trust the engine's evaluation, not just the raw win rates.

If White plays Bg2 (the inaccuracy), your plan is simple: capture on g4 with ...fxg4, attacking the bishop and opening lines. White's bishop on g2 looks active but actually leaves the kingside vulnerable. If White plays g5 (the mistake), you shouldn't retreat — instead, push d5 and prepare e5, gaining space and locking White's bishop out of the game. Against h3, consider ...fxg4 followed by ...Nf6 and ...g6, exploiting the light squares around White's king. In all these lines, remember that while you are objectively worse after 1.g4 f5, White has to navigate a minefield of inaccurate moves. Stay solid, develop your pieces, and don't be afraid to open the position.