Source Lichess
White wins at least 11.5 more of every 100 games than it loses
how far wins run ahead of losses, 90% confident
At All Ratings, French Defense: Advance Variation, Milner-Barry Gambit, Sørensen Variation keeps White's wins running +11.5% ahead of its losses at worst, 90% confident, from 2.0K Lichess games. This gambit is effective. Win rate, draw rate and loss rate come from games played December 2023 to November 2025.
| Rating | Edge at worst | White Win | Draw | White Loss | Games |
|---|---|---|---|---|---|
| 1400-1599 | −22.3% | 47.2% | 4.0% | 48.8% | 61 |
| 1600-1799 | +9.8% | 58.7% | 3.7% | 37.6% | 202 |
| 1800-1999 | +11.9% | 57.3% | 3.1% | 39.5% | 751 |
| 2000-2199 | +11.9% | 58.0% | 2.6% | 39.4% | 586 |
| 2200-2499 | −5.6% | 51.4% | 3.3% | 45.3% | 194 |
| 2500+ | −6.1% | 52.4% | 2.2% | 45.4% | 156 |
At some ratings. White comes out ahead at 1600-2199; at the other ratings we can see, the games do not show it paying. Where it pays changes with the clock, so set yours above and look again.
White wins 56.1% of French Defense: Advance Variation, Milner-Barry Gambit, Sørensen Variation games, draws 2.9% and loses 41.0%. That rests on 2.0K games from Lichess.
1800-1999. There White wins at least 11.9 more of every 100 games than it loses (90% confident), its highest Edge at worst.
Blitz. For White, the Edge at worst — how far wins run ahead of losses, 90% confident — is +8.9% in Blitz (1.2K games) and −7.4% in Rapid (142 games).
Yes. The first rating where the games do not show White coming out ahead is 2200-2499 in Bullet and 2200-2499 in Blitz. In Rapid, only 1800-1999 has enough games, and there the games do not show White coming out ahead.
The French Defense: Advance Variation, Milner-Barry Gambit, Sørensen Variation is extremely sharp: 2.9% of its games are drawn, so 97.1% end with a winner.
At 1600-2199, where White comes out ahead. At the other ratings we can see, the games do not show it paying. Where it pays changes with the clock, so set yours above and look again.
As Black, take the material, then develop fast, castle early, and trade pieces. Give the material back when doing so kills the attack. At 1600-2199, White still comes out ahead in these games; at the other ratings we can see, the games do not show the gambit paying.
This rests on 2.0K games from Lichess. Every figure carries the range it is in, 90% confident, and the table above shows how many games sit behind each rating and each clock — the fewer there are, the wider that range.