The Exchange Variation of the French Defense arises after 1. 2. 3.: you trade off your e-pawn, the e-file opens, and if your opponent recaptures with the pawns are symmetrical. As White you score 51.1% across 286,992 Lichess games between 900 and 1700 (September 2026 data). Strong players score almost exactly the same, online and over the board. Your edge out of the opening is small, and the game is decided by how you place your pieces.
Lichess games from September 2026 (30,577,162 rated 900 to 1700) and 31,791 over-the-board games rated 2300+ from the last 12 months. Data updated: October 14, 2026.
4 named lines in this family; 1 with their own page.
The opening's share in each Lichess rating band.
After 3., your opponent recaptures with in 88 of every 100 games at your level and 98 among strong players. The other common reply is , in 10 of every 100, and against it you score 55 in 100: 4. hits the queen and gains you a tempo. After , the most popular move in both groups is 4.: 37 in 100 at your level, 43 among strong players. The gap shows up elsewhere. Strong players choose 4. 30 times in 100, the public just 8, while 4. climbs to 28 in 100 at your level and drops to 3 among strong players. Steer clear of 4.: you score 46 in 100 with it.
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 87.9% | 50.7% | 98.3% | 100.0% | |
| 10.2% | 55.4% | 1.3% | no data | |
| 1.0% | 52.8% | 0.3% | no data | |
| 0.6% | 56.6% | no data | no data | |
| 0.1% | 48.8% | no data | no data | |
| 0.1% | 60.8% | no data | no data | |
| no data | no data | 0.1% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 36.7% | 51.0% | 43.2% | 82.8% | |
| 28.5% | 50.1% | 3.2% | no data | |
| 9.3% | 53.2% | 15.6% | no data | |
| 8.0% | 52.9% | 29.8% | 11.1% | |
| 6.1% | 46.0% | no data | no data | |
| 2.7% | 50.3% | 1.1% | no data | |
| no data | no data | 4.2% | 6.1% |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 51.8% | 50.3% | 43.3% | 51.9% | |
| 10.0% | 52.6% | 1.7% | no data | |
| 9.2% | 49.7% | 24.2% | 18.2% | |
| 8.8% | 52.6% | 20.9% | 29.9% | |
| 5.6% | 51.7% | 3.7% | no data | |
| 4.6% | 51.2% | 4.2% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 31.7% | 50.3% | no data | no data | |
| 19.4% | 52.0% | 69.7% | 100.0% | |
| 14.3% | 49.7% | 2.6% | no data | |
| 8.3% | 49.9% | 4.6% | no data | |
| 7.5% | 51.9% | 11.0% | no data | |
| 4.8% | 47.9% | no data | no data | |
| no data | no data | 6.6% | no data | |
| no data | no data | 2.5% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 36.6% | 50.6% | 56.6% | 80.9% | |
| 19.3% | 54.4% | 14.3% | no data | |
| 12.6% | 51.2% | 18.4% | 19.1% | |
| 9.3% | 51.4% | 6.5% | no data | |
| 7.2% | 53.6% | 1.5% | no data | |
| 4.0% | 52.9% | 1.7% | no data |
The strong players' line has a clear aim: get your king off the open e-file quickly, then lean on the f6 knight, the piece holding your opponent's castled position together. The sequence runs 4. 5. 6. 7..
After the structure is symmetrical: your d4 pawn faces the one on d5 and the e-file has no pawns on it. Neither side has a pawn break ready, so what counts is who places their pieces better. Your king is still on e1, right on the open file, and getting it out of there comes first.
4. clears g1 and covers e5, the central square on the open file. 5. clears f1 and puts the bishop on the diagonal that runs through e4, f5 and g6 to h7. With g1 and f1 empty, 6. comes on move six. That's why 4. doesn't fit: a queen on e2 blocks the f1 bishop, which can no longer reach d3, and your short castling gets delayed.
After comes the biggest gap with your level: strong players pick 5. 70 times in 100, the public just 19. At your level the most popular move is 5., in 32 of every 100 games. It hits d5, but that pawn is already guarded by the f6 knight and the d8 queen, so it wins nothing and leaves the f1 bishop at home. With 5. you score 50 in 100; with 5., 52.
Your opponent usually mirrors your setup with and ; among strong players, Black castles short after 6. almost every time. Then 7. develops your c1 bishop and pins the f6 knight to the queen on d8: if the knight moves, your bishop takes the queen. The pin matters because that knight has two jobs. It guards h7 alongside the king, the square your d3 bishop is aiming at, and it's the piece defending d5, since the d6 bishop cuts the queen off from the d-file. While the pin lasts, that knight stays tied to both jobs.
If your opponent steps off the main road with , which comes in 10 of every 100 games at your level after 4., answer 5.: the bishop comes out with check before that pawn can advance to c4 and take d3 away from it. In public games the most common pattern is both sides castling short, so expect both kings to end up on the kingside, on g1 and g8.
| Move | Strong games | Chosen by |
|---|---|---|
| 9,009 | 98.4% | |
| 9,079 | 43.6% | |
| 6,750 | 43.4% | |
| 3,105 | 70.2% | |
| 2,385 | 57.1% | |
| 1,421 | 69.7% | |
| 971 | 99.5% | |
| 891 | 43.0% |
After 4. , 5. allows , a pawn that covers d3 and takes that square away from your f1 bishop. After 6. 7. 8., it comes out on g2 instead. Better is 5., with check before the pawn arrives. At your level it rarely gets punished: you score 56 in 100, against 59.
30 in 100 play it in this position. It gives away 1.2 pawns of advantage. It scores 56.3%; scores 59.3%.
After 4. , 5. props up d4, but spends a tempo that 5. uses to develop with check. After 6. 7. 8. the c3 pawn supports nothing, though your opponent is left with an isolated d5 pawn. It's an inaccuracy you'd barely notice: 54 in 100 with either move.
15 in 100 play it in this position. It gives away 0.7 pawns of advantage. It scores 53.5%; scores 53.7%.
After 4. , 5. guards d4 with the bishop but again allows , which takes d3 away from your f1 bishop. After 6. 7. 8., your knight develops via e2 and the f1 bishop has to settle for g2. Better is 5.. At your level it's rarely punished: you score 57 in 100, against 59.
5 in 100 play it in this position. It gives away 1.1 pawns of advantage. It scores 57.1%; scores 59.3%.
After 4. , 5. attacks the e6 pawn but wins nothing: after 6. 7. 8. the queens come off, your pawn sits on c6, attacked by the b7 pawn and the b8 knight, and your knight is stuck on d1. This one does cost you: you score 51 in 100, against 60 with 5..
3 in 100 play it in this position. It gives away 0.9 pawns of advantage. It scores 50.8%; scores 60.0%.
After 4. , 5. puts the bishop on a square where it attacks nothing at all, whereas 5. develops it with check. After 6. 7. 8. your opponent has both knights out and the centre open. It's a small inaccuracy that rarely gets punished at your level: you score 52 in 100, against 54 with 5..
3 in 100 play it in this position. It gives away 0.6 pawns of advantage. It scores 51.8%; scores 53.7%.
After 4. , 5. gives check, but the queen on e2 blocks your f1 bishop. Your opponent answers with , developing as they block, and after 6. 7. 8. you castle long, with the f1 bishop still at home. At your level it's rarely punished: you score 58 in 100, against 59 with 5..
3 in 100 play it in this position. It gives away 1.2 pawns of advantage. It scores 58.2%; scores 59.3%.
| White | Black | Games |
|---|---|---|
| 178,034 | ||
| 34,454 | ||
| 18,401 | ||
| Not castled | 16,749 | |
| Not castled | 14,878 | |
| Not castled | Not castled | 11,365 |
| White | Black | Games |
|---|---|---|
| 89 | ||
| Not castled | 13 | |
| 7 | ||
| 4 | ||
| 3 | ||
| Not castled | Not castled | 2 |
1. 2. 3. 4. · 13,957 amateur games · 11 over the board
4. hits the d5 pawn. It comes up in 9 of every 100 games, and you score 53.
1. 2. 3. 4. 5. · 10,084 amateur games · 2 over the board
A different move order, 3. before trading, then 5. pins the f6 knight.
1. 2. 3. 4. 5. · C01
264 amateur games · White's score: 60.0%