The Tarrasch Variation is your 3. answer to the French Defense. Across 10,920 Lichess games between players rated 900 to 1700 (data from September 2026), you score 53.2% with it. The knight goes to d2 instead of c3, so your c-pawn stays free to support d4 from c3. Your opponent's most common reply is , which hits d4 straight away and makes you decide early how you will hold the center.
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.
16 named lines in this family; 1 with their own page.
The opening's share in each Lichess rating band.
At your level, after 3. your opponent picks in 4 of every 10 games, in 23 of every 100 and in 22 of every 100. Among online players rated over 2200, also leads, with 33 of every 100, followed by with 27. The big gap is in your fourth move. After , players at your level prefer 4., in more than 42 of every 100 games. Strong players choose 4. in 48 of every 100, while 4. stays under 7 of every 100 for them and scores just 44 of every 100. If your opponent plays instead, check the mistakes below: all of them come from that position.
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 39.9% | 50.5% | 32.9% | 47.3% | |
| 23.4% | 55.3% | 16.8% | 6.7% | |
| 22.3% | 51.9% | 27.0% | 20.1% | |
| 4.9% | 58.8% | 8.6% | 7.1% | |
| 2.9% | 58.4% | no data | no data | |
| 2.2% | 52.8% | no data | no data | |
| no data | no data | 5.8% | 9.2% | |
| no data | no data | 4.1% | no data | |
| no data | no data | no data | 5.9% |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 42.5% | 49.9% | 6.5% | no data | |
| 27.0% | 53.0% | 40.1% | 56.4% | |
| 22.2% | 50.9% | 48.2% | 43.6% | |
| 4.5% | 43.1% | 4.3% | no data | |
| 1.7% | 51.4% | 0.5% | no data | |
| 0.8% | 53.1% | no data | no data | |
| no data | no data | 0.3% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 76.7% | 51.7% | 38.2% | 33.3% | |
| 16.4% | 45.6% | 61.0% | 66.7% | |
| 6.9% | 60.5% | 0.8% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 70.6% | 48.7% | 87.7% | 81.2% | |
| 29.4% | 54.7% | 11.6% | 18.8% | |
| no data | no data | 0.4% | no data | |
| no data | no data | 0.3% | no data |
The strong players' line is all about the d4 pawn. After your opponent attacks it, and with the knight on d2 your queen no longer defends it, because the knight blocks the d-file. That is why 4. opens the center before the pressure builds: you trade your e4 pawn for the d5 pawn and force your opponent to recapture at once. If your opponent recaptures with , their queen sits in the middle of the board, but also within reach of your minor pieces, and d4 is hit twice, by the c5 pawn and the queen, with nothing guarding it.
5. brings the other knight to cover d4 and gets you ready to castle short. Your opponent still trades with , and you should not take back right away with the f3 knight: the queen on d5 just captures it and you have given a piece for a pawn. Instead, 6. develops the bishop with an attack on the queen, so you gain a tempo while your opponent still has no minor piece out. Strong players choose it in 98 of every 100 games from that position. After the queen still watches d4 down the d-file, and 7. tucks your king away before you go after the pawn.
This is where 3. pays off. After , the d2 knight jumps to b3 with 8. and joins the f3 knight against d4. If your opponent defends with , you win the pawn back with 9.. Along the way, the knight that left d2 is no longer in the way of your c1 bishop, whose diagonal is now open. Material is level, your king is castled and your pieces are out, while your opponent's queen has already moved twice.
At your level this line is rare. After 4., your opponent takes back with in more than 3 of every 4 games and with in only 16 of every 100, against 61 of every 100 among strong players. After the enemy queen stays home and only the c5 pawn attacks d4, so you have no rush to defend it. The other common choice at your level is 4.: the c3 pawn guards d4, and if your opponent trades on d4 you recapture with a pawn and keep your center. With it you score 50 of every 100. You can also play 4., which develops the knight and guards d4 in one move: at your level it scores 53 of every 100.
| Move | Strong games | Chosen by |
|---|---|---|
| 15,700 | 33.1% | |
| 5,187 | 48.1% | |
| 2,494 | 61.1% | |
| 1,514 | 87.5% | |
| 1,314 | 89.5% | |
| 1,586 | 97.6% | |
| 1,524 | 55.3% | |
| 830 | 87.3% |
After 4. 5. , 6. gives up the knight for a pawn. After 7. 8. 9. their king can no longer castle, but you are a piece down. Retreat with 6.. At your level it is always played here, and opponents often fail to cash in: you still score about 48 of every 100.
100 in 100 play it in this position. It gives away 4 pawns of advantage.
After 4. , 5. moves the same knight a third time and leaves d4 guarded only by your queen, while the enemy queen already hits it down the d-file. After 6. 7. 8. the pawn is gone. Play 5. instead, trading off the attacked knight. With 5. you score 51 of every 100; with 5., 58.
9 in 100 play it in this position. It gives away 1.6 pawns of advantage. It scores 50.7%; scores 57.7%.
After 4. , 5. saves the knight without trading it, but does nothing for d4. After 6. 7. , pawn and knight both hit d4 and the b6 queen eyes the undefended b2 pawn. You end up giving up the center with 8.. Play 5.: with it you score 58 of every 100, and with 5., 52.
8 in 100 play it in this position. It gives away 1.2 pawns of advantage. It scores 51.8%; scores 57.7%.
After 4. , 5. props up the knight with a pawn but opens the diagonal from h4 to your king. After 6. 7. 8. you have lost the e4 pawn and your castling rights. It shows up in 3 of every 100 games from that position, and you score about 46 of every 100. Play 5..
3 in 100 play it in this position. It gives away 2.8 pawns of advantage. It scores 45.5%; scores 57.7%.
After 4. 5. , the f6 knight is no longer pinned and your e4 knight is loose. After 6., grabs it and leaves your g5 bishop attacked twice. After 7. 8. 9. you are a piece and a pawn down. This trap does get punished: you score 27 of every 100.
4 in 100 play it in this position. It gives away 7 pawns of advantage. It scores 27.4%; scores 47.4%.
After 4. 5. , 6. guards the knight, but after 7. and your queen retaking on e4, checks and hits b2 and d4. Block with 9. and b2 falls; play 9. and you lose castling. At your level it rarely gets punished: you score 49 of every 100, more than with 6..
4 in 100 play it in this position. It gives away 4.6 pawns of advantage. It scores 49.4%; scores 47.4%.
| White | Black | Games |
|---|---|---|
| 5,942 | ||
| Not castled | 1,684 | |
| Not castled | Not castled | 1,087 |
| Not castled | 837 | |
| 519 | ||
| 428 |
| White | Black | Games |
|---|---|---|
| 142 | ||
| Not castled | 39 | |
| 20 | ||
| Not castled | Not castled | 13 |
| Not castled | 7 | |
| 7 |
1. 2. 3. · 4,485 amateur games · 82 over the board
The Open System starts with , hitting d4 at once; there you score 51 of every 100.
1. 2. 3. · C05
2,284 amateur games · White's score: 51.9%
1. 2. 3. · C03
638 amateur games · White's score: 58.9%
1. 2. 3. 4. 5. · C05
337 amateur games · White's score: 53.9%
1. 2. 3. · C03
193 amateur games · White's score: 51.6%
1. 2. 3. 4. · C07
131 amateur games · White's score: 46.2%
1. 2. 3. 4. · C04
125 amateur games · White's score: 58.8%
1. 2. 3. 4. 5. 6. 7. 8. · C06
107 amateur games · White's score: 55.6%
1. 2. 3. · C03
53 amateur games · White's score: 57.5%
1. 2. 3. 4. 5. 6. · C07
46 amateur games · White's score: 52.2%
1. 2. 3. 4. 5. · C03
37 amateur games · White's score: 58.1%
1. 2. 3. · C03
31 amateur games · White's score: 56.5%
1. 2. 3. 4. 5. 6. · C05
6 amateur games · White's score: 50.0%
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. · C07
1 amateur game · White's score: 0.0%
1. 2. 3. 4. 5. 6. 7. 8. · C06
1 amateur game · White's score: 100.0%