In 1,633,127 Lichess games from September 2026, rated 900 to 1700, you score 50.1% with the Caro-Kann Defense. You meet 1. with so that arrives with the pawn already backed by c6. If your opponent locks the chain with 3., your bishop comes out to f5 and your c-pawn goes after d4. Among players rated 2200 or more online, you score 47 in 100.
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.
92 named lines in this family; 3 with their own page.
The opening's share in each Lichess rating band.
Among strong players, the most repeated line is 2. 3. 4. 5. . At your level opponents spread out more: 2. comes in 4 in 10 games, against 7 in 10 among players rated 2200 or more, and 2. in 32 in 100. After , 3. shows up in 49 in 100 games at your level. After , expect 4. most often, in 35 in 100, then 4., in 21. 4., the strong players' favourite, appears in fewer than 5 in 100.
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 38.5% | 51.2% | 68.1% | 77.6% | |
| 31.8% | 49.1% | 13.2% | 14.2% | |
| 9.9% | 46.5% | no data | no data | |
| 6.8% | 49.4% | 8.6% | 5.6% | |
| 3.2% | 50.4% | 0.8% | no data | |
| 2.4% | 49.6% | 3.6% | 0.6% | |
| no data | no data | 3.7% | 2.0% |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 93.4% | 51.0% | 95.4% | 99.2% | |
| 1.9% | 52.7% | 0.1% | no data | |
| 1.7% | 52.8% | 2.8% | 0.8% | |
| 0.5% | 53.3% | 0.8% | no data | |
| 0.4% | 56.5% | no data | no data | |
| 0.4% | 57.1% | 0.2% | no data | |
| no data | no data | 0.2% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 48.9% | 50.1% | 36.3% | 58.5% | |
| 24.8% | 50.2% | 26.2% | 15.2% | |
| 18.6% | 53.7% | 22.6% | 15.2% | |
| 2.9% | 54.0% | 5.4% | 3.8% | |
| 2.1% | 56.2% | 8.6% | 7.4% | |
| 1.0% | 50.9% | 0.6% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 56.6% | 50.5% | 65.2% | 75.7% | |
| 30.9% | 47.5% | 30.4% | 23.5% | |
| 6.6% | 54.1% | 1.2% | no data | |
| 2.4% | 56.7% | no data | no data | |
| 0.8% | 59.7% | no data | no data | |
| 0.7% | 48.1% | 0.5% | no data | |
| no data | no data | 0.8% | no data | |
| no data | no data | 0.8% | no data | |
| no data | no data | no data | 0.7% |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 34.7% | 49.7% | 27.0% | 31.2% | |
| 21.0% | 51.4% | 7.1% | 4.7% | |
| 9.1% | 49.4% | 6.8% | 4.3% | |
| 7.6% | 49.9% | no data | no data | |
| 6.5% | 49.8% | no data | no data | |
| 4.5% | 58.3% | 34.8% | 35.0% | |
| no data | no data | 5.6% | 5.3% | |
| no data | no data | 5.2% | 12.9% |
The strong players' line, 2. 3. 4. 5. , does two things in a set order. First, it gets your light-squared bishop out before a pawn on e6 shuts it in. Then it goes after d4, the base of the d4–e5 chain, with your c-pawn.
develops nothing. It prepares , so the pawn that confronts e4 arrives already protected. With 3. your opponent gains space, but the chain is now fixed and e5 has a single defender, the pawn on d4. That pawn is what you will attack.
That is why comes before you move the e-pawn: with a pawn on e6, the c8 bishop would be stuck behind d5 and e6. Outside the chain, though, the bishop becomes a target for pawns. 4. prepares a push to g4, which would hit it. stops that, because g4 is now covered by your h5 pawn and by the bishop itself. Then 5. attacks the bishop, which has no defender on f5, and trades it off. Your opponent loses their light-squared bishop, and you get rid of the one the chain was going to bury.
With the bishop sorted, it is the c-pawn's turn. It already went to c6, and now it moves again, to c5, to hit d4: if d4 falls or gets traded, e5 loses its support. This is the most common plan in the structure: 6 in 10 strong online players choose it, and 1 in 2 over-the-board players rated 2300 or more. You can play it straight away with , before the bishop comes out, and at your level it works: with you score almost 53 in 100, and with , almost 50. Hitting e5 with is worse, because the f-pawn no longer guards e6.
At your level this sequence breaks off early, since 4. is rare. When it does come, the reply worth learning is : 7 in 10 strong players choose it, but fewer than 3 in 10 players at your level. Without it, the pawn reaches g4 and your bishop has to retreat.
Plan played in 50% of the 681 over-the-board games rated 2300+ with this structure.
Online, 2200+: 59.4% of 22,055. Over the board, 2300+: 50.2% of 681; the alternative, : 5.6% and 10.1%.
Plan played in 14% of the 90 over-the-board games rated 2300+ with this structure.
Online, 2200+: 10.7% of 3,379. Over the board, 2300+: 14.4% of 90; the alternative, : 3.0% and 2.2%.
| Move | Strong games | Chosen by |
|---|---|---|
| 91,989 | 68.3% | |
| 64,473 | 95.4% | |
| 61,718 | 36.7% | |
| 22,678 | 65.5% | |
| 14,822 | 34.8% | |
| 5,134 | 70.0% | |
| 3,580 | 57.9% | |
| 2,058 | 94.5% |
After 3., lets the e-pawn advance with tempo: 4. hits the knight, which goes back with , and after 5. the push 6. hits f7 and the knight on d7. Best is . It is an inaccuracy: you score 45 in 100 instead of 47, and at your level opponents often don't punish it.
12 in 100 play it in this position. It gives away 0.8 pawns of advantage. It scores 45.3%; scores 47.2%.
Against 3., shuts your c8 bishop in behind d5 and e6, the very thing the Caro-Kann is built to avoid. After 4. 5., d4 is undefended, but if you grab it with , 6. hits your queen and makes it move again. It is an inaccuracy: you score 46 in 100, against 47 with .
11 in 100 play it in this position. It gives away 0.6 pawns of advantage. It scores 46.3%; scores 47.2%.
hits e5, but after 4. the pawn has two defenders, d4 and the knight. And with the f7 pawn gone, only the c8 bishop covers e6, while your g8 knight loses the f6 square. It is an inaccuracy, but a costly one: you score 43 in 100 with it and almost 50 with .
2 in 100 play it in this position. It gives away 0.8 pawns of advantage. It scores 43.3%; scores 49.5%.
After 2. 3., hits the knight and grabs space, but the pawn drifts away from its support. 4. steps aside and 5. attacks d4; if you trade with , then after 6. the pawns on c3 and e4 control d4 and d5. Best is . You score 48 in 100 instead of 51.
13 in 100 play it in this position. It gives away 0.8 pawns of advantage. It scores 48.2%; scores 51.1%.
After 2. 3., pins the f3 knight to the queen. 4. puts the question and, after 5., your opponent keeps both bishops. Without your light-squared bishop, 6. hits b7 and you have to cover it with . Best is , but at your level opponents don't punish this: you score 54 in 100.
26 in 100 play it in this position. It gives away 1.1 pawns of advantage. It scores 54.1%; scores 53.4%.
After 2. 3., the natural move is , which keeps a pawn on d5. With , 4. takes the c6 pawn too and, after , you have no c- or d-pawn left while your opponent takes the centre with 6.. You score 48 in 100 instead of 52.
5 in 100 play it in this position. It gives away 0.8 pawns of advantage. It scores 48.5%; scores 52.0%.
| White | Black | Games |
|---|---|---|
| 739,280 | ||
| Not castled | 290,308 | |
| Not castled | Not castled | 200,498 |
| Not castled | 108,557 | |
| 105,298 | ||
| 77,012 |
| White | Black | Games |
|---|---|---|
| 775 | ||
| Not castled | 246 | |
| Not castled | Not castled | 136 |
| 107 | ||
| 78 | ||
| Not castled | 64 |
1. 2. 3.
341,227 amateur games · 682 over the board
3. locks the chain: the most played line, 341,227 games; you score almost 50 in 100.
1. 2. 3.
201,743 amateur games · 140 over the board
After 3. you recapture with and keep a pawn on d5: you score 50 in 100.
1. 2. · 151,405 amateur games
2. brings the bishop out before d4, and hits it: you score 53.5 in 100.
1. 2. 3. · 53,501 amateur games · 163 over the board
2. and 3. develop both knights without playing d4: you score 49 in 100.
1. 2. 3. 4. · 42,927 amateur games · 65 over the board
After 3. 4., develops with an attack on the knight: you score 47.5 in 100.
1. 2. · 36,165 amateur games · 8 over the board
2. props up e4 with a pawn instead of taking the centre: you score 51 in 100.
1. 2. · 24,850 amateur games · 28 over the board
2. eyes d5 before you can play : you score 46.5 in 100.
1. 2. 3. 4. · 22,879 amateur games · 66 over the board
After 3. , 4. attacks your pawn on d5: you score 46 in 100.
1. 2. 3. · 18,369 amateur games · 45 over the board
3. props up e4 with the f-pawn: you score 46 in 100 across 18,369 games.
1. 2. 3. · 17,657 amateur games
After 3., lets 4. hit the knight: you score 45 in 100.
1. 2. 3. · 16,444 amateur games · 59 over the board
After 2. , 3. props up e4 without taking the centre: you score 46.5 in 100.
1. 2. 3. 4. · 16,246 amateur games · 5 over the board
After 3. , your opponent wins the pawn back with 4.: you score 45 in 100.
1. 2. 3. 4. 5. · 14,507 amateur games · 129 over the board
After 5., opens the e-file and your f8 bishop's diagonal: you score 50 in 100.
1. 2. 3. 4. · 5,046 amateur games · 1 over the board
After 3. , 4. goes after the pawn on e4: you score 51 in 100.
1. 2. 3. 4. 5. · 4,612 amateur games · 41 over the board
After 5. the queens are gone and your opponent can't castle: you score 57 in 100.
1. 2. · 4,233 amateur games · 1 over the board
2. prepares the bishop on b2: you score 51 in 100 across 4,233 games.
1. 2. 3. · 4,229 amateur games · 14 over the board
3. defends e4 with the knight: one of your worst lines, at 44 in 100.
1. 2. 3. 4. 5. 6. · 3,467 amateur games
After , 6. gives up the knight next to your king: you score about 4 in 10.
1. 2. 3. 4. · 3,452 amateur games
After 3. , 4. hits e4 and offers a pawn for development: you score 46 in 100.
1. 2. 3. 4. · 3,127 amateur games · 27 over the board
readies the other knight for f6 without risking doubled pawns: you score 49 in 100.
1. 2. 3. 4. · B15
3,032 amateur games · White's score: 58.5%
1. 2. 3. · B15
2,729 amateur games · White's score: 54.1%
1. 2. 3. 4. 5. · B15
2,565 amateur games · White's score: 55.2%
1. 2. 3. 4. 5. 6. · B15
2,426 amateur games · White's score: 52.5%
1. 2. 3. 4. · B10
2,415 amateur games · White's score: 49.2%
1. 2. 3. · B10
2,389 amateur games · White's score: 56.5%
1. 2. · B12
2,226 amateur games · White's score: 56.6%
1. 2. 3. 4. 5. · B16
792 amateur games · White's score: 57.5%
1. 2. 3. · B10
709 amateur games · White's score: 49.5%
1. 2. · B10
672 amateur games · White's score: 43.3%
1. 2. 3. · B12
632 amateur games · White's score: 57.9%
1. 2. · B10
460 amateur games · White's score: 45.2%
1. 2. 3. 4. · B16
428 amateur games · White's score: 52.6%
1. 2. 3. 4. · B10
420 amateur games · White's score: 48.3%
1. 2. · B12
319 amateur games · White's score: 56.1%
1. 2. 3. · B10
274 amateur games · White's score: 53.5%
1. 2. 3. · B15
236 amateur games · White's score: 58.7%
1. 2. 3. 4. 5. 6. 7. · B18
216 amateur games · White's score: 69.7%
1. 2. 3. 4. 5. · B12
171 amateur games · White's score: 46.8%
1. 2. · B12
95 amateur games · White's score: 59.5%
1. 2. 3. · B12
94 amateur games · White's score: 50.0%
1. 2. 3. · B10
27 amateur games · White's score: 46.3%
1. 2. 3. 4. 5. · B12
22 amateur games · White's score: 75.0%
1. 2. 3. · B10
11 amateur games · White's score: 54.5%
1. 2. 3. 4. · B10
9 amateur games · White's score: 55.6%
1. 2. 3. 4. · B10
3 amateur games · White's score: 66.7%
1. 2. 3. 4. · B10
1. 2. 3. 4. 5. 6. 7. · B13
You answer 1. with so that comes with a protected pawn. It shows up in 5 in 100 games between 900 and 1700, and more often as ratings rise: 4 in 100 at 900 to 1100, 6 in 100 at 1500 to 1700. In every one of those bands you score about 50 in 100.
Hit the base of your opponent's chain. When your opponent plays 3., the pawn on e5 leans on d4 alone. Your c-pawn goes to c5 and attacks d4: if d4 goes, e5 loses its support. 6 in 10 strong online players pick that plan in this structure.
Get it out before the e-pawn moves. After 3., puts the bishop outside the chain before a pawn on e6 closes it in behind d5. If 4. prepares g4 to chase it, answer . When 5. offers a trade, accepts it.
It holds up at every level. Between 900 and 1100 you score 50 in 100 with it, the same as between 1500 and 1700. The plan is short to state: protect d5 with c6, get the light-squared bishop out before playing e6, then attack d4 with the c-pawn. Watch out for 3., where you score only 44 in 100.