In the Caro-Kann Advance Variation you score 50.2% on Lichess (September 2026, N = 341,227 games at 900 to 1700). After 1. 2. , you push 3.: you gain space and take f6 away from the g8 knight. Strong players answer in 66 of every 100 games, and your plan revolves around that bishop.
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.
8 named lines in this family; 1 with their own page.
The opening's share in each Lichess rating band.
Against 3. you will mostly see two replies: , which hits d4 and shows up in 28 of every 100 games at your level, and . Against the bishop, players at your level reach for 4. far more often than 4., while strong players pick 4. in 35 of every 100. After 4. the replies depend on level too. At your level they split between (35 of every 100), (32) and (28). Strong players go in 70 of every 100. Against you score 69 of every 100; against , 52.
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 35.0% | 54.4% | 14.8% | 13.1% | |
| 32.1% | 68.7% | 1.1% | no data | |
| 28.3% | 51.5% | 70.0% | 70.5% | |
| 2.0% | 53.9% | 1.2% | no data | |
| 1.0% | 72.2% | no data | no data | |
| 0.9% | 57.1% | no data | no data | |
| no data | no data | 5.3% | 4.0% | |
| no data | no data | 5.2% | 12.5% |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 51.3% | 52.3% | 58.3% | 46.3% | |
| 14.4% | 60.8% | 19.9% | 7.3% | |
| 13.7% | 43.1% | 1.2% | no data | |
| 5.3% | 59.5% | 18.6% | 46.3% | |
| 4.7% | 52.1% | no data | no data | |
| 4.2% | 45.9% | 0.8% | no data | |
| no data | no data | 0.8% | no data |
| Move | Lichess 900–1700 | White's score | Online, 2200+ | Over the board, 2300+ |
|---|---|---|---|---|
| 91.1% | 50.8% | 94.7% | 87.7% | |
| 3.6% | 69.4% | no data | no data | |
| 2.9% | 56.9% | 1.1% | no data | |
| 2.5% | 70.0% | 3.9% | 12.3% | |
| no data | no data | 0.3% | no data |
The strong players' line is 4. 5. 6. 7.. It heads for one square: g5, the one Black's h5 pawn leaves weak. You want to trade off the bishop Black brought out to f5, then plant your c1 bishop on g5, where Black's h-pawn can no longer kick it.
It starts with 3.. Your e5 pawn, backed by d4, locks the center and takes f6 away from the g8 knight. Black gets the c8 bishop out with before playing the pawn to e6, because after that the bishop would be stuck behind its own pawns.
4. goes after that bishop. Your queen on d1 already covers g4, so your g-pawn can advance to g4 and hit the bishop, with the h-pawn joining the chase. Black has to choose how to stop it. gives the bishop a retreat on h7, and that pawn still covers g5. If you get instead, as 32 of every 100 do at your level, the bishop has no h7, since the pawn still sits there, and nothing stops the push to g4. Against that reply you score 69 of every 100. With , the choice of 70 of every 100 strong players, the pawn on h5 controls g4. The price is g5: the h-pawn can no longer pass through h6 to guard it.
Now 5.. Your bishop faces the one on f5 along the d3-f5 diagonal and offers a trade. Strong players accept with in 95 of every 100 games, and players at your level in 91. After 6. your queen is developed on d3, on the diagonal toward g6 and h7, and Black no longer has the bishop it worked to bring out.
With Black completes the c6-d5-e6 chain and vacates e7. That empty square opens the diagonal between g5 and d8, which is why 7. comes in this order: the bishop arrives hitting the queen. On g5 your h4 pawn protects it, and with Black's pawn already on h5, the h-pawn can never reach h6 to chase it away. Strong players go for it in 65 of every 100 games.
Results depend a lot on level. At 900-1100 you score 53 of every 100, and at 1500-1700 that drops to 49, but strong players score 53 of every 100 online and 56 over the board.
| Move | Strong games | Chosen by |
|---|---|---|
| 22,678 | 65.5% | |
| 14,822 | 34.8% | |
| 5,134 | 70.0% | |
| 3,580 | 57.9% | |
| 2,058 | 94.5% | |
| 1,941 | 98.7% | |
| 1,904 | 68.3% | |
| 1,285 | 65.4% |
A blunder. After 4. , 5. opens the h4-e1 diagonal: your f-pawn has already left f2 and no pawn is left to block on g3. Then 6. , hitting your h1 rook, 7. 8.. Material is level, but you have lost the right to castle. At your level opponents often miss it.
5 in 100 play it in this position. It gives away 7.5 pawns of advantage. It scores 47.5%; scores 50.4%.
An inaccuracy. After 4. , 5. takes the bishop off c1, where it guarded b2, and hits b2 and d4 at once. After 6. 7. your d4 pawn is gone, and 8. kicks the queen but leaves you a pawn down. Play 5. instead, keeping the c1 bishop on b2 duty.
6 in 100 play it in this position. It gives away 0.8 pawns of advantage. It scores 47.7%; scores 50.7%.
A mistake. After 4. , Black has moved the c-pawn twice and nothing defends it: 5. grabs it. With 5. you hold d4, but after 6. the knight piles on d4 and Black has the c5-d5-e6 chain with the bishop already outside it. 43 of every 100 players choose it, and you score 49 instead of 56.
43 in 100 play it in this position. It gives away 1.3 pawns of advantage. It scores 49.2%; scores 56.4%.
An inaccuracy. After 4. , 5. invites a trade that costs you d4. After 6. the knight that guarded d4 is gone and the queen has left d1, so wins the pawn. With 7. 8. you get a lead in development for it. Play 5.. You score 44 of every 100, against 48.
15 in 100 play it in this position. It gives away 0.7 pawns of advantage. It scores 44.3%; scores 48.1%.
A mistake, though at your level it goes unpunished. After 4. , trading at once with 5. forces a pawn recapture and leaves Black with doubled g-pawns. After 5. your opponent decides: 6. 7. 8., trading once the knight is out. Even so, you score 50 of every 100, against 48.
28 in 100 play it in this position. It gives away 1 pawn of advantage. It scores 49.6%; scores 48.1%.
An inaccuracy with the same flaw as 5.. After 4. , 5. pulls the bishop off c1, where it covered b2. hits b2 and d4; after 6. 7. 8. you kick the queen, but the d4 pawn is lost. Play 5.. 2 of every 100 players choose it, and you score 48 against 51.
2 in 100 play it in this position. It gives away 0.7 pawns of advantage. It scores 48.0%; scores 50.7%.
| White | Black | Games |
|---|---|---|
| 137,226 | ||
| Not castled | 76,548 | |
| Not castled | Not castled | 47,134 |
| 20,803 | ||
| Not castled | 19,212 | |
| 14,091 |
| White | Black | Games |
|---|---|---|
| 270 | ||
| Not castled | 175 | |
| Not castled | Not castled | 73 |
| 65 | ||
| Not castled | 29 | |
| 22 |
1. 2. 3. · 95,504 amateur games · 159 over the board
strikes at d4 right away instead of developing the bishop. You score 48 of every 100.
1. 2. 3. 4. · 88,687 amateur games · 161 over the board
4. develops the knight and supports d4 before going after the bishop. You score 50 of every 100.
1. 2. 3. 4. · 18,829 amateur games · 23 over the board
4. brings out the knight, which eyes d5 and e4. You score 49 of every 100.
1. 2. 3. 4. · 8,001 amateur games · 179 over the board
4., the strong players' pick, prepares to chase the f5 bishop. You score 58 of every 100.
1. 2. 3. 4. · 6,135 amateur games · 27 over the board
4. hits the f5 bishop with a pawn straight away. You score 54 of every 100.
1. 2. 3. 4. · B12
918 amateur games · White's score: 49.7%
1. 2. 3. 4. · B12
278 amateur games · White's score: 50.0%
It is the line 1. 2. 3.. Your pawn on e5, backed by d4, gains space and takes f6 away from the g8 knight, so the center stays locked. It shows up in 11 of every 1000 Lichess games at 900 to 1700, and more often as ratings rise.
At 900 to 1700 you score 50 of every 100, so it is level overall: 53 at 900-1100 and 49 at 1500-1700. Strong players get more out of it: they score 53 of every 100 online and 56 over the board.
After 3. , strong players go 4., planning to chase the bishop with the g-pawn. After , 5. offers the bishop trade, 6. develops your queen, and once clears e7, 7. lands on a square Black's h-pawn can no longer cover.