# Lichess stats for average centipawn loss

I am not able to understand the meaning of this stats on Lichess. How can I analyze that and work on my weaker points.

Does it mean I make bad Queen moves?

Average centipawn loss (acl) is the score of engine less the score of your move. As the acl decreases your accuracy increases. If your acl is lower (close to zero) you are a strong player.

Example

``````game_number: 1, move_number: 20, move: d4d5, score: 20,
engine_move: g1f3, engine_score: 30
cl: 30-20 or 10 (cl=centipawn loss)

game_number: 2, move_number: 36, move: b1c3, score: 18,
engine_move: f3g5, engine_score: 40
cl: 40-18 or 22

acl = (10+22)/2 = 16

The 2 is from the 2 moves with engine evaluations.
``````

You are right your queen moves are less accurate compared to other pieces.

Most players or all players have lower accuracy for queen moves - including me. This is normal because the queen has the greatest mobility of all pieces, it can move like a rook and a bishop, it takes more time to calculate accurately.

Mine

To improve upon it just practice analyzing positions with queens still on the board. Another way is without looking on the board imagine the queen is located in E4 square and any other squares, on an empty board except the queen, where the queen can go? F3, D5 ... The speed of locating the squares will help a lot in blitz or bullet games.

Magnus Carlsen
I tried to analyze the 100 latest Lichess blitz games of Magnus using stockfish 14.1. Analysis start at move 12. Maximum error is set to 1000 cp. The queen also has the highest error.

``````   pieces  cploss  totalmoves
0    pawn      39         618
1  knight      25         488
2  bishop      35         394
3    rook      45         566
4   queen      58         370
5    king      33         531
``````

• Damn son, beat those knights into shape and you'll be seriously challenging Magnus! Mar 3 at 10:17