Free online Chess server. Play Chess now in a clean interface. No registration, no ads, no plugin required. Play Chess with the computer, friends or random opponents.
Play
Create a game Tournament Simultaneous exhibitions
Learn
Chess basics Puzzles Practice Coordinates Study Coaches
Watch
Lichess TV Current games Streamers Broadcasts (beta) Video library
Community
Players Teams Forum Questions & Answers
Tools
Analysis board Opening explorer Board editor Import game Advanced search
Sign in
Reconnecting

Average centipawn value not calculated correctly?

Asked by studentt
Tags question about analysis
Activity Viewed 232 times, last updated
1
I have a self-made game:

lichess.org/Ymq1nWek/black

I try to verify the average centipawn loss for the white player, which is 72:

100 * (0.2 + 1.0 + 0.8) / 3 = 66.67

- 0.2 because the initially evaluation gives 0.0 score but after 1.a4, the evaluation drops to 0.2
- 1.0 because the evaluation before 2.Ra2 is +0.4 but after 2.Ra2, the evaluation drops to -0.6
- Similarly for 0.8

How does the lichess software derived 72, when the correct average centipawn loss should have been 66.67?




2 Answers
2
Answered by jmcampbell
1) The initial evaluation is +0.1, not 0. So the calculation is:
100 * (0.3 + 1.0 + 0.8) / 3 = 70

2) The engine only displays one decimal place of precision. So what you think is a 1.0 might actually be a 1.04, and instead of 0.8 it might actually be 0.82. This example would give an average centipawn loss of 72.
1
Answered by DoTheMath
Lots of good information in the comments and posted URLs:
lichess.org/qa/103/what-is-average-centipawn-loss

A post made by user kettwiesel has the guts of it:
lichess.org/forum/general-chess-discussion/average-centipawn-loss#6

Only registered members with one week of lichess activity can contribute to the Q&A.