lichess.org
Donate

Why is the average rating for moves in the lichess opening database so much higher than for players?

When you mouse over a move in the opening database set to "lichess", the average rating for 1.e4 is 1906. The average rating on the site is 1500. I doubt stronger players play than many more games on lichess, at least not enough to explain this. What's going on? Are weaker player's games just not added to the database?
e4 is the most common opening in chess. You can't just base the rating on one move. If you want to be more accurate, go further into the opening database.

There is also very little room for error in the opening besides center forks, random blunders, and traps (most of which require a specific opening to use).
@ThatChapThere said in #1:
> When you mouse over a move in the opening database set to "lichess", the average rating for 1.e4 is 1906. The average rating on the site is 1500. I doubt stronger players play than many more games on lichess, at least not enough to explain this. What's going on? Are weaker player's games just not added to the database?

You are telling the truth. My match does not appear in the database at all. This is probably due to the fact that the matches of the players with high scores are followed more.
@ChessVip11 said in #3:
> You are telling the truth. My match does not appear in the database at all. This is probably due to the fact that the matches of the players with high scores are followed more.
Are you on the "Masters" tab or the "Lichess" tab of the database?
@InkyDarkBird said in #2:
> e4 is the most common opening in chess. You can't just base the rating on one move. If you want to be more accurate, go further into the opening database.
>
> There is also very little room for error in the opening besides center forks, random blunders, and traps (most of which require a specific opening to use).

You misunderstand my post. I'm not trying to test how good of a move 1.e4 is. I'm asking why the average rating for a move which is played at similar frequencies at all levels has an average rating much higher than the average player. This suggests a biased sample.
afaik games below an average rating of (some number, 1400? 1600? 1200?) are not included. basically because nobody cares what very low rated players play in the opening, since it is basically random anyway.
@glbert said in #6:
> afaik games below an average rating of (some number, 1400? 1600? 1200?) are not included. basically because nobody cares what very low rated players play in the opening, since it is basically random anyway.
No. We do play seriously. Take us seriously too or else I will defeat you any second we start playing!
friend, it's not about what i think. i take every opponent seriously.
@glbert said in #6:
> afaik games below an average rating of (some number, 1400? 1600? 1200?) are not included. basically because nobody cares what very low rated players play in the opening, since it is basically random anyway.

Thought so. Has there been an official statement from lichess on this, or did they just make the database and not tell anyone the criteria? Couldn't find anything on google.
> Has there been an official statement from lichess on this, or did they just make the database and not tell anyone the criteria?

i dunno. but if you look at the settings you can have for the database, it's pretty clear that it's the case anyway from the rating subsets you can pick.

This topic has been archived and can no longer be replied to.