Your network blocks the Lichess assets!

lichess.org
Donate

Highest Percentile current games to be broadcast rather than Highest Rated

while waiting for a game I see games where it is clear that it will be all about making the time cutoff or terrible blunders in the remaining time, as due to the Highest Rated game policy we tend to see Bullet where the ratings are absurdly high considering the quality of play. Apparently this stems from a decision to make a human's bullet rating similar to their standard or blitz rating, where for 960 all the ratings are much lower due to computer analysis showing the "accuracy" to be lower. I want to see a game where the chess itself is still interesting. 960 is never shown as a result of Highest Rating policy. Highest Percentile will fix this.

while waiting for a game I see games where it is clear that it will be all about making the time cutoff or terrible blunders in the remaining time, as due to the Highest Rated game policy we tend to see Bullet where the ratings are absurdly high considering the quality of play. Apparently this stems from a decision to make a human's bullet rating similar to their standard or blitz rating, where for 960 all the ratings are much lower due to computer analysis showing the "accuracy" to be lower. I want to see a game where the chess itself is still interesting. 960 is never shown as a result of Highest Rating policy. Highest Percentile will fix this.

@themiddleway said in #1:

Apparently this stems from a decision to make a human's bullet rating similar to their standard or blitz rating, where for 960 all the ratings are much lower due to computer analysis showing the "accuracy" to be lower.
Rating calculations do not have anything to do with accuracy or engine analysis in general. All that matters are ratings and rating deviations of the players and the result (white won / draw / black won).

IMHO the reason why the upper tail is so much stronger in bullet (e.g. 2500 is 98.2% percentile in bullet, 99.6% in blitz, 99.9% in rapid and in classical it already rounds to 100.0%) is that much fewer players are able to actually think fast enough to be able to "play chess" in bullet time controls, therefore the rating distance from an average player is higher.

I want to see a game where the chess itself is still interesting. 960 is never shown as a result of Highest Rating policy.
You can select time control class or variant in the left column of https://lichess.org/tv if you want to watch slower time control games or Chess960.

@themiddleway said in #1: > Apparently this stems from a decision to make a human's bullet rating similar to their standard or blitz rating, where for 960 all the ratings are much lower due to computer analysis showing the "accuracy" to be lower. Rating calculations do not have anything to do with accuracy or engine analysis in general. All that matters are ratings and rating deviations of the players and the result (white won / draw / black won). IMHO the reason why the upper tail is so much stronger in bullet (e.g. 2500 is 98.2% percentile in bullet, 99.6% in blitz, 99.9% in rapid and in classical it already rounds to 100.0%) is that much fewer players are able to actually think fast enough to be able to "play chess" in bullet time controls, therefore the rating distance from an average player is higher. > I want to see a game where the chess itself is still interesting. 960 is never shown as a result of Highest Rating policy. You can select time control class or variant in the left column of https://lichess.org/tv if you want to watch slower time control games or Chess960.

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