Whats diffrent between real life rating and online chess rating ? • page 2/2 • General Chess Discussion • lichess.org

Great question.

From what I've seen:

There are players whose FIDE and Lichess scores are identical.
There are players whose FIDE scores are higher than the Lichess score.

But for most of the time, FIDE is anywhere from slightly lower (50 points), to significantly lower (200 points), with the majority of the time it being closer to the "200" side of things.

@OtmanBelghard Great question. From what I've seen: There are players whose FIDE and Lichess scores are identical. There are players whose FIDE scores are higher than the Lichess score. But for most of the time, FIDE is anywhere from slightly lower (50 points), to significantly lower (200 points), with the majority of the time it being closer to the "200" side of things.

Kilbroney

#12

patzer81 said it all. Game Over.

doom12384

#13

If I'm not mistaken, you can find the standard deviation of all FIDE ratings. The number of standard deviations you are above the mean lichess rating should be approximately equal to the number of standard deviations you are above the FIDE mean, assuming that the lichess playerbase for the given time control is representative of the FIDE playerbase for the comparable time control (which is probably not a valid assumption).

Either way, there should be a better way to calculate it than simply plus or minus some value. Of course, this also doesn't take into account time controls, which will throw the calculations off by quite a bit. 15+15 "classical" is probably more like a FIDE Rapid rating. Make sure the data sets you're comparing are actually comparable. This is the main reason why the assumption that the lichess playerbase is representative of the FIDE playerbase is likely invalid. Comparing lichess blitz and rapid to FIDE blitz is probably more valid than anything involving slower time controls.

At the end of the day, you can approximate it reasonably well, but there are a lot of factors that affect the approximation. Take it all with a grain of salt.

If I'm not mistaken, you can find the standard deviation of all FIDE ratings. The number of standard deviations you are above the mean lichess rating should be approximately equal to the number of standard deviations you are above the FIDE mean, assuming that the lichess playerbase for the given time control is representative of the FIDE playerbase for the comparable time control (which is probably not a valid assumption). Either way, there should be a better way to calculate it than simply plus or minus some value. Of course, this also doesn't take into account time controls, which will throw the calculations off by quite a bit. 15+15 "classical" is probably more like a FIDE Rapid rating. Make sure the data sets you're comparing are actually comparable. This is the main reason why the assumption that the lichess playerbase is representative of the FIDE playerbase is likely invalid. Comparing lichess blitz and rapid to FIDE blitz is probably more valid than anything involving slower time controls. At the end of the day, you can approximate it reasonably well, but there are a lot of factors that affect the approximation. Take it all with a grain of salt.