Correlation between Fide rating and online rating

This information comes from a survey with a sample of 370 people. It shows that for other websites, like chess.com, the correlation between Fide rating and online rating is stronger than for lichess. Why do you think this is the case? Looking forward to an interesting discussion!

chess.stackexchange.com/questions/189/whats-the-correlation-between-fide-rating-and-online-rating

I don't know how reliable a sample of 370 is, what if they were all from the same area? But who knows.

the sheet had absolutely no measurement of correlation. different pools have different means to the absolute values will not be same. As for correlation: They are definately poor anyway. estimating fide from online would result of accuracy of +-200 pts which you can get just by playing few games against rated player at home

You see similar lines, right? „Different“ with method.

No correlation would mean a cloud of random points. One can discuss the strength of the correlation, it is not that weak in my opinion.

Correlation was actually high. Chess.com blitz to fide had r^2 = 0.75 which mean 75% of variation in FIDE rating can be explained by variation in chess.com rating.

But as makeing estimate. There was simple linear regressin model done in one of the answers and estimate of STD deviation. SD=200 means that estimate done by the formulat woud hit the mark in +-200 pt accuraty about 67% time. And with about 96% time with accuracy of +-400 .

So theya re obviously correlated but still if you want toknow how well you would do OTB then attending OTB is the way to go