Hi @KasparGarriow
The elo rating system is an old one as Arpad Elo invented this rating system as an improvement over the original rating system developed in 1950 by Kenneth Harkness. Arpad Elo's system was supposed to be an improvement to this and by 1960, it was already pretty well accepted. Various improvements have happened in the years from them to now - including the similar glicko rating system and glicko 2.0 system. I mention this because "elo" has become an informal way to say "rating." Technically, chess.com doesn't even use the outdated elo system. If I am up to date, chess.com uses glicko 2.0 for calculating its ratings.
This is all extra history though; this is the basics behind how chess.com calculates rating:
https://support.chess.com/article/210-how-do-ratings-work-on-chess-com
If you want more of the math side, here is a pdf of the glicko rating and is basically what chess.com computers. http://www.glicko.net/research/gdescrip.pdf
Glicko 2.0 is virtually identical; it is just that 2.0 changes the Rating Deviation (RD) statistic (called k-factor in elo if I recall) and that basically means the "confidence" that chess.com article refers to. This RD statistic is what helps someone reach their actual rating faster. If someone were to just gain (roughly) 8 rating points per win, then it would take a long time for someone with a new account to reach their true ability if they are a high rated player and similarly, this would be a lot of games where they would face players way below their rating. This RD part is what calculates how much more to increase the rating by, so that someone can reach their true rating faster.
Hello all,
I would like to know how the Elo point gain is calculated. The stronger the opponent in relation to me, the more points - clearly!
But how much difference means how many points?
For example: For every 50 points my opponent has more than me, I get 1 point more?
Thanks for your help!