- Blind mode tutorial
lichess.org
Donate

Rating difference between LiChess and Chess.com is about 100 points? (Chess.com Elo is lower)

These are probably related to players selecting whom they are playing with i.e. there is pool inside a pool. And I am not sure it is even true. here is answer to on gazillion similar threads about ratings . This is fide to lichess with verified fide ratings
---------------------- clip -------------
here is comparison which is best directed effortr so far. and definately good enough that making next one is futiel
www.reddit.com/r/lichess/comments/mdvbkp/yet_another_lichess_vs_fide_rating_comparison/
and why it is futile. Please see scattergram of variaous lichess rating vs verified FIDE ratings
https://i.imgur.com/Txd4qBH.png

accuracy of any prediction is so low that they dont have even entertainment value
---------------------------- clip ---------------
as you can see the prediction accuracy is at best about +- 200 points. There is no reason to assume that lichess-chess is anymore less random. there are plenty of reason why this happens. In fide ratings the actual fide rating can be very inaccurate due fact that only some otb tournaments are fide-rated. Also rate people playing drunk/half a sleep is probably very different.

These are probably related to players selecting whom they are playing with i.e. there is pool inside a pool. And I am not sure it is even true. here is answer to on gazillion similar threads about ratings . This is fide to lichess with verified fide ratings ---------------------- clip ------------- here is comparison which is best directed effortr so far. and definately good enough that making next one is futiel www.reddit.com/r/lichess/comments/mdvbkp/yet_another_lichess_vs_fide_rating_comparison/ and why it is futile. Please see scattergram of variaous lichess rating vs verified FIDE ratings https://i.imgur.com/Txd4qBH.png accuracy of any prediction is so low that they dont have even entertainment value ---------------------------- clip --------------- as you can see the prediction accuracy is at best about +- 200 points. There is no reason to assume that lichess-chess is anymore less random. there are plenty of reason why this happens. In fide ratings the actual fide rating can be very inaccurate due fact that only some otb tournaments are fide-rated. Also rate people playing drunk/half a sleep is probably very different.

its amazing how much confusion there is about ratings.

Chess.com uses the Elo ratings system
Lichess uses Glicko-2 ( a modified version of the Elo rating which includes a volatility in ratings when very few games have been played within a certain timeframe)

https://lichess.org/faq#ratings

The rating is calculated by your performance in games against other players in the same pool (same group of people and same time control). It is a RELATIVE measure of your success, not absolute. Therefore there is no expectation that you will have the same rating when you play in two different pools of players. In the stronger pool you will have a lower rating.

Since we are talking about different pools, it is very possible that in one group there are fewer very strong players than in another one so that the relative ratings in the two groups would be different for a subset of players in the high end of the pools than in the low end.

its amazing how much confusion there is about ratings. Chess.com uses the Elo ratings system Lichess uses Glicko-2 ( a modified version of the Elo rating which includes a volatility in ratings when very few games have been played within a certain timeframe) https://lichess.org/faq#ratings The rating is calculated by your performance in games against other players in the same pool (same group of people and same time control). It is a RELATIVE measure of your success, not absolute. Therefore there is no expectation that you will have the same rating when you play in two different pools of players. In the stronger pool you will have a lower rating. Since we are talking about different pools, it is very possible that in one group there are fewer very strong players than in another one so that the relative ratings in the two groups would be different for a subset of players in the high end of the pools than in the low end.

@EmaciatedSpaniard said in #12:

its amazing how much confusion there is about ratings.

Chess.com uses the ELO ratings system
Lichess uses Glicko-2 ( a modified version of ELO which includes a volatility in ratings when very few games have been played within a certain timeframe)

Could you tell me exactly what ELO stands for please?

@EmaciatedSpaniard said in #12: > its amazing how much confusion there is about ratings. > > Chess.com uses the ELO ratings system > Lichess uses Glicko-2 ( a modified version of ELO which includes a volatility in ratings when very few games have been played within a certain timeframe) > Could you tell me exactly what ELO stands for please?

Actually my chess.com Blitz rating is 2325. Sounds reasonable.

Actually my chess.com Blitz rating is 2325. Sounds reasonable.

ELO doesn't stand for anything. Its the name of the physics professor who came up with the Elo rating system, Arpad Elo. So i should not have written "ELO". Correct is "Elo".

here is more on Elo:
https://en.wikipedia.org/wiki/Elo_rating_system

ELO doesn't stand for anything. Its the name of the physics professor who came up with the Elo rating system, Arpad Elo. So i should not have written "ELO". Correct is "Elo". here is more on Elo: https://en.wikipedia.org/wiki/Elo_rating_system

"ELO doesn't stand for anything."

Wrong. It stands for Electric Light Orchestra.

"ELO doesn't stand for anything." Wrong. It stands for Electric Light Orchestra.

@EmaciatedSpaniard said in #15:

ELO doesn't stand for anything. Its the name of the physics professor who came up with the Elo rating system, Arpad Elo. So i should not have written "ELO". Correct is "Elo".

here is more on Elo:
en.wikipedia.org/wiki/Elo_rating_system

Thanks for clearing that up and while we are on the topic can you also explain where elo fits into chess.com ratings as you said?

@EmaciatedSpaniard said in #15: > ELO doesn't stand for anything. Its the name of the physics professor who came up with the Elo rating system, Arpad Elo. So i should not have written "ELO". Correct is "Elo". > > here is more on Elo: > en.wikipedia.org/wiki/Elo_rating_system Thanks for clearing that up and while we are on the topic can you also explain where elo fits into chess.com ratings as you said?

So i did a bit of research and there too i didn't get it right.
It turns out that chess.com (according to their support site) also use a Glicko rating system with a volatility in the rating representing how accurate your rating is. I.e. there your rating will also go up and down more if you didn't play a lot of games in a while.

So i did a bit of research and there too i didn't get it right. It turns out that chess.com (according to their support site) also use a Glicko rating system with a volatility in the rating representing how accurate your rating is. I.e. there your rating will also go up and down more if you didn't play a lot of games in a while.

Hmmm according to chess goals as well as people who play on both sites, below 2400 Chesscom has higher level of play corresponding to rating. However, around 2400 could be a bit higher or lower, it skews toward lichess.

But rating is subjective, and should only be used as a measure of progress. Especially for online chess. Your primary question should be: Are you improving?

There are also other aspects that impact rating on lichess and Chesscom. Including membership (no ads to slow you down), and the ability to premove a bunch of moves. If you start the premoving before your opponent you have a distinct advantage. As a result, if you can put out 8 pre moves in a row, it is hard for your opponent to stop your premoves especially in drawn bishop endgames. As a result, it is difficult for your opponent to premove. However, if your opponent has much more time than you do, because the premoves cost .1 seconds, you will run out of time eventually.

On lichess, you can utilize tricks to regain the premove advantage. As there is only 1 premove limit. One way is to stop for a brief second to get your opponent off rhythm.

This is also ignoring that you can also stop your opponent from premoving by preventing the premove from occurring with a check ect. But that is more related to chess ability instead of basic parameters.

In other words, you have to learn different styles as well. And as a result for lower time controls there can be even greater variation.

Anyways hope I helped answer your question.

Hmmm according to chess goals as well as people who play on both sites, below 2400 Chesscom has higher level of play corresponding to rating. However, around 2400 could be a bit higher or lower, it skews toward lichess. But rating is subjective, and should only be used as a measure of progress. Especially for online chess. Your primary question should be: Are you improving? There are also other aspects that impact rating on lichess and Chesscom. Including membership (no ads to slow you down), and the ability to premove a bunch of moves. If you start the premoving before your opponent you have a distinct advantage. As a result, if you can put out 8 pre moves in a row, it is hard for your opponent to stop your premoves especially in drawn bishop endgames. As a result, it is difficult for your opponent to premove. However, if your opponent has much more time than you do, because the premoves cost .1 seconds, you will run out of time eventually. On lichess, you can utilize tricks to regain the premove advantage. As there is only 1 premove limit. One way is to stop for a brief second to get your opponent off rhythm. This is also ignoring that you can also stop your opponent from premoving by preventing the premove from occurring with a check ect. But that is more related to chess ability instead of basic parameters. In other words, you have to learn different styles as well. And as a result for lower time controls there can be even greater variation. Anyways hope I helped answer your question.

@EmaciatedSpaniard said in #18:

So i did a bit of research and there too i didn't get it right.

It's amazing how much confusion there is about the ratings.

@EmaciatedSpaniard said in #18: > So i did a bit of research and there too i didn't get it right. It's amazing how much confusion there is about the ratings.

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