"BTW, at chess.com the starting rating can be anywhere from 800 to 1800 for a regular sign-up. Potential members are given options, various devices have restrictions, every sign-up procedure (by facebook), Android, Iphone for examples all have different boxes available to check your starting rating."
Yes, but this is a relatively new thing, and the vast majority of old accounts started at 1200, and 1200 is the default rating (which is why I said for unknown skill in my post - you have to change it to something else, if chess.com has no information, it starts the player at 1200).
And yes, established ratings are closer and closer to Elo, because rating deviation is approaching zero, which is what Elo actually is (Glicko with rating deviations of zero). Of course, ratings don't stay established in glicko systems. As soon as you stop playing a game, it starts to creep back up.
And the answer to will the same player achieve the same rating at different sites is definitely a no in most cases, and there is no reason to expect it to happen.
"BTW, at chess.com the starting rating can be anywhere from 800 to 1800 for a regular sign-up. Potential members are given options, various devices have restrictions, every sign-up procedure (by facebook), Android, Iphone for examples all have different boxes available to check your starting rating."
Yes, but this is a relatively new thing, and the vast majority of old accounts started at 1200, and 1200 is the default rating (which is why I said for unknown skill in my post - you have to change it to something else, if chess.com has no information, it starts the player at 1200).
And yes, established ratings are closer and closer to Elo, because rating deviation is approaching zero, which is what Elo actually is (Glicko with rating deviations of zero). Of course, ratings don't stay established in glicko systems. As soon as you stop playing a game, it starts to creep back up.
And the answer to will the same player achieve the same rating at different sites is definitely a no in most cases, and there is no reason to expect it to happen.
Agreed. A players rating varies from site to site. There exists several reasons for this. But maybe my point was misunderstood, which is: The different ratings are not due to how ratings are calculated by a site. Very similar gains or losses of points for established players occurs once players have played 100's of games.
Tanks summary of Glicko 1 & 2 and ELO was excellent. Chess sites use slightly differing systems to establish a players rating. A starting point, the ratings of initial opponents and how established they are, are all factors that can make for big swings when 1st playing.
To make 1 exception. At CC 6 years ago when I signed up there several options were available to select my initial rating, on my Android device. Only 1 option was available on my pc. This is not something new. CC utilized many developers for differing OS's and many options were written differently.
Agreed. A players rating varies from site to site. There exists several reasons for this. But maybe my point was misunderstood, which is: The different ratings are not due to how ratings are calculated by a site. Very similar gains or losses of points for established players occurs once players have played 100's of games.
Tanks summary of Glicko 1 & 2 and ELO was excellent. Chess sites use slightly differing systems to establish a players rating. A starting point, the ratings of initial opponents and how established they are, are all factors that can make for big swings when 1st playing.
To make 1 exception. At CC 6 years ago when I signed up there several options were available to select my initial rating, on my Android device. Only 1 option was available on my pc. This is not something new. CC utilized many developers for differing OS's and many options were written differently.
For what it's worth my ratings are within 100 points at Lichess compared to CC. I can be very streaky, especially in 10/0. Over weeks time my rating can +/- 100 points.
I played 1000's of correspondence games at CC and many 100's now at Lichess. My rating is only slightly higher here. I have ideas as to why this is so, but in the end it is only speculation.
Many find their ratings here higher, many find it lower. It may have much to do with the interface, how comfortable players are, lag issues, playing friends, devices used etc; etc.. the list is long.
We see many "new" players saying ratings are inflated at this or that site. Often, these are relatively new members. See what they have to say after 6 months and a 1000 games.
For what it's worth my ratings are within 100 points at Lichess compared to CC. I can be very streaky, especially in 10/0. Over weeks time my rating can +/- 100 points.
I played 1000's of correspondence games at CC and many 100's now at Lichess. My rating is only slightly higher here. I have ideas as to why this is so, but in the end it is only speculation.
Many find their ratings here higher, many find it lower. It may have much to do with the interface, how comfortable players are, lag issues, playing friends, devices used etc; etc.. the list is long.
We see many "new" players saying ratings are inflated at this or that site. Often, these are relatively new members. See what they have to say after 6 months and a 1000 games.
Just for the record, it's lichess.org. Not lichess.com.
Just for the record, it's lichess.org. Not lichess.com.
When people say ratings are "inflated" at this or that site, they are often comparing the ratings to ratings from another pool of players which is outside the scope of the rating system. It is not designed to do this at all, so the comparisons are meaningless. Conversations regarding inflation or deflation relative to other pools of players should be ignored or corrected in this regard.
When people say ratings are "inflated" at this or that site, they are often comparing the ratings to ratings from another pool of players which is outside the scope of the rating system. It is not designed to do this at all, so the comparisons are meaningless. Conversations regarding inflation or deflation relative to other pools of players should be ignored or corrected in this regard.