- Blind mode tutorial
lichess.org
Donate

What is the highest rating possible?

#1

Does anyone know what the highest rating possible is?

Does anyone know what the highest rating possible is?
#2

I don't think there is any „Rating Cap“ so to speak, which means theoretically one could go all the way up to 10.000 rating and beyond. However in practice nobody has even reached 3500 in any rating (legit) yet, and with deflation being a thing, it is likely no one will ever reach such ratings.

I don't think there is any „Rating Cap“ so to speak, which means theoretically one could go all the way up to 10.000 rating and beyond. However in practice nobody has even reached 3500 in any rating (legit) yet, and with deflation being a thing, it is likely no one will ever reach such ratings.
#5

Kenneth Regan is a computer scientist and FIDE's advisor on fair play and he once put together a model that said a perfect rating would be about 3600, although I think I remember an interview on Perpetual Chess last year where he modified that. Here's the article about his work from 2014: http://www.uschess.org/index.php/June/How-To-Catch-A-Chess-Cheater-Ken-Regan-Finds-Moves-Out-Of-Mind.html

Kenneth Regan is a computer scientist and FIDE's advisor on fair play and he once put together a model that said a perfect rating would be about 3600, although I think I remember an interview on Perpetual Chess last year where he modified that. Here's the article about his work from 2014: http://www.uschess.org/index.php/June/How-To-Catch-A-Chess-Cheater-Ken-Regan-Finds-Moves-Out-Of-Mind.html
#6

When you win you add rating points, with the number of points being determined based upon the difference between your previous rating and the opponent's rating.

If you become better by far than anyone else on Earth (including computers) it seems as if the increment added to your rating would gradually approach zero as a limit. And, indeed, if only integer increments are permitted, it might become zero, unless "1" is a minimum by convention.

That could be looked up. But I have lunch to eat and dogs to pet.

So -- if the minimum increment is 1, then rating is limited only by lifespan.

If the minimum increment is 0, then rating will peak at some point.

If increments can be real numbers less than 1, then rating will approach a peak asymptotically.

As a practical matter, though, it seems that rating DOES have a peak (although that could increase if genetic engineering is taken over by dark forces and bent to their selfish will decades from now).

When you win you add rating points, with the number of points being determined based upon the difference between your previous rating and the opponent's rating. If you become better by far than anyone else on Earth (including computers) it seems as if the increment added to your rating would gradually approach zero as a limit. And, indeed, if only integer increments are permitted, it might become zero, unless "1" is a minimum by convention. That could be looked up. But I have lunch to eat and dogs to pet. So -- if the minimum increment is 1, then rating is limited only by lifespan. If the minimum increment is 0, then rating will peak at some point. If increments can be real numbers less than 1, then rating will approach a peak asymptotically. As a practical matter, though, it seems that rating DOES have a peak (although that could increase if genetic engineering is taken over by dark forces and bent to their selfish will decades from now).
#7

https://i.imgur.com/Og1kDJn.png

Think again lads. (Yes those accounts are closed, but with manipulation, levels above Leela's 3200 on this site are possible)

https://i.imgur.com/Og1kDJn.png Think again lads. (Yes those accounts are closed, but with manipulation, levels above Leela's 3200 on this site are possible)
#8

It is based on winning probabilities, it depends on the pool. If you don’t know now you know.

It is based on winning probabilities, it depends on the pool. If you don’t know now you know.
#9

Indeed, @Sarg0n , the relative ELO ratings of two players DO carry with them a determinable probability that one of them will win. And yet that lovely fact does not address whether or not ratings have a "peak." Indeed, what that "peak," if any, is -- 3,500 or even more (as was well demonstrated to be possible by @Leux_OW) -- is even a third question yet!

But thanks for reminding or (for some, no doubt) teaching us that ELOS can be examined mathematically in the wonderful light of probability.

Indeed, @Sarg0n , the relative ELO ratings of two players DO carry with them a determinable probability that one of them will win. And yet that lovely fact does not address whether or not ratings have a "peak." Indeed, what that "peak," if any, is -- 3,500 or even more (as was well demonstrated to be possible by @Leux_OW) -- is even a third question yet! But thanks for reminding or (for some, no doubt) teaching us that ELOS can be examined mathematically in the wonderful light of probability.

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