Chess.com ratings realistic or inflated?

Sort:
PsychoNautics

I play live games and correspondence on a few different sites and my ratings are all in the 1500's.. inflated as compared to OTB? maybe.

I personally do not have any intention of OTB chess anytime soon so the online rating is important to me as it tracks my progress.

DonnieDarko1980

Maybe it's got something to do with different time controls:

In correspondence I used to have a rating of 1500-something (on another chess site, not on CC - I don't currently play correspondence).

On CC I'm somewhere between 1000 and 1100 in blitz and between 1100 and 1200 in standard (which means 15-15 minutes for me - I don't play long games online).

I don't have an official OTB rating, but I'd estimate it somewhere around 1300-1400 - in my local club there are mostly 1500s and I mostly lose against them, however sometimes there's a win, at one time a draw against a 1600, ... Well, I'm going to start my first OTB tournament soon, so then we'll see :)

My rating in Fritz rated games (with long time controls) is also around 1300-1400.

So I'd guess, the faster the time controls, the weaker I play. Now I know that there are people where it's the opposite way, great blitz players that are not that good in long games, but if I'm in the majority, that would be an explanation why correspondence ratings seem inflated compared to OTB ratings for some people.

Of course, there's also the fact that the CC rating is Glicko, while the official ratings are basically Elo - but I think you can't say in general that Glicko ratings are higher than Elo?

BalticKnight

Correspondence games generally produces higher ratings. In one way it's easy to explain since the player is allowed to touch pieces, consult books and databases and even if Zeitnot exits it's per move and not game. On the other hand, the sum of the rating in the population should be the same, so there must be a greater inflow in cc than in otb.

erik

we're going to do something about the slowly upward creeping ratings (which are, admittedly, inflated). it is mostly due to people timing out: inactive players timeout and lose rating points while shuffling them up to the active players who gain them and spread them around. my goal is to, over time, reduce everyone's ratings by 2-300 points at the top, and then on down. we will do it slowly and subtly. 

TheOldReb

I believe online ratings are inflated when compared to established otb ratings, both uscf and fide. I have played online on several different sites and on all of them , but one, my online ratings are from 100-300 points higher than my otb ratings.

If someone tells me they are 2800 on some online site I am not impressed , it means nothing to me.

However if someone tells me they are 2800 in otb chess I am VERY impressed and probably talking to one of only 2 or 3 people !

kenneth67
erik wrote:

we're going to do something about the slowly upward creeping ratings (which are, admittedly, inflated). it is mostly due to people timing out: inactive players timeout and lose rating points while shuffling them up to the active players who gain them and spread them around. my goal is to, over time, reduce everyone's ratings by 2-300 points at the top, and then on down. we will do it slowly and subtly. 


I am glad you will be doing this. As mentioned in my forum topic: 

http://www.chess.com/forum/view/suggestions/opponent-time-out

there should be a point system which awards less rating points to default wins if an opponent "times-out" thereby avoiding inflated ratings. Is this part of your plan Erik?

-X-

I don't have an otb rating but I'm ... gonna guess ... 2300ish or so.Laughing

rooperi

Well my highest Chessa (South Africa) rating was about 1450. Chessa seems to correlate pretty well with FIDE. Here I break 1700 every now and then. So, yeah, I suppose it's overrated.

openings_inactive
The questions can be re-phrased as is there a way to compare a 700 player on 500 player on say another scale. For that you will have to know their percentile in their repective pools. Compare a plyer who is member of both pools. And analyse your rating compared to that indicator player. More the number of players in common to both pools,it would be easier to compare and the ratins would have a stronger correlation.
Dragec
Kenneth, did't you get(in your thread) a valid answer why a win by time-outs shouldn't give you less points by default? Some people don't resign, they time-out instead, a win is not less deserved then, and it shouldn't be penalized.
kenneth67
Dragec wrote:
Kenneth, did't you get(in your thread) a valid answer why a win by time-outs shouldn't give you less points by default? Some people don't resign, they time-out instead, a win is not less deserved then, abd it shouldn't be penalized.

I understand that is the rule, so I will have to agree with you then, Dragec.

Kacparov

my rating is already deflated by 300 points and I wouldn'tlike to have it deflate even more :(

TheOldReb
Kacparov wrote:

my rating is already deflated by 300 points and I wouldn'tlike to have it deflate even more :(


 Your otb rating is over 2500 ?! 

Kacparov

2303-1965=338

TheOldReb

where does the 1965 come from ? Its not your CC nor live rating here....

Kacparov

CC

Loomis

Kacparov, it would be very easy for you to keep your chess.com rating from being deflated compared to your OTB rating. Just play your OTB games like they are blitz games, don't ever use more than 5 minutes total in a game. Your ratings will quickly converge to each other

TheOldReb

Here is a GM that has a very deflated rating here .......

http://www.chess.com/echess/profile/LuckyTiger

TheOldReb

Loomis...... did you go to the Duke/Bama game ?  Duke did score a TD , the first one Bama has allowed this season......

Raghav

Hello Everyone,

I believe chess.com is a great site to increase one's efficiency.To know where we stand infront of other good player's.As if you have seen there are many GM,NM and other high rated player sometime's play with lower rated player.So we can get an idea and also which part need to be improved.Comparing rating's of FIDE with chess.com rating,I believe quite is unjustified.

Thanks.