ELO rating ?

Simple question :

Is Chess.com rating = ELO rating ?

Can I claim that my Chess.com rating IS my Elo rating ?

No.

NO !

Then does anybody know how they compare ? Is there a formula (using rational and/or irrational numbers) that converts one rating to the other ? You know - like degrees F to degrees C ...

there is no comparison between on-line ratings and either USCF or ELO rating--much different pools.

Not at all

You might find more correlation with a UFO rating.

 Like what?  So you cant have a run of bad games on one site or lucky break on another?  Nonsense to talk about one example indicating many.

A lucky break or a run of good games won't have a real affect on your long term average. What will make a difference between different ratings other than just the pool of players and the time control is the playing conditions. Some people perform better with a real board and pieces, some do better with a 2D virtual board and for some it doesn't make a difference. Some people don't like the way certain interfaces look or feel -- for example a person might like to click-and-click rather than drag and drop and the option might not be available at some server.

I'm sorry, but I just feel some of the stuff you claim/say is complete nonsense, this is no exception.

You cannot compare ratings between different pools of players with any reasonable degree of accuracy, particually when the game is completely different and correspondence, blitz, bullet and even online standard are completely different from real otb chess. Even between these four rating pools on chess.com there is no consistency. There is really no comparison, and certainly no formula that works even most of the time.

 You are only backing up my claim that one result does not extrapolate into all 1500 players corresponding to 1600 on the other site.

Some people do better with 2D boards, some people do better with a real physical board. In this case, not only does the transition from online to OTB not have the same effect, the effect is the opposite.

I don't think whether an interface is "ugly" matters as much as whether a player is accustomed to a particular interface and can't get used to a new site. Even more important than aesthetics is functionality (which I mentioned in my previous post).

I actually think the atmosphere can make a big difference to many people -- especially people with no OTB experience when they go to play their first few OTB tournaments, and this is exactly the people who are trying to translate their online rating to an OTB one.

Well keep in mind that your ELO rating is only a rough guide of your actual playing strength. It can vary according to opponent quality, playing conditions, general variance, recent improvement or decline in skills.

Comparing one rating pool to another will also have a certain amount of error to it, so by combining the two, the possible range your playing strength can fall into becomes pretty huge. 

You can get a very rough ballpark by comparing, but that's it.

Wow I didn't expect so many replies :)

I've struggled to reach 1400+ on both Blitz and Standard

But the top guys have Elo ratings of 2800+ --- does that mean I'm HALF as good as they are ?

Joooooking !

Some people take it more seriously than others too.  For a while the time I had for some online blitz games was usually right before bed.  I enjoyed the games, but I was pretty tired, I'd play a few then log off.  Now I mostly play for fun when I'm in the mood.  I know some people though take their online games pretty seriously and wont play if they feel like they wont do well... one guy I know warms up with a few tactics first even!

That to me explains why you can have 3 guys rated, say 1800 on site A and they're all probably within 200 points of each other on site B, but it's really not a sure thing.  On FICS or ICC you see sometimes people who have their blitz higher than their standard or vice versa.  Or in their notes you see they're a USCF expert, but rated 1800 blitz.  Then you run into Mr 2200 blitz whose notes also claim their USCF is 2100 and you notice their peak blitz was 2400+ ?!

There's probably a conversion formula but nobody knows it !

e.g. Elo = Chess.com x Sqrt(10) / Pi

or something of that nature

Has anybody even got BALLPARK guidelines

e.g. a small table with ROUGH equivalents ...

Chess.com 1400 = Elo 1350 or 1300 or 1250

Celsus, imagine you collect your data and find that you have 10 guys rated 2000 on chess.com and their FIDE ELO ratings range anywhere from 1600 to 2200.

How does your formula handle that? Do you think it would be possible to say Chess.com 2000 = ELO XXXX?

It was once claimed, years ago, that USCF ratings tended to be around 200 points below correspondence ratings. But I think there is a larger gap now, and that there is a lot more variation, so it is even harder to guess at it today. And as others have pointed out, there are many different styles of play, none of which simulate the type of game you play to establish your ELO.

Thanks Mouse for the rough guideline.

Loomis I would have thought that 10 guys with similar ratings on Chess.com would in reality have a much narrower range of Elo ratings.

Yes I would think it possible to say Chess.com 2000 = Elo XXXX plus or minus YYY where YYY is not such a large number.

Chess.com faq has information on how they derive their ratings, I believe off the top of my head they use the glicko, not ELO.

You may be surprised how not similar the elo ratings will be.

Here's a (possibly extreme) example. Chess.com user Kacparov is a FIDE Master, his FIDE elo is 2291. His Chess.com rating is 1941. My USCF rating is about 1900 -- that is, I am not as good a tournament player as Kacparov, if I had a FIDE ELO, it would be much lower than 2291. However, my Chess.com rating is over 2200 -- 300 points higher than Kacparov.