Does anyone feel there rating is inflated, Although i've only completed 9 games. I feel as though my rating is inflated.
My impression is that, on average, the turn-based ratings are very inflated, but the long live chess ratings are pretty comparable to USCF and the live blitz are lower than USCF by around 100pts.
A chess.com rating can only be said to be inflated if compared to a chess.com rating from the past.
Does anyone feel there rating is inflated, Although i've only completed 9 games. I feel as though my rating is inflated.
My impression is that, on average, the turn-based ratings are very inflated, but the long live chess ratings are pretty comparable to USCF and the live blitz are lower than USCF by around 100pts.
I've observed this as well.
I'd say for 90% of people this formula will be true.
USCF - chess.com CC - add 150-250 points
USCF - chess.com Live long - add 1-50 points or subtract 1-50 points
USCF - chess.com quick or blitz - totally indeterminable
TadDude - surely everyone could potentially be at precisely the correct rating? Why must everyone be under/overrated?
Here is an explanation of Glicko and RD. http://math.bu.edu/people/mg/glicko/glicko.doc/glicko.html
Relevant excerpt "Because a player in the Glicko system has both a rating and an RD, it is usually more informative to summarize a player's strength in the form of an interval (rather than merely report a rating). One way to do this is to report a 95% confidence interval. The lowest value in the interval is the player's rating minus twice the RD, and the highest value is the player's rating plus twice the RD. So, for example, if a player's rating is 1850 and the RD is 50, the interval would go from 1750 to 1950. We would then say that we're 95% confident that the player's actual strength is between 1750 and 1950. When a player has a low RD, the interval would be narrow, so that we would be 95% confident about a player's strength being in a small interval of values."
Currently your rating is 2382 and RD is 92. So you can be 95% confident that your actual strength on the site is between 2198 and 2566.
Yes yes yes, I'm well aware how Glicko works. I can be 95% confident it's in that range, but I may in fact genuinely be 2382 ;)
Okay I will cede the point, but after your next game you will either be under or over rated :)
Does anyone feel there rating is inflated, Although i've only completed 9 games. I feel as though my rating is inflated.
My impression is that, on average, the turn-based ratings are very inflated, but the long live chess ratings are pretty comparable to USCF and the live blitz are lower than USCF by around 100pts.
I've observed this as well.
I'd say for 90% of people this formula will be true.
USCF - chess.com CC - add 150-250 points
USCF - chess.com Live long - add 1-50 points or subtract 1-50 points
USCF - chess.com quick or blitz - totally indeterminable
As opposed to using the inappropriate term "inflation" [which I and others have casually done], I like to use the above idea of equivalents. It's rough justice but I'd expect to see an equivalent display of skill between a 1600 USCF and an ICC 1750 in long standard play. Rough indeed but there's not much else we can say when comparing different rating pools and widely differing TCs, the skill sets for them, and how radically wide the range of seriousness is for the varying types of online play, especially vs. organized OTB.
Why bother comparing it anyway? This is one rating for this site only and O.T.B ratings are different again, even in each individual countries rating systems. Correspondent/turn-based online ratings are different from one site to the next, so why not just compare within this sites ratings which is plenty as is?
If u want to compare O.T.B., play the tournaments and compare them to other O.T.B ratings such as ur national ratings to the F.I.D.E. ratings.
Just my thoughts on this issues of always comparing ratings from other places to this one.
Personally I do not have much interest in comparing my own rating to any online ones, but as the question comes up here quite a bit, it's handy to have some form of an answer.
Yours is a good one too.
Any reasonable rating system (Elo,Glicko,BCF etc.) becomes less accurate as the number of players (in the pool to be rated) increases.
Why bother comparing it anyway? This is one rating for this site only and O.T.B ratings are different again, even in each individual countries rating systems. Correspondent/turn-based online ratings are different from one site to the next, so why not just compare within this sites ratings which is plenty as is?
It's only natural to want to compare I suppose. I haven't played a rated OTB tournament in years, but I am sometimes curious to know how strong I'm playing in USCF ratings terms.
Some people deny that, for instance, ICC or Chess.com ratings can be compared to USCF or FIDE, but I really don't know why that would be the case. They all work on pretty much the same system (e.g. a 100pt rating difference between players should average to an appx. 2 to 1 scoring probability for the stronger player), and a simply formula correcting for the difference from one rating pool to another should be adequate to give a rough approximation.
Any reasonable rating system (Elo,Glicko,BCF etc.) becomes less accurate as the number of players (in the pool to be rated) increases.
Are you sure they don't become more accurate?
Live games provide better competition than most sites but c.c. games are inflated.
haha
Any reasonable rating system (Elo,Glicko,BCF etc.) becomes less accurate as the number of players (in the pool to be rated) increases.
Are you sure they don't become more accurate?
Yes. The greater the number of players in the pool,the less likely it becomes that any two players will ever actually play one another.
Does anyone feel there rating is inflated, Although i've only completed 9 games. I feel as though my rating is inflated.
My impression is that, on average, the turn-based ratings are very inflated, but the long live chess ratings are pretty comparable to USCF and the live blitz are lower than USCF by around 100pts.
I've observed this as well.
I'd say for 90% of people this formula will be true.
USCF - chess.com CC - add 150-250 points
USCF - chess.com Live long - add 1-50 points or subtract 1-50 points
USCF - chess.com quick or blitz - totally indeterminable
this may be right but i know a player on chess.com whose USCF rating is about 200 points higher than his online rating of 1900 or so. i suppose it's different from player to player. i think a lot of it depends on who you're playing and how often you're playing.
What does less accurate even mean in this context? It's a self referential measurement that tells you how you stack up compared to other players who've been rated in the same pool. It's not like there's a constant measuring stick that can be used to gauge how accurate it is from one point in time to another. It simply is.
Think of a meter (OK, yard....) stick that doesn't have measurements on it but has tick-marks labeled 0%, 25%, 50%, 75% and 100%. Whether the stick is actually a meter long or a mile long doesn't mean that those marks are any more or less accurate. Now add a bunch more marks (i.e. players in the pools) so that it reads 0%, 1%, 2%, 3%.... all the way up to 100%. Again, the accuracy of those marks is entirely independent from the size of the stick. Also, by adding marks we haven't added accuracy at all -- 25%, 50% and 75% are all right where they were before -- but rather granularity.
My rating then must be very inflated.
I had an otb rating of 1646 when i stopped playing tournaments in 2004. My turn based rating here is over 2200+ which is a 600+ difference. So either i got better or ur estimation cannot be true and it's still just something u cannot define accurately enough to compare.
Does anyone feel there rating is inflated, Although i've only completed 9 games. I feel as though my rating is inflated.
My impression is that, on average, the turn-based ratings are very inflated, but the long live chess ratings are pretty comparable to USCF and the live blitz are lower than USCF by around 100pts.
I've observed this as well.
I'd say for 90% of people this formula will be true.
USCF - chess.com CC - add 150-250 points
USCF - chess.com Live long - add 1-50 points or subtract 1-50 points
USCF - chess.com quick or blitz - totally indeterminable
this may be right but i know a player on chess.com whose USCF rating is about 200 points higher than his online rating of 1900 or so. i suppose it's different from player to player. i think a lot of it depends on who you're playing and how often you're playing.
Well I left a 10% margin for that. In my case my live chess rating and CC rating are almost the exact same (though my CC should be 1900 or so pretty soon after I finish up a few more games), the reasons for this are varied.
What does less accurate even mean in this context? ............
It means that the greater the number of players in the rating pool, the less likely it is that a rating difference of 'x' will correctly predict the outcome of a game yet to be played between two players who have never played one another.
It'd be nice to be able to come up with some estimate of what the site thinks your OTB elo would be based on your online rating. It'd be possible you'd just have to collect data on member's actual USCF/FIDE/ etc. ratings and do some statistics.
Although, I imagine that how they correlate might vary by level, e.g. there are alot less players represented in the chess.com rating pool at the higher ratings then there would be in the USCF rating pool.
Does anyone feel there rating is inflated, Although i've only completed 9 games. I feel as though my rating is inflated.
My impression is that, on average, the turn-based ratings are very inflated, but the long live chess ratings are pretty comparable to USCF and the live blitz are lower than USCF by around 100pts.
A chess.com rating can only be said to be inflated if compared to a chess.com rating from the past.
I'm just using loose language - taking USCF as a norm and comparing another ratings pool to say that the ratings there are "inflated" as compared to USCF.
I'm not speaking of the phenomenon of ratings inflation where the ratings of players at a given strength increase without their actual playing strength increasing.
At correspondent chess maybe u r a master level who knows.