When I joined with default rating of 1200, initially opponents were pushovers. Now in my last game with an oppo of 1600+ I just barely managed to win (suddenly saw a line to get his queen). I set up 2 traps - he fell in neither (third I made it appear like an oversight on my part by which he would get a piece by a queen fork, and he fell for it)! Even after losing the exchange, he put up a tremendous fight. I was afraid it would go into endgames - but thankfully it did not (he had a rook and a bishop (bishop pair) and two pawns for my queen, then only 1 bishop and 2 pawns, then only bishop). I infer that my true rating may be somewhere between 1600-1800. Am I right, or my self assessment is wrong?

Thanks in advance for any input.

Just realize your Chess.com rating is not the same as your USCF/BCF/FIDE/ICCF, etc. rating.

I agree! Thanks CarlMI

I can't really comment on anything because who knows if your opponents "true" rating is 1600. The more games you play, the more accurate your chess.com rating will become. But CarlMI is right, it only reflects your true chess.com rating.

how much do different ratings normally vary? i.e. chess.com compared to fide rating?

Once you've played a good few games. Lets just say 20, fo the sake of throwing a number out there. And you opponent rating is almost the same as you actual rating. Then it should be accurate. I've played 200 games, my rating has bounced between 1500-1550 for 10 months and my average opponent rating bounces around the same. So, I'm confident that my chess.com rating is between 1500-1550.

If I had looked at your rating just out of noseyness, I would have thought that you were overrated at the moment. But it's early days, too hard to tell.

Under the Glicko system used on this site, you get closer to your "true" rating as your RD gets smaller. If you play more and more games, without significant time off between, your RD will get smaller. And the RD of your opponent helps to avoid the problem of not really knowing what his or her "true" rating is.

Thank you all for your opinions! Housewrecker, I think you are right - my opponent's true rating is probably higher than what he currently has - apparent from his loss-list (he lost to 1700+ and 2000+ oppo).

Your Glicko RD is 155. That means your true rating is likely in the range 1563-1873.

Manchero> i.e. chess.com compared to fide rating?

Correspondence chess and OTB chess differ in too many ways for them to be a useful comparison--time pressure, access to analysis board, access to databases, etc. I know someone rated 1000 points higher here than OTB. But in general, the difference is less than 300 points. If you're training for OTB, you might prefer live chess (30+5).

ive played nearly 200 games i believe my rating is above 1500 but below 1700 at present im about 1580 ,i play live chess in england and am rated about 80 ecf,at best i was just over 100 which is still fairly weak.

I don't really like the fact that some people study theory for 8 hours every day, but when into unknown lines, they bluff. And also when some good players play stupid opening lines. I think that for a true rating, there should be two ratings assigned to each player.

 I had a look at the ECF website but could not find a comversion table. Do you have any idea what 80 ecf is in comparision to other rating systems?

Thanks guys you were awsome!

This was very useful. I was wondering when your rating would start to level out and average.

Just thinking out aloud - Erik, wouldn't it be nice if a guy's rating was written as (1718±165) instead of simply (1718) ? Thanks.

The OTB rating differs from correspondence just because prepered knowledges:

- knowledges in openings overall;

- quality of lines of your opening repertoir;

- amount of memorized endgames knowledges.

- and other things like that. :)

For example one who has little knowledges about this all can have OTB rating much lower then correspondence rating, because in correspondence he is much more relay on his skill and already used to play like this, able to found good opening moves just by skill and long thinking. The same is about endgames. While OTB he has no time for such digging and just overplayed by prepered opponents, even if they are weaker and he can beat them in correspondence. Prepered opponents can have OTB rating much more higher then correspondence because the same reason. But more often reason to have correspondence rating much more then OTB - is just cheating, any kind of it, that cant be performed OTB.

 No it wouldnt, because that 165 also is approximation, it would be like 1718(+-165(+-123(+-87(+-25)))) or something like that, looks not pretty good, right? :) The point is that you can look for number of games played, the statistics of results and opponents and estimate the opponent's RD by yourself, that is enough to know. Even USCF rating is not precise because one 1900 can learn chess 10 hours per day for a month while another 1900 was just drinking all that month. In tournament they both will be rated as 1900 (from previous tournament), while first is probably already 2000 and second is 1800 (for example).

That would be awesome if we got to use differential calculus to find our true ratings.  Sort of like simulating our rating attained so far by graphing which can be done by finding a certain regression.  Then you could find the derivative to approximate the upcoming change in rating.  Really farfetched and awful idea, but it sounds scientific.

x equals negative-b, plus\minues the square root of b-squared - 4ac, all over 2a

No, wait . . it's the other one . .  hmmmm.  Killa, now you've got me all confused :p

There is a conversion table in de la Maza's "Rapid Chess Improvement" (Everyman Chess, 2002, p.13):::

USCF Moniker   USCF     FIDE     BCF

Class E             1000       900       37.5

Class D             1200      1100      62.5

Class C             1400      1300      87.5

Class B             1600      1500      112.5

Class A             1800      1700      137.5

Expert             2000       1900      162.5

Master/NM       2200       2100      187.5