Inflated Rating

 Beyond the fact that you haven't shown the statement to be false, how does believing the ratings to be accurate a serious impediment to the improvement of chess skills?

But everyone gets the same advantage from the additional time then.  The games might be of higher quality, agreed, but this won't manifest itself as higher ratings overall.

If I make better use of that additional time than others (and given my slow time-per-move stat it's arguable that I do compared to someone with a very quick time-per-move) then perhaps my rataing will be slightly higher than it might be in another pool, but it stands to reason then, that the individual who moves very quickly will have a lower rating than he might in another pool that has less opportunity for this type of disparity in usage patterns.

At the end of the day it just underscores the fact that you can't reliably compare between pools.

That's for Vote Chess. This is just a guess based on my observations as a player here.

 shak: I don't know what you mean when you see you don't believe it to be true.  With a large enough sample size, the ratings should accuratly reflect relative strength of player (within the same pool).

Of course no one is arguing that the lower rated person cant beat the higher rated person, just over many games, you would predict the higher rated player to win more (the percentage would be relative to the rating disparity)

As for the 'impediment' part.  I believe his point is that the lower rated player plays with less confidence against higher rated players., i.e they play not to lose rather than playing to win.  I don't know how serious an impediment this really is, but it is possible it has some effect.

My rating is deflated at the moment.

The ratings here may be a little high compared to USCF, but not wildly so. Historically I am a high 1800s/near 1900 player in tournaments. Here I vary between 1900s and over 2000. Then again, I play better chess here, having the advantages of more thought time, less pressure, less need for physical stamina, ability to use Game Explorer and openings books in the opening, etc. So I find the rating commensurate. If I could play the same quality of chess I play online in tournaments, I probably would be at USCF 2000.

Glickman conquers dot.com land, Australia & USA.

...we shall fight Glickman on the beaches, we shall fight Glickman on the landing grounds, we shall fight Glickman in the fields and in the streets, we shall fight Glickman in the hills; the ECF & FIDE shall never surrender...

Average rating stats for the various chess.com playing versions are shown on this site, the USCF all player ave is a little lower, USCF tourney ave slightly higher, ECF & FIDE ave a bit higher still, so there might be some form of comparison possible as might be expected.

The weakness of chess.com ratings is that if a player loses little or not at all the grade goes far too high. Sorry to finish on a negative I must be forgetting the sandwich communication technique.

>:)

To compare ratings I think one should first find a federation which allows as much cheating OTB as chess.com.

Just report those that you consider cheaters. Let your engine run a deep analysis on the suspect games and send it along - like I care. But you should really stop moaning about cheating everywhere.

Cheating is a serious issue, but it isn't the current topic.

 Please read the statement in question carefully.   If it were true, no player's rating would ever change significantly.  Ratings are basically records of past performance. 

Improvement, on the other hand,is all about a demonstration that a player's established rating is not only inaccurate,but also too low to reflect his/her current strength.

If a player believes that his/her current rating is 'accurate', does that not imply that said player has lost the will to improve?

Er, no it doesn't mean that at all. Rating is a reflection of past performance... up to the current day, so in the present it is more or less correct. My current rating roughly approximates my current chess skills. This in no way means that I'm not seeking to improve in the future, or even now rather. A user may have a current rating of 2000 (say) which is a correct approximation, but studies hard and hopes to be 2100 in a few months. This doesn't equate to believing you are 2000 in the present means you don't want to improve.

If a sprinter concedes that his time in the 100M is accurate does that mean that he has no will to improve?

The perceived fidelity of the measurement device has absolutely no bearing on willingness to improve your performance against it.

 Are you 14 rating points stronger today than yesterday? If not, then your rating as of yesterday was inaccurate.

Oh yeah? And what happens after someone is reported? 

As for being on topic, the top players here have 2800 CC and 0 OTB. What relation between ratings are you seeking?

 I'm sure that the benefits of diamond membership are very fine and well worth the expense.

However, at last report, they do not include unlimited access to the OTB ratings of everyone at Chess.com.  Nor do they include the right to control who is and who is not allowed to be included in the Chess.com rating lists.

Can't you read? I said a correct approximation. This doesn't mean it's correct down to the last point, no. To take my case - my rating fluctuates round and about the 2400 mark at the moment. In some months I hope it will fluctuate around the 2500 mark. Geddit?

@ shakmatnykov

I'm curious what your thesis is here exactly. If it is to point out that ratings are not perfect predictors of future performance, well, no argument here. Players may improve or fade, or have (un)lucky streaks, or one player may have another's "number", in national ratings pools there may be geographically isolated groups where players mostly play only amongst themselves and then have ratings that are dissociated from the larger pool etc. There are a whole host of factors that can potentially skew players ratings, so the knowledgable person looks at another's rating and understands it is just an approximation.

But I get the impression that you are saying something beyond that and think ratings are all but useless for predicting future performance.

 As you correctly point out,a chess rating is not a definite measurement of chess skill that has the same 'weight' in all locales within a given pool. My first remark was directed at that fact, and also toward the fact that as the number of players in the pool rises, the factors that you have mentioned also increase.

I am ,as you have noted, now saying something beyond that,which has more to do with the mental attitude of players who desire to improve both their chess skills and their ratings.

Such a player must continuously endeavor to 'prove' that his/her current rating is,in fact,too low to accurately reflect his/her current playing strength.

I've already pointed out why this isn't necessarily correct.

 Yes,but since I can't read, I was not aware of it.