when dealing with the chess.com the rating system is unfair on players. take this for an example you could be playing and offer a draw in which you would gain rating, then 2 days later you beat someone and your rating sky rokets then your opponent accepts draw and then you lose alot of rating. is this unfair should we make it that when you play an opponent your ratings are locked and cant change untill the end. would this make it more or less fair? 

If you take a look at the formulas used you'll see that it won't actually matter much if you had gotten the extra score from the draw before you got the score from the win the later score would be lower than in your scenario and your final score would be just about the same. 

don't take ratings seriously.....come on I am a lousy chess player who play like crap. And I am rated 1833 !!!

 You can be serious with that can you !

Everything is fair. Chess.com use Glicko-2 rating system.Write in google Glick-2 and read something about that rating system.

Take your rating to the bank and see how much it is worth. Why not concentrate on chess instead? If you have a strong interest in numbers, I'm sure there are some lovely math sites with all sorts of numbers for you to enjoy.

actually, a rating is supposed to be an accurate representation of your rating. and the most accurate representation is the most RECENT representation, which is why we use it.

we used to have it the other way. then people complained so we changed it. now we have it this way, and some people don't like it. we can't make everyone happy :) sorry! 

gateman234 wrote:

when dealing with the chess.com the rating system is unfair on players. take this for an example you could be playing and offer a draw in which you would gain rating, then 2 days later you beat someone and your rating sky rokets then your opponent accepts draw and then you lose alot of rating. is this unfair should we make it that when you play an opponent your ratings are locked and cant change untill the end. would this make it more or less fair? 

---

 I hate chess.

I wonder how accurate these ratings are , I know a 1100 Fide player rated at 2000 on a correspondance chess website. Maybe chess.com should run a poll for those players who have otb ratings to see how close their chess.com rating is to their otb one .

good idea! new poll coming...

You will probably find that the OTB rating will be less than for CC. The reason for this is that you have much longer to think about your moves. In otb you have on average 3 minutes to make a move and the game will last only a couple of hours, maybe 6 depending on what time control you are using. In CC you can have hours to make a move, consult opening books, look at game databases and, dare I say it, consult computer programs. But the poll does sound interesting. I hope we have a lot of otb players otherwise it will be mis represented.

Red

Redserpent, so everybody should have a higher CC rating than OTB rating?

 

The bottom line is that there is very little meaning in comparing ratings from two different rating pools. The accuracy of chess.com ratings only pertains to how well they compare two players playing at chess.com.

 

Because games can last so long in CC chess, I think a system of re-rating games could provide the most accurate ratings. Why is a game that finished today more accurate than a game that finished last week when it's possible that the game lost today was due to an error made a month ago? It would not be easy to figure out what is the best way to generate ratings in this environment. One issue is that a system where games get re-rated might be more accurate, but it would be less transparent to the user what is going on. Despite the fact that it might be possible to generate more accurate ratings, it's probably not worth confusing the users. :-). 

basing ratings changes on the most recent score makes sense.  your score should get more accurate the more games you play, thus making appropriate adjustments for the present moment

But ratings are comparitive, not absolute. So you can't have everyone's rating being higher.

 

sirsc: "basing ratings changes on the most recent score makes sense."

 And it probably does is the most accurate system that is also transparent to the user. There are probably more accurate systems that are also more mysterious.

Your blunders are reduced and your analysis is improved.  Your opponents blunders are reduced and their analysis is improved.  It shouldn't make any difference. 

My rating here is higher than it 'should' be.  Or maybe my OTB rating is lower than it 'should' be.  I tend to think the latter since I am in UK and British grades are notorious for lagging - it can be that your current rating is based largely on results from 3 yrs ago.  3 yrs ago I was not very interested in chess and often took draws in better positions or even resigned in drawn positions to avoid another evening playing an adjournment.

I have seen other players post the exact opposite, i.e. their rating here is lower than OTB. 

 

Whether your rating will be lower or higher on chess.com versus OTB depends on what you do with the different conditions.

 

If in OTB you tend to think too long and end up blundering in time trouble, that should be much less of a problem here. However, playing a complex game with one move every few days it's easy to forget some of your analysis. Playing many games simultaneously on chess.com can also lead you to take each one less seriously and do less analysis compared to an OTB tournament game where you are investing hours of your time against a single opponent and are much more likely to be focussed on that game.

 

If opening theory is a strength of yours OTB, then that advantage may be nullified here if your opponents is using a reference source. Of course, if you or your opponent play "non-book" openings that will be less of a factor too. 

 

So the definitive answer is ... it depends.  ;-)

 

 

 

Is it ok to use opening references while playing on here?  If so, I had no idea.  That may seem silly, but I didn't know what would be considered cheating.

Irrespective blunder reduction with longer times and analysis and books and the like, the ratings are MUCH higher here than in the uk at least.  1100 players otb can play a fairly solid game, and certainly would not commit the same mistakes as players do here.  Even some 1500 players on here can hang multiple pieces or miss forks even only one move away.  A uk 1500 player in real life would probably grind me down steadily for a win, whereas here he would be more likely to throw his pieces away or play f6 in the opening or something.  In otb chess you would have to work for a rating of 1200 in the uk by beating people who have probably more experience than you have when you are u/g, but here you are given a provisional 1200 even if you only just learnt how the pieces move and it seems people around this rating sometimes play almost completely randomly.  I think there are going to be more total beginners on a chess website like this, than in an otb chess league and this is bound to affect the lower end of the ratings scale.  I think also this is difficult to compare because USCF starts players off at 1200 also (i think?) whereas provisional rating in the uk is 1000 so maybe this website is similar to how grades are otb in the us. 

 

On the subject of people playing better on chess.com with all the time and analysis and whatnot, I find, and i'm sure i'm not alone, that i play better otb provided i have a decent amount of time, than i play when i have too much time, and when i have to come back to each position 'cold' and perhaps having played other games for a few consecutive moves in the mean time, and maybe i havent seen a game for a week and then have to remember what i was doing.   

Reb wrote:  

---

I dont think fide ratings go that low. Please tell me a player with 1200 or less fide rating? I think fide only publishes 1600 and up?

---

FIDE now goes down to 1400 but they themselves admit the ability variation at that level is massive.

I don't buy into the point that the rating is unfair, the Glicko system is probably the best method to grade people because it takes into account game activity. A player rated 1500 who has played 2 games in the last two years is likely to vary in ability much more than a seasoned, 30 game a season 1500.

A rating, afterall, is a number which reflects your current ability in chess. I wonder whether some people treat the rating as an outward measure of their self-worth.