If a player participates in both blitz,standard,and online chess,there should be a pattern to their ratings,they should increase with increasing time limits and should have roughly the same ratings difference between time limits.For example,a grade difference between each type of chess between blitz,standard and online.When there is a huge difference between standard and blitz on one hand,and online on the other,then the online rating becomes a bit suspect;It's hard to fathom someone who can play online chess at a 2400-2500 elo level,yet can't rise above 1700 in blitz.In the cheaters forum thread there is a list of banned players,a very long list,many on that list fell into the above described pattern.
Why Are Correspondence Ratings So Inflated?
Sort:
Jan 23, 2014
0
#23
I beg to differ that Internet Blitz is severely deflated because everyone starts at 1200, and with the high mass volume, everybody is underrated, and so ratings stay low:
Case in point:
USCF Regular - 2094
USCF Quick - 1963
USCF Blitz - 2069
USCF Correspondence - 2018
FIDE - 2071
Chess.Com Correspondence - 2038
Chess.Com Blitz - Under 1700!
badger_song wrote:
If a player participates in both blitz,standard,and online chess,there should be a pattern to their ratings,they should increase with increasing time limits and should have roughly the same ratings difference between time limits.For example,a grade difference between each type of chess between blitz,standard and online.When there is a huge difference between standard and blitz on one hand,and online on the other,then the online rating becomes a bit suspect;It's hard to fathom someone who can play online chess at a 2400-2500 elo level,yet can't rise above 1700 in blitz.In the cheaters forum thread there is a list of banned players,a very long list,many on that list fell into the above described pattern.
But I have a Rapid or Blitz - not sure which one - 10 min control of 1100 (I didn't mean to play this game - I thought I was signing up for 30 MIN) My standard is 1400 - 300 points difference.
I will probably never play 10 min again - at least not in the near future. If things go as planned - I will be a 1600 player by the end of the year (I hope that is realalistic) that is a 500 point spread -
This means my online rateing could be as high as 2100 - useing your equal spread therory - there is no way in the world I could play 2100 level chess - even if you gave me a decade per move.
Sorry - I do not think that therory works as an indicater.
More likely to me - is when you run the game back through Fritz or whatever - if the opponinete always makes the highest rated move - he is cheating.
But then again - is the highest rated move really always the best? I read the notes of some GM games against computers - and they showed how the computer thinks vrs how a top level human thinks, thier conclusion - a Super GM will always beat a computer because of human creativity.
Jan 23, 2014
0
#25
Raymond,it's not a theory,nor is it predictive,the ratings red flag is raised when the pattern,as described, is present,and the player is active in the various catagories as listed.Furthermore,those players banned for cheating,who indeed are active in the various catagories of chess offered by chess.com, usually follow the above described pattern.It's just one of many factors that eventually cause site admins to start looking at a player's game history and examining said games,persumably with chess engines.What exactly was your objection again?
Badger_Song: I guess the key phase there would be "active". If I have only one 10 min game, that does not make me active in the catagory - esspecially if I have say, 100 games in the standard and online catagory.
Therefore, the 10 min catagory would have to be dismissed since there is no game trail there to follow.
But know for the million dollar question: Why would you cheat?
What do these people gain from this. No knowledge is gained - the computer did all the work, No titles are gained, and No monies are earned. What is the point? If you play by yourself - and the allowed material - you at least walk a way from a game with more knowledge. You have "earned" something for your time.
I simply do not see what is to be gained by cheating in a chess game -
Jan 23, 2014
0
#27
Cheating has little to do with the reasons you list,the most common reasons for the type of cheating that is seen in chess and other games, are purely psychological,and have much to do with self-image as percieved by others.The primary goals are subjective things,like group approval and fame,titles ,victories and prizes won are just a means of getting there.
Jan 23, 2014
0
#28
There is no such thing as a "1700 player" or "1700 play".
Ratings can't be compared across different player pools. Inflation/deflation refers to variation in ratings across time within a single player pool.
Get it straight, guys.
The misconceptions of people and self image. I recently played a game with the guy who taught me how the peices move and some fundamentals - after the game he said: "Your game has really improved, why don't you come down to the club this Thursday? I will introduce you around."
This made me feel good - I put alot of effert in to this over the months. And now, it was being acknowledged by "acceptance". That seems to be good self image to me.
Being listed in a fourm topic as banned for cheating - seems to be the opposite of acceptance and positive self image among the group - Heck you have been completely outcast by the group.
Oh well -
I guess it depends on what you are looking for out of chess. I just want to play a good hard fought game - win or lose, I want you to say: "That was a tough game."
It's not clear to me that cheating would necessarily raise the average rating. Cheating would raise the ratings of those who are cheating, but that would depress the ratings of the other players. So the average should stay about the same. The average of the active players might go up or down depending on the relative ratings of those players that quit playing on a regular basis. If all the low rated players quit playing, then the average of the active player pool would rise, whereas conversely if high rated players quit, then the average would decline.
Jan 23, 2014
0
#32
Ziggyblitz wrote:
I've read that Online ratings are inflated by roughly 300 points compared to OTB. Seems about right to me.
mines is opposite lol
This is the difference between a more diverse and an inflated pool ( compared to the original pool):
Diversity has nothing to do with inflation and inflation has nothing to do with diversity.
sammynouri wrote:
Where did you get the plots from?
I have created them with Wolfram alpha based on the formula of the normal distribution:
pt22064 wrote:
It's not clear to me that cheating would necessarily raise the average rating. Cheating would raise the ratings of those who are cheating, but that would depress the ratings of the other players. So the average should stay about the same. The average of the active players might go up or down depending on the relative ratings of those players that quit playing on a regular basis. If all the low rated players quit playing, then the average of the active player pool would rise, whereas conversely if high rated players quit, then the average would decline.
Cheating can 'deflate' or 'inflate' the pool depending on circumstances (mainly how many games they are playing when they are banned). But in almost all cases I would guess it 'deflates' the pool. A player removes points from other players, and is then banned, and the rating points are therefore removed from the pool. This would cause 'deflation'. However, if they are playing enough games (also to an extent based on the RD of opponents), it is possible (but unlikely), they will put more than the amount of rating points that they have taken back into the pool, and therefore inflate it.
None of the arguments such as time taken on move, tools used etc are convincing. That just distributes the points differently, it doesn't inflate or deflate the pool as a whole. It's slightly more complicated than this and could have a slight effect, but not significant.
The most significant factor for me would be people who join, start games and timeout, and are then removed after closing their accounts.
watcha wrote:
sammynouri wrote:
Where did you get the plots from?
I have created them with Wolfram alpha based on the formula of the normal distribution:
Great site. Used to have them in my favs, but removed it because I never use it.
We all know cheating exsists - but I wide rating spread doesn't indicate that cheating is taking place.
There could be many factors:For young players, like myself - I have only been playing chess 5 or 6 months now, the pressure of OTB kills us. We get in time trouble alot. We become aware of time troubles in the past and don't look at all the variations. Then, Whammm, Blunder or obvious wrong move.
I do not know how many games I have gone back and relooked at and felt like running my head into a brick wall - "how stupid can I be!?!"
Young players who are serious about learning this game (being a 800-900 and having people score minitures on you is not my idea of a fun hobby) have a lot of material, books, databases, manuscripts, videos, GM games, we have the infomation - we just do not have it at instant recall in our minds yet. Give me 3 days to find a move on the other hand and I probably find a much better move than I would OTB - plus the time difference gives me time to formulate a stredegy - not something easy for a young player to do OTB.
I also think not everyone approaches correspondence the same. For me it is a great learning tool - I love looking at all my canidate moves, and exporing every variation on each one before I make a move - it gives me times to really study positions. I think a lot of correspodence chess plays play more instintivly as if they are playing OTB.
I do not have the expereince that most everyone here has, so I maybe treading in deeper waters than I want - but these are a few of my observations - but again 5 months is not enough time compared to those of you with life long expereince.
Raymond
I think you hit the nail on the head, I am a younger player myself (well if you can call University age young) but I still play at the speed say a real young gun would. Same energy and passion, OTB and online. I hardly use any time at all, and the funny thing is, I do well against people who do use lots of time to calculate and analize. But to adress the nerves you have, calming down is important, find something that loosens you up, and centers yourself so that you can play really well, be it get to sleep earlier the night before a tournoment, listent to music, etc.