Author: Brian Katz
Date: 10:17:24 05/26/02
Go up one level in this thread
On May 25, 2002 at 16:37:46, Jeroen van Dorp wrote: >>Let's not forget that correspondence ratings are probably inflated a lot. >>The best players at this specialty probably never played a single game. >>For instance, I have no doubt in my mind that Kasparov will eat alive any >>correspondence player by a huge margin. What would be the rating of Kasparov if >>it decides to really go into it? 3000?. IMO, They are just from different >>levels. It is very often to see correspondence ratings that are 200 points >>higher than their respective FIDE rating. > >I partially agree. > >I think "inflated" means "higher than you should expect based on average >performance". ut that can never happen, as rating is "related to the >opposition". > >In that case there is no inflation for any chess form, including correspondence >chess. That your CC rating is higher than an OTB rating, is not necessary - each "Pool of players" is fine point. If you were over 2500 strength at real over the board tournaments such as grandmaster and Super GM's and all you played against on a chess server was 1300-1400, it would take quite a long time befor your rating raised to a much higher level. Once your rating is 350 - 400 points above the opposition you gain from ZERO to ONE point. So, that being said, an 1800 server rated player who is really a GM (2500 and above) would have to win HUNDREDS of games. The math is not that hard to calculate. 1800 to 2500 is 700. One would have to win at least 700 straight games to get his rating where it really should be. Of course you would not even gain a full point in every win, so it would be even more than 700 games. AND if god forbid you lose a game by being outplayed or due to an unfortunate disconnect, then you will lose 32 POINTS, setting you way back again. Then there is always the possibility that you will be playing a 1400 who is really your strength, (2500) who is in the same boat you are in, and you lose, and again there go your points. POOL of PLAYERS is definitely a big factor. For a player to take the Server ratings seriously is a big mistake. There is always the possibility that on a given day a 1400, 1500, or 1600 etc. player will play against a real 1400 strength player who decides to use a PROGRAM such as FRITZ 7 and loses badly. Now this poor player who is let's say 1600 gets crushed by his supposed 1400 opponent, doubts his own playing strenth. I myself have given up playing rated games for the most part on ICC. I am a real 1930 to 2000+ USCF rated player, and have been over 2000 for many years, (Sometimes I dip below like I did 2 years ago to 1920 due to a horrendous tournament, presently 1930, and have not played in more than 3 tournaments since then, to bring my rating back to normal. In any event, I have given up on rated server play and only play unrated. So please any of you lower rated real tournament players just starting out, do not use the server Ratings as an accurate monitor of your real playing strength. Use your over the board results as your true test. Otherwise some younger players can and will get real discouraged and that could have a disastrous effect in some cases. Now that I have rambled on endlessly, and bored a few if not many, I would like to hear some views on this, pro or con. Regards Brian discipline has it's experts - but not so illogical. > >If I play on an online server, and restrict myself to playing 3,0 or 5,0 blitz, >my rating on that server is like 1300 or 1400. When playing longer time >controls, but also rapid play like 20 minutes a game, my rating is already >1600-1650. When playing OTB with tournament controls, my rating rises to >1750-1850. When playing correspondence chess, I have a rating on that particular >server of 2000-2100. > >The "progress" is almost linear and related to increasing use of time. Of course >it is quite logically that -when having more time to think and thus more time to >analyse- strenght will improve until the point of your "natural maximum >strenght". > >As rating is always something "related to the pool of players", a CC rating >can't be compared to an OTB rating. >As you state it yourself, that even goes for Kasparov. > >The only reason for applying ratings is to get an impression of strenght of an >opponent in the pool you play. > >I even don't know if I could win playing against an OTB 2000-2100 player, he or >she playing with tournament time controls, me playing at CC time controls. Those >two ratings are maybe not from the same "pool of players". And maybe if they >were, not everyone might have the same rise in strenght with prolonged time >controls - ratingwise. > >But -again- that is not the purpose of ratings. > >Ratings are always relative, and thus are never inflated, they just tell >something about your strenght between your opponents playing the same game. > >J.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.