Originally posted by ark13Yes, but while that player would decrease, all above that (new) player would increase.
Yes, it would. But for every player that joins that's better than that player, his rating would decrease slightly. So the rating average wouldn't change, the range would just expand. And it wouldn't change much.
If the average is of newcomers is slightly higher than 1200 like you said, I would expect that everyone above this number would, ceteris paribus, rise over time.
I'm sure it would be easy to prove (or disprove) this mathematically by looking at the ratings calculation formula, but I'm lazy today. Since I'm usually lazy for these type of things, I don't know if I'll try looking at it tomorrow or not...
Originally posted by PalynkaYes, you're correct that since the starting rating is 1200, and the average rating is 1300, there should be a slight increase in average rating over time. You just weren't making yourself very clear.
Yes, but while that player would decrease, all above that (new) player would increase.
If the average is of newcomers is slightly higher than 1200 like you said, I would expect that everyone above this number would, [i]ceteris paribus[/i ...[text shortened]... ngs, I don't know if I'll try looking at it tomorrow or not...
However, several people on this site have done studies showing that rating inflation isn't happening. I can't explain it.
Originally posted by ark13No, I mean the same player, without improving, if he's better than the average of newcomers, will see his rating improve over time.
Yes, you're correct that since the starting rating is 1200, and the average rating is 1300, there should be a slight increase in average rating over time. You just weren't making yourself very clear.
e.g. If my rating is now around 1500 and I keep playing exactly at the same level as before I would increase my average over time IF the number of participants increase.
If the number of entrants is equal to the number of people who drop-out, I see no reason for it to change, but if the number of entrants is bigger then I would expect a rise if I'm an above average player and a decrease if I'm a below average player.
Originally posted by ark13There was a post a while ago that mentioned that there were a lot more 2000+ players at the site, for example. There are many factors here, of course, including the number of good players that is attracted to the site.
However, several people on this site have done studies showing that rating inflation isn't happening. I can't explain it.
PS: I just realised that most players are still rising in rating (not reached their balance) or they are fairly stabilized (reaching their balance). If this is true, as the better players reach their balance, those stabilized will also grow with them as losses with better players will cost them less points.
I asked this question partly because my RHP rating appears terribly unstable at present. I've just won a string of games, with one loss and one draw in the mix. My rating has climbed 100 points in less than two weeks; 200 from my low point one month ago. I play on several sites, as well as several OTB tournaments per year, and my ratings vary substantially. Consequently, I'm interested in comparing my experience with that of others.
My current USCF OTB is 1568. I also have a USCF correspondence rating is 1805, but I haven't played a USCF postal game in about five years. My RHP rating 1665. Another site where I play online correspondence rates everyone new at 1500 through the first 20 games. At the end of my first 20, I was ~2200, but I then dropped to about 2150, and then rose above 2300. Currently I'm 2283 and have a rating floor of 2100. Rating floors may contribute to the rating inflation on that site.
I perceive that RHP has rating deflation, rather than inflation. But my current rating does not bear that out, unless I have recently taken a jump upwards in skill (a possibility due to some changes in my training regimen and opening repertoire).