I suspect that Paw is correct. I do not believe that this is the only reason but it makes sense that it would have an effect. 64, the fact that experienced players are below their peak suggests that his glicko theory is correct. They are probably at the same standard but the ratings have deflated.
Can anyone explain the discrepancies between ratings?
Sort:
Scottrf wrote:
I like the explanation, and it probably has some truth, but standard/bullet ratings are similar despite what I imagine are much more bullet games played.
It is definitely true, but there are lots of factors. For instance, the width of the pool of players has an effect, the number of times people play each other in a row has an effect (in bullet, many times players play each other a lot of times in a row, and each successive game will have a smaller effect since RDs of both players are decreasing each game--therefore, the number of games played total in that pool is less of a factor). etc etc
64whitemice wrote:
Not sure of that, pawpatrol - I've come across a few players who've played thousands of blitz games over several years, and most of them are well below their highest rating, ie they don't seem to have improved at all; on the contrary, they seem to have got worse. According to your theory they should be winning more often the more games they've played. And doesn't the RD settle down after around only 30-40 games? That's not enough for a measure of their general chess experience I think.
No, you are not understanding what I said. Whether a player improves or not is irrelevant. RD never settles, either. If you play 40 games in a week and then don't play for a year, your RD will be close to the same as it was before you played any games at all.
MCFan wrote:
I suspect that Paw is correct. I do not believe that this is the only reason but it makes sense that it would have an effect. 64, the fact that experienced players are below their peak suggests that his glicko theory is correct. They are probably at the same standard but the ratings have deflated.
This is a better interpretation, yes.
Not sure of that, pawpatrol - I've come across a few players who've played thousands of blitz games over several years, and most of them are well below their highest rating, ie they don't seem to have improved at all; on the contrary, they seem to have got worse. According to your theory they should be winning more often the more games they've played. And doesn't the RD settle down after around only 30-40 games? That's not enough for a measure of their general chess experience I think.