|
|
|
Topic: RATINGS CALCULATIONS
| |
|
|
Author
| Message |
|---|
|
|
| Rating changes calculated from the rating at the end of the game are prone to manipulation by the players |
I don't see that as 'manipulation'. If you delay your game to minimize your loss or maximize your gain you are actually allowing the system to calculate your new rating more accurately, so you are not causing any harm in doing it. And besides, you are still bound by the game clock, so you won't be able to stall a game that much anyway.
And as I also mentioned, drastic rating fluctuations only happen to provisional players during their first few games, so when you are past that you won't be making any significant wins/loses in points by stalling a game.
Miguel
|
|
|
This is form the USCF so I can only say that I was wrong 
Game are rated based on the rating of each player at the end of the game, not the beginning. Many players are in more than one tournament at the same time, their rating may be constantly changing. Does that answer the question?
From
Stephanie Colley
Assistant Correspondence Chess Director |
|
|
|
As I see it, there are only three workable possibilities to calculating the rating change:
1. Use the ratings at the beginning of the game.
2. Use the ratings at the end of the game.
3. Use an average of the two.
Statistically, after playing enough games, I believe that these will all approach the same overall rating. For the short term though, I agree with Miguel and the USCF, the ratings at the end of the game should be the latest, most current, and therefore the best ratings to use.
--p
|
|
|
The calculations for your side are correct, you won 30 points after this match.
This is calculated as follows:
max(2287-1848,350) = 350
new_rating = old_rating + 0.04 * 350 + 16 = 1848 + 14 + 16 = 1878
Now dnptrs losing only one point looks wrong and I will investigate the implementation of the algorithm, even though at first glance I couldn't spot any obvious problems in the computation.
Miguel |
Miguel, after all changes, how is the formula? Is the about page updated?
This 350 maximum, it is used in the stablished ratings, it is not mentioned in the about page (although it is mentioned in the uschess fed.).
What happens when the value used is:
max(1848-2287,350) = 350
The correct should be -350 ?
Thanks for helping...
Beco.
|
| Previous 1 2 3 |
|
|
|
