Timeline for How to calculate the FIDE Percentage Expectancy Table
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 31, 2018 at 7:53 | comment | added | Glorfindel♦ | Yeah, maybe they have a slightly different formula for the tails. It could be that they don't want people to gain rating from playing a lot of games against much weaker opponents, and therefore using a slightly higher expected score. | |
| Oct 31, 2018 at 7:49 | comment | added | user1205901 - Слава Україні | I checked some calculations in MATLAB. I get that if I enter for example 1/(1 + 10^((300-392)/400)) it equals 0.6294 just like you said. However, if I enter 1/(1 + 10^((300-838)/400)) I get 0.9568, when the table seems to indicate that a difference of 538 rating points should accord to .97. The table seems to be slightly 'out' in relation to your formula as soon as we get into high probabilities. | |
| Oct 31, 2018 at 7:32 | history | edited | Glorfindel♦ | CC BY-SA 4.0 |
added 185 characters in body
|
| Jan 18, 2018 at 15:53 | history | edited | Glorfindel♦ | CC BY-SA 3.0 |
added 171 characters in body
|
| Jan 18, 2018 at 15:46 | history | answered | Glorfindel♦ | CC BY-SA 3.0 |