| Angelo De Pa1ma 2005-04-11, 5:59 pm |
|
I actually think the system is way better than my fantasy method. Thank
goodness I have nothing to do with USCF ratings (or y'd all be calling me
Grossmeister).
Statistics be damned: Had I played five 2300 players my score most likely
would have been 0-5. Had I played ten of them my score would probably be
0-10.
Thanks again everyone for explaining this to me. It was most instructive.
"George John" <george@neosoft.com> wrote in message
news:1112640374.495627.311320@g14g2000cwa.googlegroups.com...
quote:
> Angelo De Pa1ma wrote:
>
> Angelo,
>
> [SNIP]
>
> times
> your
> weaker
>
> Just to be certain there is no misunderstanding, the formulas used to
> compute rating changes are the USCF's and not mine.
>
> The USCF doesn't publish 'performance ratings'. The calculation method
> using +400/0/-400 & opponents' ratings is a good hand calculation way
> to estimate performance ratings, but I think the calculator provides a
> better way to do this; although, it uses a method that would be VERY
> time consuming to do by hand or even using a non-programmable,
> hand-held scientific calculator.
>
> The online calculator performance rating calculation uses exactly the
> same formula the USCF uses to calculate 'expected scores' for
> non-provisional players. All I have done is write a 'search' algorithm
> that finds the rating which has an expected total score closest to the
> actual total score.
>
> So, WRT to your 2005 USATE performance, a 2300 rating was most likely
> to score 4.5 against opponents with the ratings you faced, (IOW, had
> the estimated score closest to actual).
>
> [SNIP]
>
|