That's a difficult question, but, gun to my head, I'd say 50. You want to talk ratings? I can get on IRC.
-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
Smogon Forums
Antar
Aldaron
I'm at work ATM, just wasting time brainstorming cause bored
Antar
I really think you would have liked COIL given the opportunity to test it. Yes, it has issues when number of battles is low, but if what you're looking for is a clean cutoff for reqs, I think it's ideal.
Aldaron
We can probably fine tune some parameters with COIL to see what we want ya.
about calculating GXE: Given a player rating R and a rating deviation RD:
GLIXARE Rating = 0, if RD > 100
GLIXARE Rating = round(10000 / (1 + 10^(((1500 - R) * pi / sqrt(3 * ln(10)^2 * RD^2 + 2500 * (64 * pi^2 + 147 * ln(10)^2)))))) / 100, otherwise
that is in x-act's thread...do you know where he gets some of those numbers from?
about calculating GXE: Given a player rating R and a rating deviation RD:
GLIXARE Rating = 0, if RD > 100
GLIXARE Rating = round(10000 / (1 + 10^(((1500 - R) * pi / sqrt(3 * ln(10)^2 * RD^2 + 2500 * (64 * pi^2 + 147 * ln(10)^2)))))) / 100, otherwise
that is in x-act's thread...do you know where he gets some of those numbers from?
Antar
Yes. I know where he gets ALL of those numbers from.
GXE = prob. of winning against random person on the ladder = prob. of winning against an unknown player.
To get the values on that page, you have to plug in the rating of an unknown player as "R2" into the formula for E as found here: http://en.wikipedia.org/wiki/Glicko_rating_system
GXE = prob. of winning against random person on the ladder = prob. of winning against an unknown player.
To get the values on that page, you have to plug in the rating of an unknown player as "R2" into the formula for E as found here: http://en.wikipedia.org/wiki/Glicko_rating_system
Antar
But note that when X-Act wrote his formula he assumed that the distribution of player strengths had a std.dev of 350, so the rating of an unknown player was 1500±350. My analysis has shown that the correct unknown rating is actually 1500±130. So the formula for GXE now is different than it used to be.