I wrote to dougz asking for clarification on his TS implementation, but I'll ask the same questions here in case any of you know the answers. The quotes are from
http://web.archive.org/web/20130116154350/http://dominion.isotropic.org/faq/In a nutshell: skill is measured on a scale that goes from roughly 0 to 50 points. (Actually skill can be any number, but 99.8% of players should fall in the 0–50 range.) The skill range column is a 99.8% confidence interval — the system is 99.8% sure your true skill lies somewhere in that range. New players are assigned a skill of "25 ± 25", which is to say, we don't really have any idea what that person's skill is.
Does 25 +/- 25 mean that you're starting with mu=25, σ=25? If so, then 0-50 would be more like a 68% CI, right? The default for the other implementations I've seen is 25 +/- 8.66. That is, σ=25/3.
The level column is the low end of your range, rounded down to an integer and clamped to the range [0, 50]. If we ignore the clamping, it is a conservative skill estimate in the sense that we are 99.9% confident that you are at least that skillful.
Did you eventually remove this on the upper end? I see a lot of people on your leaderboard who would be at large negative levels but are listed at 0. But I also see lespeutere and -Stef- over 50.
I've set β = 25, γ = σ0 / 100 (applied daily), and the draw probability at 5%.)
Empirically, I find more like a 1.75% draw rate. Was there any special reason for chooing 5%?
@Obi Wan Bonogi: You voted for isotropic's TS algorithm "without the decay," but dougz doesn't mention a decay in mu or a daily increase in σ. Can you clarify?