I went through the trouble of deriving the formula for the standard deviation for a players hourly win rate and arrived at
This differs from the formula in the article by replacing 1/(total time) on the outside with 1/n. There is no way that I can see to rectify these results. I can only conclude that either a) the forumla was mis-typed or the source of the original formula was in error.Statistics: Posted by sfairgrieve — Mon Oct 02, 2006 9:55 pm
]]>