View Single Post
  #28  
Old 01-14-2007, 01:18 PM
cwar cwar is offline
Senior Member
 
Join Date: Dec 2005
Location: Cwar LLC
Posts: 2,491
Default Re: Buy in requirements

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
The variance of a binomial distribution is n*P*(1-p)
s.d = sqrt[n*P*(1-p)] but this assumes p is constant for each game .

[/ QUOTE ]

I didnt' assume that p was constant for each game. It varied between 50 and 70 percent. In my simulation, I first picked what the winrate would be, then, based on that, decided if the player won.

[/ QUOTE ]

but assuming u sit down first in a sng, u should calculate with p=0.6 as u dont know anything about your random opponent, dont u? sitting there with a good or bad opponent is part of luck, so your variance shouldnt be reduced if u do well vs certain opps and do bad vs others. pls correct me if i m wrong.

[/ QUOTE ]

Taking the average of your winrate does seems like it would reduce variance IMO but much much easier.
Reply With Quote