Check out our brand new Local Poker Communities! Get updates and interact with poker players in your area.
Visit the United States Poker Community | Visit the California Poker Community | Read more about the Launch of P5s Local

Standard deviation sngs and mtts

Reply to Thread
  1. timwatly timwatly is offline

    Phoenix, AZ
    United States

    Posts: 4
    Joined: Nov 07

    Nov 8th, 2007 (11/8/2007 1:23am)

    I used sharkscope to download huge excel spreadsheets of the top players and got out my stat book to figure out how many sngs it takes to have a an idea of what your roi is give or take 2 percent within a 95% confidence interval (and i am no statistician)

    I was getting numbers of like 15 thousand sit n goes that needed to be played. Is there anyway this is right or am i screwing up some math? And According to my stats if I am at lets say a 3 percent return on only 1400 games my roi is 3.5 give or take 7.5 percent (95% confidence interval)

    I always thought 1k games was close to the amount needed to be played. So what am i doing wrong, or is this perhaps correct?

    Also i was calculating variance on the 180 person tourneys and came up with like 30,000 tourneys needing to be played to be within plus or minus 10% roi. Can that be right.

    If you have only played 1k 180 person tourneys and average 100 roi (your true return is anywhere from 50 to 150 percent at a 95% confidence interval.

    I always here 1k or 500 is what you need to play.........is it really like 15k sit n goes?
  2. chrispy2005 chrispy2005 is offline

    IL
    United States

    Posts: 988
    Joined: Oct 06

    Nov 8th, 2007 (11/8/2007 1:41am) in reply to timwatly

    1k sngs is definitely not enough to determine true variance.

    30k seems a bit much but that is 5% significance level

    but for all practical purposes of keeping track of stats, you don't need to find a true variance, so a few thousand sngs is enough.

    example on ROI: over 1000 4.40$ 180 mans say you have 100% roi. so 4400 invested and a 4400 profit.

    say you win your 1001st sng for 216 (4.40 invested so profit = 211.6)
    new roi after 1001st sng : 4611.6/4404.4 = 104.704% roi.

    one sng can effect your roi here by 4.704 percent because the sample size is only 1k
     
Reply to Thread

Similar Threads