[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Probability to win a unlimited session of 10 games

Posted By: eXtreme Gammon
Date: Tuesday, 6 September 2011, at 5:31 p.m.

In Response To: Tournament Format and Prize Sharing By Dr. Garal (Daniel Murphy)

For short session it seems difficult to use normal distribution. they may be fancy way to determine that, but a good approach is to use a monte carlo simulation.

For instance, assuming the following (actual data from my own games, jacoby for the vast majority)

  • lose 4 points: 7.3%
  • lose 2 points: 20.3%
  • lose 1 point: 19.4%
  • win 1 point: 20.0%
  • win 2 points: 20.7%
  • win 4 points: 12.3%

    I ignored for simplicity any score>4 (they represent 2.7% of the games: 0.9% lost, 1.7% won)

    That's equivalent to 0.217 ppg (or 108 elo difference, 3.29 PR)

    on a 10 games session the outcome is, after 11,229 10-game sessions

  • 59% wins
  • 5% draw
  • 36% losses

    I am not sure it fits Dr. Garal proposal, i've been to lazy to try to understand it. if anyone is interested i can rerun the simulation with different parameters.

    • Messages In This Thread

     

    Post Response
    Your Name:
    Your E-Mail Address:
    Subject:
    Message:

    If necessary, enter your password below:

    Password:
     

     

    [ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

    BGonline.org Forums is maintained by Stick with WebBBS 5.12.