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

BGonline.org Forums

PR difference affecting the gameplay

Posted By: Phil Simborg
Date: Wednesday, 11 March 2015, at 2:46 a.m.

In Response To: PR difference affecting the gameplay (Karol Szczerek)

I doubt that we'll ever really be able to determine this except for bots. How to apply this to human beings would be nearly impossible.

For one thing you can take two players with the same PR average and there could be a big difference in which one wins more often when playing a bot vs playing other human beings or even playing each other.

A two PR difference of 3 PR to 5 PR is probably very different from a two PR difference from 8 PR to 10 PR.

Even if it was theoretically possible to come up with some kind of formula, it would take far too many games between two humans to get a statistically relevant sample, and it would be nearly impossible to definitively say what the players' true PR was to start with.

Again I guess you could have eXtremeGammon play itself at a lower level a million times and see what the two PR's ended up and the results, but that still wouldn't tell you much about predicting human results.

I am not criticizing asking the question--I'd love to know the answer myself. I'd love to tell my students that if they improve by X PR they will win Y more. I just don't think the answer can be found with any degree of reliability.

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.