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

BGonline.org Forums

THEN see this RO

Posted By: Daniel Murphy
Date: Sunday, 9 January 2011, at 4:12 p.m.

In Response To: THEN see this RO (Chuck Bower)

Thanks, Chuck. I did look first at the Simborg position, and didn't believe the rollout that said Double/Pass. I would not have doubled, although I would not have been able to say how close to the D/T line it was. Stick's comment that people tend to double backgames too because people take them too late is spot on -- so, when I look at the Simborg position, a big part of my practical cube thinking is "I and nearly everyone I know is going to be taking this for a long time."

Then looking at the Bower position, I didn't believe that having the 4 point instead of the 5 point could make such a huge difference. And then I saw that Petter Bengtsson has apparently found the problem with the Simborg rollout. So there's some XG numbers to quantify "believe" -- 0.120 difference? Maybe. 0.360 difference No way. I did note that the 95% confidence intervals in both Chuck and Petter's rollouts are about 0.040

The Position of the Day rollouts ought to show their parameters.

At the POTD page, I thought this comment by "admin" was interesting, although I don't agree with it: "Whenever you disagree with or don’t believe the rollout from XG, there are basically two things you can do: 1) more rollouts and 2) play the game out many, many times yourself until you have some statistical proof that the bot is wrong."

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.