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

BGonline.org Forums

Bias in how most do XG analysis?

Posted By: Timothy Chow
Date: Thursday, 31 July 2014, at 9:21 p.m.

In Response To: Bias in how most do XG analysis? (Bob Koca)

This is correct. Didn't we discuss this all before (sophomore slump and so forth)? For example, when there was all the discussion about Stick's low PR and about rolling out selected decisions.

If you get multiple chances to get something "right" then you'll obviously do better than if you get only one chance.

The only thing I would quibble with is calling this a "bias." PR is a complicated thing and in my opinion, the definition of PR should include the entire algorithm, including what multiple passes you do under which conditions. Then it's not "biased"; it is what it is. One algorithm for computing PR might tend to produce higher or lower numbers than another slightly different algorithm, of course, so if you care about these fine distinctions then you need to clarify exactly what species of PR you mean. But to call it a "bias" implicitly requires you to say that one particular algorithm is privileged over the others, and I don't see the justification for that. There's no Platonic ideal PR. If you try to imagine a Platonic ideal PR then you immediately encounter philosophical knots like why XG's specific threshold for what constitutes a "trivial" decision belongs to the Platonic ideal but its sequence of rules for re-evaluations aren't.

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.