| |
BGonline.org Forums
Let's Be Consistent -- An Error is an Error
Posted By: Rich Munitz In Response To: Let's Be Consistent -- An Error is an Error (Bill Riles)
Date: Monday, 20 July 2009, at 9:51 p.m.
I agree with you Bill. Nobody should be second guessing, and errors are errors.
But perhaps there is some value for keeping a separate statistic for "induced error rate". Basically, how much equity did your opponent give up by making wrong T/P decisions in response to your own "wrong" D/ND decisions.
By keeping a separate stat, the pure error rate would remain a well defined, universally understood standard, but for those who are interested in the induced error effect, they would have a way of seeing that too.
I honestly believe that a bot would achieve greater success against humans (perhaps even against other bots), if it would not assume perfect cube decisions on behalf of its opponent, but rather if it would assume that say 5% of the time the opponent would make an error on their side of the decision. Basically, program the concept of "practical" cube decisions.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.