| |
BGonline.org Forums
Is a .09 cube error "huge"?-- *Spoilers*
Posted By: Maik Stiebler In Response To: Is a .09 cube error "huge"?-- *Spoilers* (Dmitriy Obukhov)
Date: Tuesday, 16 August 2016, at 5:35 p.m.
"Gain" is not an absolute concept. It requires comparing different situations or courses of action. You like to compare doubling vs. not doubling against a given opponent. I, or Albert, it seems, are interested, at least in this thread, in comparing not doubling against a perfect opponent vs. against a particular non-perfect opponent. Both are fruitful ways of thinking. Sometimes both can be used equivalently: Casper uses your thinking here to show that not doubling a 4 roll/4roll position against an opponent who always takes a 3roll/3roll is an 0.051 error. It is no coincidence that -0.051 = -0.156 + 25/36 * 0.151. Against a perfect opponent not doubling a 4/4 is a 0.156 error, while taking a 3/3 is a 0.151 error. By not doubling against the 3/3-taking player I sacrifice 0.156 now, but gain 0.151 later in 25 of 36 cases from his imperfection.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.