| |
BGonline.org Forums
XG's definition of "obvious"
Posted By: Timothy Chow
Date: Tuesday, 29 December 2009, at 7:16 p.m.
While trying to understand the details of XG's PR, I encountered a subtlety that others might have an opinion about.
XG's philosophy is to exclude trivially obvious decisions. In particular, if (according to the bot) the correct cube decision is "no double," and furthermore if doubling would be a 0.200 EMG error or worse, then the decision is considered obvious.
If you examine the fine print, the threshold for non-obviousness is even higher than that. Suppose the correct cube decision is "no double," and the equity you would lose by doubling is less than 0.200 EMG. Is such a decision non-obvious? Not necessarily! If your absolute EMG equity is less than -0.900 (assuming you don't double), then the "no double" decision is considered obvious, even if you would lose less than 0.200 EMG by doubling.
My question is, are there situations where one's equity is less than -0.900, and the correct decision is "no double," yet the "no double" decision is not an easy one in practice? I recently read an article by Steve Sax at gammonvillage.com about a match he played against Stick, where Stick correctly doubled in a situation where Stick's equity was something like -0.7 or -0.8. If there can be correct doubles when the equity is -0.8, then surely there exist incorrect doubles when the equity is -0.9, whose incorrectness is not obvious?
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.