| |
BGonline.org Forums
unusual match
Posted By: Timothy Chow In Response To: unusual match (christian munk-christensen)
Date: Wednesday, 13 July 2011, at 1:57 p.m.
CMS wrote:
LAR is not perfect.
Nothing is.
I guess that the LAR is also just based on the bot's evaluation of the equity associated with the best play of the possible rolls.
Certainly the bot's evaluations come into play. The advantage of the LAR is that in the long run, it's unbiased, and therefore any mistakes in the bot's evaluation will average out.
In the short run, though, LAR has too much noise. My suggestion is not to use the LAR, but to use the difference in total MWC errors. Like PR, total MWC error presumes that the bot is perfect, but it has several advantages that PR doesn't, as I explained here.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.