| |
BGonline.org Forums
unusual match
Posted By: Henrik Bukkjaer In Response To: unusual match (Timothy Chow)
Date: Thursday, 14 July 2011, at 10:07 p.m.
Actually I don't care if bot beats CMC or the other way around consistently.
It's the estimate (from analysis) vs. actual result.
Bot plays CMC 10000 games, says (based on the luck-adjusted analysis) that it should have won 5300 games. It won actually 5050 games. Now was it just unlucky? Or was it's estimate wrong (thus room for improvement).
In what sort of games and vs. which opponents do the estimate/analysis get close to actual result and for which ones are the gap big? There you can see which area to improve on!
Also - it could be fun to then analyze CMC play week players, and see if the gap opens up? Maybe CMC performs relatively better vs. weakplayers, compared to the bot?
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.