| |
BGonline.org Forums
Stick vs. mfic final game cube action
Posted By: Chase In Response To: Stick vs. mfic final game cube action (Christian Munk-Christensen)
Date: Monday, 19 April 2010, at 12:55 p.m.
The only evidence I know of to support my hunch that humans win less than METs suggest when trailing is the Heinrich database, which, as Henrik mentioned, may be outdated due to style changes.
I think human tendencies swing in both directions. For example, humans tend to cube too late when leading at 2-away scores. On the other hand, humans also tend to cube too late when trailing at scores like 3a2a or 4a2a, and do not play as aggressively for the gammon as bots do. Back on the first hand, humans also tend to take too deeply at these scores. (Some of these tendencies were documented in the Big Brother project -- again, old data.) These effects, and others, will cancel each other out to some extent. My guess, and it's only a guess, is that they very slightly favor the leader. It would be great if someone could do as Henrik suggests and gather a large database of modern-day matches and analyze them. I'd like it to include more than just world class players, however.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.