| |
BGonline.org Forums
More data to support making the 2 pt with opening 64
Posted By: David Rockwell In Response To: More data to support making the 2 pt with opening 64 (Steve Mellen)
Date: Tuesday, 9 February 2010, at 3:06 a.m.
Steve...there is a substantial likelihood that a bot's play errors will be systemic.
That may be true in a backgame where the bot will face the same type of position 20 moves in a row. In the opening, the game is fluid and far more likely to generate positions of different type move to move which will not be susceptible to systemic errors. Furthermore, the type of position seen will change from player to player (first one on the bar, then the other). When this happens, errors will be somewhat offsetting. Systemic errors are not as likely to be found in the opening as they will later in the game.
Steve...There's just no scientific basis for attributing significance to a miniscule difference like that when there's so much white noise in the countless evaluations that take place along the way.
There is a scientific basis. It is called statistics. If one cared to do so, one could derive a correspondence which would show, for example, that if .004 is insignficant, then an error of X would be seen Y% of the time. If X seemed bigger than we could expect to ever see on a 100k rollout, then the premise that .004 was insignficant would be doubtful.
Steve...I don't have the sort of intuition that permits me to say ".004 means something, but .001 is just noise."
I do. As a life and health actuary working for an auditing firm, I get paid to do this every day. My expert testimony is that .004 means something even though it is far from certain. .001 is largely noise.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.