Using "Many" plies early in the game gets you results that don't have much use at all.
If you think about it -- say you're choosing between opening with A or with B. And so you open with A, and then assign random racks to the end of the game to see who wins, and then you do the same with B... and repeat this. Well, there's so much randomness in the whole remainder of the game, that it will take millions and millions of iterations before you have a meaningful sample of the possible combinations between here and the end of the game. Whereas, if you go two or three moves out, there are many fewer combinations, and you can draw a meaningful conclusion about the state of the game in a couple moves' time with hundreds or thousands of iterations, not millions. Interesting side question -- has anyone calculated how many iterations you would need for a "Many" sim on the opening move to be statistically meaningful to whatever given degree? Has anyone run sims for that long? Chris Lipe On Mon, Nov 24, 2008 at 11:29 AM, Phil Robertshaw <[EMAIL PROTECTED]> wrote: > I don't sim games as often as I should. I've been experimenting with 4- > ply sims, between 500 and 1000 iterations. Wondering, though, if I > should use 'many' plies, which takes considerably longer. What is the > general combination of plies and iterations which most people use? I > find 4 plies generally gives a pretty good indication of the best > moves, but often disagree with the results. > > Phil R > >
