Thanks friends for your responses.
Allow me to return to the original question, which really had nothing to do with RA indices, but was my attempt to find a simple explanation for how the Index for each alternative was determined.
In reviewing the output for the EV maximizing indices in the 2/3/11 post by Southpaw above, it appeared that the Index for the vast majority of choices fell at the TC where the "Diff" between the EVs for the 2 alternatives crossed over from negative to positive. This made sense to me.
However, there were some exceptions, which triggered my question, dealing with 15 DD situations & 1 Surrender (8,8 v X).
It turned out that the Surrender situation was a bug in the program that Norm is fixing. All the others involve the choice of putting more money on the table & Norm had decided that where the increase in EV was insignificant, then the crossover point should not be the Index, and I was curious as to what is the quantification in the program of "insignificant".
It appears that for the 15 DD situations, the "Diff" is .002162 or less, & the Index is kicked up to the next higher TC; for all the other DDs, it appears that the Diff is higher than .002162 and thus the Index is at the crossover point.
So if someone asked me for a simple explanation of how each Index was determined, I would be inclined to say the following.
Simulations were run dealing billions of hands to determine the EV of each
of 2 alternatives for each possible decision point at every possible TC. In general, the Index falls at the TC where the difference in EVs for the 2 alternatives crosses over from negative to positive.
However, if the choice involves putting more money on the table, then to avoid taking unnecessary risk, the Diff in EV must be greater than .002162
(or however this is quantified in the program?); if not, the Index kicks up to the next higher TC.
Norm,
Please let me know if this explanation is accurate,
or if not, please correct it.
Also, how would you specify the quantification point?
Thanks.
MikeyD
