Getting "Exact" Indeces trhough Simulation

From Stanford Wong's BJ21

Posted by MathProf on 20 Nov 1997, at 4:33 a.m., in response to Rounding, truncating, posted by Pete Moss on 17 Nov 1997, at 9:43 p.m.

Good Points about calculating Indeces. To summarize what I thought was one important poitn you made: If you are getting indeces by simulation, you should generate the indeces with the same rounding convention that you will use in playing.

Now have you or anyone given any thought to getting "esact" decimal critical values thorugh simulation. By this mean, an index of 1.38 or 2.56. (The term "exact" is not the best choice of words for what I am describing.) The way I see to do this is as follows:

Sample the various decks with TC from 1 to 2. For some, the play will be positive, for other the play will be negative. From each deck, we caluclate the Expected Gain (EG) from making th play. Again for some TCs of 1 this will be EG will be positive and for some it will be negative. Then form a regression line between EG adn TC and see where that line intersects the axis (EG=0).

I have thought about trying to do this, but have never actually done it. Any thoughts on this methodology...

Of course, the advantage to having and publising this index, is that then the user could make their own fully-informed rounding decisions.

Responses