--- In vpFREE@yahoogroups.com, "dunbar_dra" <h_dunbar@...> wrote:
>
> Perhaps you're remembering a discussion about longterm at this link:
>
> http://members.cox.net/vpfree/FAQ_LT.htm
>
> ...especially the part posted by John G. Zaroff part way down the page. Zaroff credits TomSki with calculating how many hands it would take to have a 95% chance of being within 1% of EV.
>
> That's an interesting measure, but I don't think either Zaroff or TomSki meant for that to be an official technical definition of long term.
>
> --Dunbar
>
This is John Zaroff. Long Term doesn't really have a fixed definition. The Tomski definition gives a way of calculating one definition of long term. It really has to be defined in terms of % away from expected and a confidence interval.
If you want to know how long you have to play before you exactly achieve the actual expected value, it is really infinite. This is similar to the limits definition from calculus. By choosing an appropriate value, you can make the result as close to the limit as you want but you never hit it.
If you want 1% deviation, that is one value. If you want 0.1% you need 100 times as many hands.
You can calculate how many hands to get how close, but which one of these calculations is 'long term' is up for discussion.
