--- In vpFREE@yahoogroups.com, "5-card" <5-card@...> wrote:
>
> Frank,
>
> For my personal definition of long-term I like a post from Tomski.
>
> Years ago Tomski posted some possible long-term numbers on Skip's board. I copied a few of them. I didn't copy the pay scales, but 10 or more years ago FP DW was 100.76% and FP DDB was 98.98%.
>
> "For a 99% confidence of being +/- 1%
> DDB requires 2,783,536 hands.
> DW requires 1,713,354 hands
> JB requires 1,293,635 hands"
>
> Food for thought.
> With perfect strategy a $1 9/6 DDB player after 2,783,536 hands has a 99% confidence of having a loss between $2,783 and $281,137.
Long term and short term aren't strict mathematical terms. They mean different things to different people. I think the commnon misconception is that once I hit long term, I'll be at 100.76% playing deuces wild. That isn't true. I wrote a paper on long term a while ago and posted it on the files section but it has since disappeared. I will have to look it up.
For JOB, I have to play 1,293,635 hands to have a 99% chance of being +/- 1%. So, after 6,468,175 coin in, I have a 99% chance of within 64,682 coins of expected value.
In 1,293,635 hands I have will average a hair over 32 royals. Only 75% of the time will I be +/- 6 royals from that average value. That means 25% of the time, I'll be at least 28,000 coins further away from expected value ( +/- 7 roayls different). In JOB, the royal is about 85% of the variance. Think of video poker as 2 games going on at the same time. One is all the pay hands except for the royal and the other is the royal. The first one is pretty well behaved after 1.2 million hands. The number of royals, not so much. Because it is such a high pay and infrequent hand, it adds 85% of the variance but is only 2% of the over all payback.
