vpFREE2 Forums

A silly, simple question that shows my ignorance

I think that would result in a huge variance value that would be
totally unrealistic since no player tries all 32 ways of playing
every hand.

Sorry, I should have made it clear that I only considered all
the possible draws of the optimum play for each dealt hand. It
was obvious for me (no need to considers cases that don't
happen), but I guess that it wasn't clear in what I wrote.

Actually, for the case of 1000 players each playing one hand, I can
say that exactly 0% (none) of those players can be more than one SD
below expectation.

That was exactly my point.

How many hands is "enough" is subjective, and I don't think it is
important. The Risk of Ruin for a given game and bankroll is much
more significant [...]

Agreed - although I'll argue that "enough" can be defined quantitatively,
but indeed it's not what matters for VP play.

JBQ

···

On 8/22/05, Dan Paymar <Dan@optimumplay.com> wrote:

Agreed - although I'll argue that "enough" can be defined

quantitatively,

but indeed it's not what matters for VP play.

"Enough" can be approximately defined:

N0=variance/(er-1)^2 hands

for example for fpdw, N0=26/(1.0075-1)^2 = 462,222 hands

N0=84% chance of net winning, 16% of net losing (1sd)
4xN0=98% (2sd)

assumes normal distribution so only valid if N0 > royal cycle

···

--- In vpFREE@yahoogroups.com, Jean-Baptiste Queru <jbqueru@g...> wrote: