Here's how I look at the problem. Suppose I have $100 (or $1000) to gamble
with. I want to gamble, but I don't have a precise goal except that I don't
really want to lose it all quickly. I have two games to choose from. The
first is a lottory that has a very high payout and has an ER or 101%. The
second is a VP game that has an ER of 99.9%. So do I sit down at the
"losing" VP game, or do I buy 100 lottery tickets and play the "positive"
game?
Dick Kalagher
Skip,
It seems then that what is commonly called a max-EV strategy in some
posts here (which would not be your vocabulary) includes playing a
massive number of hands. I was confused about this. I thought for a
while a max-EV strategy just meant to play well and not hold cards
for which the EV was not maximum (as compared to other possible holds
in the original five cards). I think the name is not happily chosen.
The max-EV strategy can never be maximum because it increases
indefinitely and only ends when one quits VP for good, or dies. A
better term would be limitless-EV strategy or indefinitely increasing-
EV strategy. But that is being too punctilious maybe. As long as
people understand, it is all the same. I believe I understand now,
knock on wood.
Thanks,
E.
E
That confusion is the reason I defined my terms as I went along.
Skipmubowor wrote:
>
> Hello Skip Hughes,
>
> Just a comment to say it is hard to keep these notions of EV and
ER
> straight since people are not consistent. What you below call EV
> seems precisely what Frugal and the Dogma here call ER. Further
you
> do not give a name to what the Dogma calls EV, which is certainly
not
> what you call ER, since you say EV, ER do not depend on the
number of
> hands played. Apparently you call ER (expected return) the return
of
> a machine, what WinPoker calls the total return, and Leny Frome
the
> payback.
>
> Once one realizes what whoever is writing means by whatever term
is
> being used, then one can understand what the person is talking
about.
>
> Lately a lot of talk has been dedicated to max-EV strategy. By
that I
> understand several things. First, in the casino you are playing,
play
> at the machine with the highest return (using your terminology)
and
> when playing, use the strategy that chooses to hold the cards that
> have the highest EV (again, according to your terminology).
Further,
> according to the Dogma here, I would assume that a max-EV strategy
> also involves playing as many hands as possible in the shortest
> amount of time possible (speed, multi play).
>
> A strategy that is not a max-EV strategy, then, would mean either
> choosing a machine with a return less than the best in the given
> casino one has chosen to play; and/or playing a machine in a way
> that, for one reason or another, one chooses out of a first deal
of
> five cards, to hold a number of cards that is not the number of
cards
> that has the highest EV. Another way not to play a max-EV strategy
> would be to just linger on a machine playing very slowly. A person
> might want to do this, for example, to get a free drink.
>
> If what I have said above is right, then maybe I'm beginning to
see
> the light. If not, since you are a patient person, I would beg
you to
> set me straight. Thanks,
>
> E
>
> --- In vpFREE@yahoogroups.com, Skip Hughes <skip-hughes@e...>
wrote:
> > I'm going to get away from Steve's point a little here. Those
who
> are
> > still pondering the meaning of video poker life should read the
> above
> > carefully. The EV of a particular held hand is simply the
average
> result
> > from holding that hand. With a 52-card deck, if one holds four
> cards
> > there will be 47 results to average, with three cards there
will be
> 1081
> > and so on.
> > The return of a game is simply the average result of each of
the
> > 2,598,960 different hands when selecting the hold that has the
> highest
> > EV for that hand. While it's obviously true that a large sample
of
> hands
> > is more likely produce a result that is closer to the Expected
> Return of
> > the game, those numbers (EV/ER) have nothing to do with the
number
> of
> > hands played. They are simply averages that apply to each hand.
> >
>
> vpFREE Links: http://www.west-
point.org/users/usma1955/20228/VP/Links.htm
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
--- In vpFREE@yahoogroups.com, Skip Hughes <skip-hughes@e...> wrote:
> I'm not alone. Tom Robertson seemed to have read the phrase
> "optimal EV" as another way of saying "max-EV," just as I did. If
> you read your Dogma by substituting the phrase "max-EV" at
> each point where you had "optimal EV" then I think you'll
> understand my confusion.As I said, I used "optimal EV" trying to cover all possibilities, although
I believe that " max-EV" is accurate and correct (and less confusing) when
accompanied by the "different goals may choose other strategies"
disclaimer.Accordingly I've just reworded the Dogma as follows:
"In routine, everyday play anyone trying to maximize their profits will
always choose the highest EV option when playing a hand.
Tom, this is exactly what I _thought_ you were saying, and it is also
what I'm claiming to be misleading. Surely "highest EV" and "max-EV"
are the same concept. The craps/roulette example I gave was intended
to illustrate how "choose the highest EV option" can (strange as it may
seem) FAIL to "maximize profits."
This is a counterintuitive result, and that is why I went to great length
to explain it. Your change in wording moves closer to what I've
called "the party line" rather than toward "strange but true." I'm
thinking that you have not completely understood my post. You say
that I'm preaching to the choir, but when you echo the message
of the sermon back to me it comes back different. This tells me that
I've failed to accurately convey the intended message. Like I said
to someone else, if there hasn't been some point where you've said
to yourself "that can't be right" then you probably haven't understood
what I'm saying.
You said in a recent post that I'm anti max-EV, but that isn't the case.
My point isn't that max-EV is "bad" because it isn't. My point is that
max-EV is just one alternative among many possible objectives, and
that none are inherently superior to the others.
Players with
different goals may choose other strategies, but in doing so they lower
their profit expectations."
But "profit expectation" is not absolute, it too is relative to the player's
objective. Different objectives require different optimal strategies in order
to maximize profit expectations relative to the chosen objective.
> I'll write another review with suggestions for your consideration.
Thanks. I believe that the Dogma's statements are appropriate and correct
for VP players as written, but I would like them to be accurate from all
angles, and include all necessary disclaimers.
I'll write up my suggestions, but I'm guessing that you will probably not
like them.
On Sunday 12 October 2003 11:57 am, vp_FREE wrote:
On 12 Oct 2003 at 11:20, Steve Jacobs wrote:
This is starting to get a little silly, nit-picky, etc. etc.
Here's an email I received from Dean Zamzow in April in response to a
question I had about methodology. If you read it, you will see that
in the case of max-EV, the term "highest ER" was used by him
interchangeably. That's not the main point, though, and he's not in
the business of realizing every tiny nuance of politically corrrect
vp parlance. The main point is that he describes his methodology for
determining best holds.
"Out of a 52 card deck there are 2,598,960 possible dealt 5 card
hands where
order doesn't matter. WinPoker looks at every one of these dealt
hands. Then with each hand, every possible hold combination is
evaluated
(32 of them). With each possible hold there are 1 possible hands
when
holding all 5, 47 possible hands when holding 4, 1081 possible hand
outcomes when holding 3 and so forth. Each one of the possible hand
outcomes is added together then divided by the total number of
outcomes on
that hold to determine that holds ER (this is what you see in the
details
screen in WinPoker). The hold with the highest ER is the best play.
Each
best play ER is added together for each possible deal (2,598,960) and
then
divided by the coin in to arrive at the machines return. So there
you have
it ... every possible hand is looked at to give the theoretical
return with
perfect play.
Thanks,
Dean"
Now remember, ER is a long term estimate. Playing each individual
hand optimally is not long term, but the results of playing each hand
optimally are. I hope this helps to clear up confusion among some who
aren't sure what's long term, what's short term, and what's somewhere
in between.
lb
I'll write up my suggestions,
Sounds good.
but I'm guessing that you will probably not like them.
I'll be the judge of that <g>.
Thanks,
vpFREE Administrator
On 12 Oct 2003 at 18:53, Steve Jacobs wrote: