vpFREE2 Forums

Multi-Strike Poke: Computing the Return

#1

The "free ride" feature is separate from the deck being dealt. Where it
appears on the screen means nothing. If you go to the ledgaming.com site,
you can access the Bob Dancer article from Strictly Slots. That gives the
percentage for every game on the product of 98% or better. There is a way
to compute the return on the above products but it not worth the trouble.
See the wizardofodds.com site which does that for FPDW. (It agrees with the
Bob Dancer column.)

···

[Original Message]
From: brumar_lv <brumar_lv@yahoo.com>
To: <vpFREE@yahoogroups.com>
Date: 2/26/2004 1:28:36 PM
Subject: [vpFREE] Multi-Strike Poke: Computing the Return

There appears to be no way to compute the return of this game using
WinPoker or FrugalVP. First, there is the "Free Ride" card, with an
unknown probability. Second, the payoff multiples (2X, 4X, 8x)
are outside the scope of VP software. Does anyone know of a way to
compute the return?

It's interesting the "Help" instructions say "Each hand is dealt from
a separate 52 card deck", yet the Free Ride feature means there are
actually 53 different possible deal possibilities.

vpFREE Links:

http://www.west-point.org/users/usma1955/20228/VP/Links.htm

Yahoo! Groups Links

#2

Go to the archives of my Frugal Fridays column, beginning with June 20, 2003. I discuss this game at length in a 3-part series, including how to use FVP or other software to make strategy charts for the various lines for any game in that format.

···

________________________________________
Jean $¢ott - Go to http://www.FrugalGambler.biz
for VP software and strategy cards; "frugal" books;
and the NEW "Tax Help for the Frugal Gambler."

[Non-text portions of this message have been removed]

#3

--- In vpFREE@yahoogroups.com, "Howard Stern" <howard.w.stern@e...>
wrote:

The "free ride" feature is separate from the deck being dealt.

Where it

appears on the screen means nothing. If you go to the ledgaming.com

site,

you can access the Bob Dancer article from Strictly Slots. That

gives the

percentage for every game on the product of 98% or better. There is

a way

to compute the return on the above products but it not worth the

trouble.

See the wizardofodds.com site which does that for FPDW. (It agrees

with the

Bob Dancer column.)

Thanks for the feedback from so many of you regarding Multistrike. I
will check out the sources mentioned when I have time. My primary
interest is whether any version has a return equal to 9/6JorB.
Anything less than 99.5% has no chance of every favoring the player
at the SunCoast, even on double point days.

Still, I am surprised the "free ride" feature does not influence
(improve) the return. For example, if I were playing 9/6 JorB and
periodically got to play a game for nothing (a free ride), wouldn't
that improve the return by some unknown percentage? I shouldn't have
used my "53 card" analogy because that implies we know the odds of
the free ride feature (one chance in every 53 cards dealt). Maybe
I'm missing something here, because I've only watched others playing
so far.

#4

brumar_lv wrote:

Still, I am surprised the "free ride" feature does not influence
(improve) the return. For example, if I were playing 9/6 JorB and
periodically got to play a game for nothing (a free ride), wouldn't
that improve the return by some unknown percentage? I shouldn't have
used my "53 card" analogy because that implies we know the odds of
the free ride feature (one chance in every 53 cards dealt). Maybe
I'm missing something here, because I've only watched others playing
so far.

I don't think the intent was really to imply that the free ride
feature (and frequency) has no impact on game return.

In fact, if you removed the free rides from 9/6 Jacks MS, the expected
return (playing the "6/4/2" strategy) falls to 88.6%.

- Harry

#5

snip:

Still, I am surprised the "free ride" feature does not influence
(improve) the return. For example, if I were playing 9/6 JorB and
periodically got to play a game for nothing (a free ride),

wouldn't

that improve the return by some unknown percentage?

To understand why the "free ride" feature does not improve the game
significantly, Look at the following simplified "game": you are
dealt 1 card, if it is a spade you are paid 4 (3 to 1), if any other
suit you loose your bet. This would be a 100% game. However it would
be a bad bet to play on a second level for an additional bet unit
and be paid double, since your chance of making it to the second
level is only 1 in 4.

Relating this to real life JoB, according to standard strategy you
bust 54.54% of the time. Therefore w/o the "free ride" the game
would be too favorable to the house. Adjustments are also made in
standard strategy to lessen the bust percentage such as keeping 2
high cards over a low pair on the first level, etc. Only between
such modified strategy and the free rides does multi-strike become
playable.

David

···

--- In vpFREE@yahoogroups.com, "brumar_lv" <brumar_lv@y...> wrote:

#6

brumar_lv wrote:

--- In vpFREE@yahoogroups.com, "Howard Stern" <howard.w.stern@e...>
wrote:
> The "free ride" feature is separate from the deck being dealt.
Where it
> appears on the screen means nothing. If you go to the ledgaming.com
site,
> you can access the Bob Dancer article from Strictly Slots. That
gives the
> percentage for every game on the product of 98% or better. There is
a way
> to compute the return on the above products but it not worth the
trouble.
> See the wizardofodds.com site which does that for FPDW. (It agrees
with the
> Bob Dancer column.)
>
Thanks for the feedback from so many of you regarding Multistrike. I
will check out the sources mentioned when I have time.

With all due respect, and saying this as nicely as I can, you need to do that -- check on the sources provided to you -- before you post very many more questions or comments like this one: b

> Still, I am surprised the "free ride" feature does not influence
> (improve) the return. For example, if I were playing 9/6 JorB and
> periodically got to play a game for nothing (a free ride), wouldn't
> that improve the return by some unknown percentage?

I make that suggestion with all courtesy and kindness, as a friend, because that is precisely what _is_ discussed in probably every one of those "sources mentioned", including my own at http://velek.com/ms in which my calculations _PROVE_ that the ER for JoB-9/6 is _IMPROVED_ from the regular ER of 99.5439% to an ER of 99.7918% -- an _IMPROVEMENT_ of .25%, as Bob Dancer had initially reported.

  My primary
interest is whether any version has a return equal to 9/6JorB. Anything less than 99.5% has no chance of every favoring the player
at the SunCoast, even on double point days.

Your problem is probably going to be in _finding_ any MultiStrike games that actually have a good paytable. Everyone of them in Tunica, Mississippi, is absolutely horrible, with some of them being downright disgraceful (how does something like JoB-6/5 sound, IIRC)?

Still, I am surprised the "free ride" feature does not influence
(improve) the return. For example, if I were playing 9/6 JorB and
periodically got to play a game for nothing (a free ride), wouldn't
that improve the return by some unknown percentage?

See above.

  I shouldn't have
used my "53 card" analogy because that implies we know the odds of
the free ride feature (one chance in every 53 cards dealt). Maybe
I'm missing something here, because I've only watched others playing
so far.

When you play a non-Joker game, there are only 52 cards. The "Free Ride" feature is not represented by a separate card, but rather is sort of like a random 'back of the card' that is displayed until the card is turned over. Moreover, if you go to the paytable screen for any game, it will give you the specific percentages for the occurrences of Free Rides at each level. These vary with each game, and even with each paytable. Using that specific information, which is _supposed_ to be just as accurate and reliable as the paytable itself, we can make precise calculations of the ER and variance. I haven't gotten around to looking at the "other sources" mentioned besides Bob Dancer's article and the LED Gaming web-site, so I don't know, for example, what sort of info the "Wizard of Odds" provides, but I'm confident that it gives the same type of detailed info, and probably explains it more clearly than my page.

Now I've got to do a few other things, and eat, before I calculate the variance for the Double Pay Poker game you requested.

Cheers.

Bill Velek

#7

the variance for the Double Pay Poker game you requested.
Cheers.
Bill Velek>>

Whoops ... and now I realize I forgot to thank you for the
calculation of the the initial deal return. Very cleverly done, and
I thank you. I also should have known better than to even mention
the "free ride" issue until I checked out those sources. I just
wanted to thank those who responded to my inquiry before too much
more time passed.

As to the variance calculation, I have absolutely no clue how to do
it, so I await your result with great interest. Please explain how
you did it too.

···

--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:

Now I've got to do a few other things, and eat, before I calculate

#8

Bill Velek wrote:

Now I've got to do a few other things, and eat, before I calculate
the variance for the Double Pay Poker game you requested.

brumar_lv replied:

As to the variance calculation, I have absolutely no clue how to do
it, so I await your result with great interest. Please explain how
you did it too.

I'll be keeping an eye out for it too, Bill. Given the covariance
between pays on dealt hands and the subsequent drawn hands, not to
mention that standard covariance implicit in multiplay, the
calculation is a very challenging prospect.

- Harry

#9

Harry Porter wrote:

Bill Velek wrote:
> Now I've got to do a few other things, and eat, before I calculate
> the variance for the Double Pay Poker game you requested.

brumar_lv replied:
> As to the variance calculation, I have absolutely no clue how to do
> it, so I await your result with great interest. Please explain how
> you did it too.

I'll be keeping an eye out for it too, Bill. Given the covariance
between pays on dealt hands and the subsequent drawn hands, not to
mention that standard covariance implicit in multiplay, the
calculation is a very challenging prospect.

Well, Harry, I just posted it ... or I guess I should say _part_ of it ... as far as I am capable of going right now without taking a remedial course or two :frowning:

Actually, you will notice that I do mention in that post that I am unsure of how to continue once the variance is calculated for the first 'dealt' hand stage. Intuitively, I want to say that for _this_ type of a game, it seems appropriate to just average the first stage and the second stage, but I can't back that up with any math. And as you point out ... and as I also acknowledged in my other post ... things do get hairy when we get into multiplay, etc. That is why I specifically mentioned your name along with a few others as folks who might possibly know how to do this. I would very much like to see it done if it is other than averaging. I hope you or someone else will take it from here.

Thanks, and I look forward to seeing some comments in that thread. It is under the subject line of ...
"[vpFREE] How to calc variance: JoB-9/6 and Dbl Pay Poker examples"

Bill Velek