vpFREE2 Forums

Ultimate poker

#1

Saw 2 new games at Texas Station yesterday. MG/MD 3/5/10 play (near the North parking garage and in front of Quizno's) Bad pay tables such as 6/5 Bonus and Colorado Deuces, but an interesting quirk. A second pay table adds multipliers on the next hand after a hit, so a 4 of a kind pays 7 times and straight flush 13 times (on deuces). If, for example, you get a sf on the first line, the next hand will pay 13 times normal on that line. So far, so good, but the really bad part is that all bets cost 10 coins per line and pay as if you bet 5.

Anyone want to take a stab at evaluating this?

···

--
Love, IS!!!
(see IS - Spiritual Insights, a FREE book with NO ads at:
http://www.is1.org )

#2

Good luck. It's an incredibly difficult game to analyze. The problem is
there are just too many situations to even come close to doing it by brute
force, but the trick is absolutely ingenious (at least I thought so).

···

On Sat, Jan 10, 2009 at 9:29 AM, Dennis Krum <is@is1.org> wrote:

  Saw 2 new games at Texas Station yesterday. MG/MD 3/5/10 play (near the
North parking garage and in front of Quizno's) Bad pay tables such as
6/5 Bonus and Colorado Deuces, but an interesting quirk. A second pay
table adds multipliers on the next hand after a hit, so a 4 of a kind
pays 7 times and straight flush 13 times (on deuces). If, for example,
you get a sf on the first line, the next hand will pay 13 times normal
on that line. So far, so good, but the really bad part is that all bets
cost 10 coins per line and pay as if you bet 5.

Anyone want to take a stab at evaluating this?
--
Love, IS!!!
(see IS - Spiritual Insights, a FREE book with NO ads at:
http://www.is1.org )

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

#3

Just off the top of my head, I think you can arrive at a decent
approximation...

First evaluate the return of the pay table as is without considering
multipliers. Take that number and multiply it with the multiplier
for each payout so that you have a different product for each line in
the pay table. So just for the sake of an example let's say the base
return is .45 (I think I recall you saying that the paytable is akin
to a 5-coin paytable, but you are putting in 10 coins), so if the
multiplier for a full house is 10, 10 X .45 is the approximate added
bonus of hitting a full house. Add that number to the payoff for a
full house. Do this separately for every line in the paytable. Now
that you have a new adjusted paytable, evaluate the return of that
and repeat the process. I would guess that each iteration of the
process should probably get you closer to the true return.

Now the interesting problem with this method is that the numbers you
are using are working with the assumption that the current hand has
no multiplier. Obviously, using the example, adding 4.5 to the value
of the full house has a much greater impact when the full house pays
9 than when the full house pays 90. In generating a strategy it
would probably be worth generating a different strategy for every
multiplier value and seeing how close they are to one another. If
there is quite a difference, you might actually want to play a
different strategy for every value of multiplier for the current
hand. That is to say, if your current multiplier is 10X play one
strategy, if it is 5X play another. This is different than with STP,
where the value of the next hand is irrelevant. The strategies don't
change in STP because every pay gets the same multiplier so their
strengths stay the same in relation to each other. Here the value of
the next hand when added in will be a greater or lesser percentage of
the current pays depending on what your current multiplier is.
Basically the value of the NEXT hand has a different amount of impact
depending on what your current pays are. Whether this change in
impact is enough to warrant memorizing a different strategy for every
multiplier remains to be seen.

Like I said, this is just my initial guess as to how to come up with
an approximation. I would be interested to hear from any of you as
to whether you think this would work (and of course why you think so).

···

--- In vpFREE@yahoogroups.com, Dennis Krum <is@...> wrote:

Anyone want to take a stab at evaluating this?
--

#4

Just thought of another thing. Is it very noticable when you have a
multiplier on a current hand? I could easily imagine people waiting
around these machines for a person to leave when they just had a
winning hand. Then they would swoop in to use the multiplier that the
person left behind. I would guess that most people would not leave if
the just had a winning hand, but I'm sure someone will do it. Just as
I'm sure someone will be waiting for it to happen.

···

--- In vpFREE@yahoogroups.com, "pokegimp" <wincerwj@...> wrote:

--- In vpFREE@yahoogroups.com, Dennis Krum <is@> wrote:
>
>
> Anyone want to take a stab at evaluating this?
> --

#5

pokegimp wrote:

Just off the top of my head, I think you can arrive at a decent
approximation...

First evaluate the return of the pay table as is without
considering multipliers. Take that number and multiply it with the
multiplier for each payout so that you have a different product for
each line in the pay table. So just for the sake of an example
let's say the base return is .45 (I think I recall you saying that
the paytable is akin to a 5-coin paytable, but you are putting in
10 coins), so if the multiplier for a full house is 10, 10 X .45 is
the approximate added bonus of hitting a full house. Add that
number to the payoff for a full house. Do this separately for
every line in the paytable. Now that you have a new adjusted
paytable, evaluate the return of that and repeat the process. I
would guess that each iteration of the process should probably get
you closer to the true return.

Now the interesting problem with this method is that the numbers
you are using are working with the assumption that the current hand
has no multiplier. Obviously, using the example, adding 4.5 to the
value of the full house has a much greater impact when the full
house pays 9 than when the full house pays 90. In generating a
strategy it would probably be worth generating a different strategy
for every multiplier value and seeing how close they are to one
another. If there is quite a difference, you might actually want
to play a different strategy for every value of multiplier for the
current hand. That is to say, if your current multiplier is 10X
play one strategy, if it is 5X play another.

Very strong approximation, with two changes.

-- Treat the base return as 100% ... close enough for most games
you're likely to play (this is treatment akin to that in working with
MS).

-- Subtract "1" from the multiplier in adjusting the payouts. Even if
you fail to hit on this hand, you still get to play the next hand.
You only add the incremental bonus for hitting a hand.

You can calculate the adjustments on a "5-coin" basis. If the
multiplier produced by a given win is 3x, add 10 coins to that hand's
payout.

As you allude, this analysis yields a unique strategy for each
possible multiplier that applies to the current hand in play. This is
analogous to MultiStrike, where there's a unique strategy for each
Level (multiplier).

A likely larger number of potential multipliers makes this much more
difficult than MS to get one's hand around for optimum play. It's
possible that some decent simplifications can be made without
sacrificing too much ER to keep things workable.

- H.

#6

As an added note: Analyzing the return/variance of pokegimp's
suggested strategy is a relatively straightforward.

#7

Dennis Krum wrote:

Saw 2 new games at Texas Station yesterday. MG/MD 3/5/10 play (near

the

North parking garage and in front of Quizno's) Bad pay tables such as
6/5 Bonus and Colorado Deuces, but an interesting quirk. A second pay
table adds multipliers on the next hand after a hit, so a 4 of a kind
pays 7 times and straight flush 13 times (on deuces). If, for example,
you get a sf on the first line, the next hand will pay 13 times normal
on that line. So far, so good, but the really bad part is that all

bets

cost 10 coins per line and pay as if you bet 5.

Anyone want to take a stab at evaluating this?

Question: what happens to the pay table after a second consecutive
winning hand?

For example, you follow up a straight flush with a three of a kind;
what do the multipliers look like for the hand after the 3K? Does the
SF multiplier reset to 1?

-- Don

#8

You would get the multiplier associated with 3OAK. The cards and multiplier
has no memory beyond the immediately preceding hand.

···

On Sun, Jan 11, 2009 at 9:48 AM, Don Del Grande <del_grande@earthlink.net>wrote:

  Dennis Krum wrote:
>
> Saw 2 new games at Texas Station yesterday. MG/MD 3/5/10 play (near
the
> North parking garage and in front of Quizno's) Bad pay tables such as
> 6/5 Bonus and Colorado Deuces, but an interesting quirk. A second pay
> table adds multipliers on the next hand after a hit, so a 4 of a kind
> pays 7 times and straight flush 13 times (on deuces). If, for example,
> you get a sf on the first line, the next hand will pay 13 times normal
> on that line. So far, so good, but the really bad part is that all
bets
> cost 10 coins per line and pay as if you bet 5.
>
> Anyone want to take a stab at evaluating this?

Question: what happens to the pay table after a second consecutive
winning hand?

For example, you follow up a straight flush with a three of a kind;
what do the multipliers look like for the hand after the 3K? Does the
SF multiplier reset to 1?

-- Don

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

#9

I've been playing this game for fun on-line and free at the video
poker website, www.videopoker.com It sure is a lot of fun, but I'm
not sure if the games listed will follow the same paytables and find
their way into the real casinos.

The straight flush on deuces wild pays the highest multiplier at x 12,
and so does the full house on DBL, DBL, back at x 12.

I guess my question would be when playing deuces wild, would you
deviate from perfect strategy from the paytable and try to go for more
straight flushes, so on the NEXT hand, you have the highest multiplier
available then, the x 12 ??? It's awesome to see a x 12 multiplier
staring back at you on 10-hands though, when you are dealt a straight
flush on deuces wild. So I've been playing around and holding like,
2,7,10 or 2,7,jack just to see how many straight flushes I can capture
to get the x 12 to come out on the next play then.

I've been up/down and all around with this game, the variance has to
be quite high. If you get a lot of winning hands back to back, you
sure can win a lot of credits FAST. Or get dealt a string of pat
winning hands all in a row. But like anything else, the downside is
that doesn't happen too often.

But I have to say, it's a blast to play !

-Len

You would get the multiplier associated with 3OAK. The cards and

multiplier

has no memory beyond the immediately preceding hand.

> Dennis Krum wrote:
> >
> > Saw 2 new games at Texas Station yesterday. MG/MD 3/5/10 play (near
> the
> > North parking garage and in front of Quizno's) Bad pay tables

such as

> > 6/5 Bonus and Colorado Deuces, but an interesting quirk. A

second pay

> > table adds multipliers on the next hand after a hit, so a 4 of a

kind

> > pays 7 times and straight flush 13 times (on deuces). If, for

example,

> > you get a sf on the first line, the next hand will pay 13 times

normal

···

--- In vpFREE@yahoogroups.com, "Jason Pawloski" <jpawloski@...> wrote:

On Sun, Jan 11, 2009 at 9:48 AM, Don Del Grande <del_grande@...>wrote:
> > on that line. So far, so good, but the really bad part is that all
> bets
> > cost 10 coins per line and pay as if you bet 5.
> >
> > Anyone want to take a stab at evaluating this?
>
> Question: what happens to the pay table after a second consecutive
> winning hand?
>
> For example, you follow up a straight flush with a three of a kind;
> what do the multipliers look like for the hand after the 3K? Does the
> SF multiplier reset to 1?
>
> -- Don
>
>
>

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