I don't live near an Indian Casino, so I'm not particularly interested one way or the other, but in view of the lengthy and numerous discussions in vp/gambling circles about Indian Casinos, I thought those of you who are interested ought to take a look at this article by John Robison in the Casino City Times, dated today -- http://casinocitytimes.com/
Wanted to post the link here in case there are some of you who do not subscribe.
Cheers.
Bill Velek
all indian tribal gaming is regulated by the NIGC:
http://www.nigc.gov/nigc/laws/igra/igra_index.jsp
but regulations aren't very specific:
"Class III Gaming
(d) (1) Class III gaming activities shall be lawful on Indian lands
only if such activities are--
(A) authorized by an ordinance or resolution that--
(i) is adopted by the governing body of the Indian tribe having
jurisdiction over such lands,
(ii) meets the requirements of subsection (b) of this section, and
(iii) is approved by the Chairman,
(B) located in a State that permits such gaming for any purpose by
any person, organization, or entity, and
(C) conducted in conformance with a Tribal-State compact entered into
by the Indian tribe and the State under paragraph (3) that is in
effect. "
and for Thunder Valley:
http://www.auburnrancheria.com/html/facts/01.html
Law Enforcement Jurisdiction
The Sheriff will have authority to enforce all state criminal laws,
except state gambling laws, on all trust lands. However, prior to
entering any facility on trust lands, the Sheriff's office will
notify the Tribe's public safety or security director and cooperate
with Tribal officers.
In my mind, minimum video poker regulation would require random
independent inspection of roms to ensure they have not been altered
(this assumes the original manufacturer's roms have been verified as
random and do not contain switches that would allow someone to alter
random behaviour) and there must be an independent board that will
hear player-casino disputes (such as casino refusal to pay a jackpot)
and will investigate allegations of players or casino workers.
Nevada, New Jersey and Mississippi have these regulations in place.
In my mind, these jurisdictions offer fair gambling. Note, it is
still possible to get cheated, even in Nevada, but at least the odds
are significantly reduced. Regulation (i.e Law) doesn't eliminate
crime, but it does reduce it.
I don't live near an Indian Casino, so I'm not particularly
interested
one way or the other, but in view of the lengthy and numerous
discussions in vp/gambling circles about Indian Casinos, I thought
those
of you who are interested ought to take a look at this article by
John
Robison in the Casino City Times, dated today --
Wanted to post the link here in case there are some of you who do
not
I don't live near an Indian Casino, so I'm not particularly interested
one way or the other, but in view of the lengthy and numerous
discussions in vp/gambling circles about Indian Casinos, I thought
those
of you who are interested ought to take a look at this article by John
Robison in the Casino City Times, dated today --
Wanted to post the link here in case there are some of you who do not
subscribe.Cheers.
Bill Velek
Thanks for the post, Bill. I can give some perspective on the
situation in NM and how federal law influenced the outcome here.
_Any_ law that legalizes _any_ type of gambling for _any_ length of
time will open the door for "compacts" as they call them in NM.
Casinos were outlawed also in NM, but "Casino Nights" are legal to use
once a year for fund-raising events. The political parties chose to
keep this priveledge rather than close the loophole, thinking they
could use the old patronage system to muscle a big piece of pie from
the compacts. The Pueblos merely fell back on the federal Indian
Gaming Law, opened their LV style (house banked) casinos anyway, and
negotiated on their terms. This worked, as they cut the original pie
they would "pay" in half. It could be safely assumed that Oregon also
had a similar loophole, they refused to close it and lose a
fund-raising, VIP elbow-rubbing opportunity. What's left? If states
would open up the possibility of Racinos or "non profit" casinos (one
or two per) metro area, they will be able to leverage a Nevada-style
gaming commision, spot inspection control over all casinos within
state boundaries. A charity casino operating games on the same
margins as the Palms in LV would scare the daylights out of the 4
major tribal casinos here in Albuquerque, I believe they would agree
to immediate revision of the reporting and control requirements.
--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:
* Bill Velek wrote: I thought those
of you who are interested ought to take a look at this article by John
Robison in the Casino City Times, dated today --
http://casinocitytimes.com/
In other words no one. The B.I.A.s sole power is check which types of
gambling are permitted and operated.
[Non-text portions of this message have been removed]
(snip) take a look at this article by John
Robison in the Casino City Times, dated today --
Wanted to post the link here in case there are some of you who do
not
--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:
subscribe.
Cheers.
Bill Velek
*********************************************************************
Following your link, I came across another article by John Robison
which contained a common error, a misconception that is not immnuned
in 'gambling gurus', not even one with a master's in computer
science (see also Peter Griffin's _Extra Stuff_):
re: your 15 January 2004 Casino City Times Article
Mr. Robison:
On the answer to Steve's question, you wrote:
"Twenty-five thousand spins isn't really that many. If you were
playing video poker, you'd have only just passed the 50/50 point of
having hit a royal flush."
You were wrong. The 50/50 point of having hit a royal flush occurs
at (natural log 2)*Royal Flush Cycle.
For Jacks or Better, this point occurs at 27,997 hands;
for full pay Deuces Wild, this point occurs at 31,387 hands;
for 10/7 Double Bonus, this point occurs at 33,304 hands.
You failed to distinguish between the MEAN and the MEDIAN, and made
the flawed observation that the 50/50 point occurs at half of the
value of the MEAN.
Please be kind enough to publish a retraction as a service to your
readers.
Sincerely,
L. Wluiki
(snip) take a look at this article by John
Robison in the Casino City Times, dated today --
Wanted to post the link here in case there are some of you who do
not
subscribe.
Cheers.
Bill Velek
*********************************************************************
Following your link, I came across another article by John Robison
which contained a common error, a misconception that is not immnuned
in 'gambling gurus', not even one with a master's in computer
science (see also Peter Griffin's _Extra Stuff_):re: your 15 January 2004 Casino City Times Article
Mr. Robison:
"Twenty-five thousand spins isn't really that many. If you were
playing video poker, you'd have only just passed the 50/50 point of
having hit a royal flush."You were wrong. The 50/50 point of having hit a royal flush occurs
at (natural log 2)*Royal Flush Cycle.
I may be misunderstanding something. Isn't the natural log of 2 the
power to which 10 must be raised to get 2? That's .30103. But the
point in a cycle at which there is a 50% chance of hitting it occurs
about 69.3% of the way into it.
For Jacks or Better, this point occurs at 27,997 hands;
for full pay Deuces Wild, this point occurs at 31,387 hands;
for 10/7 Double Bonus, this point occurs at 33,304 hands.
Assuming these figures are right, which all seem to agree with my
69.3% figure, how does his conclusion that 25,000 spins can't be just
past the half-way point follow? The only criterion mentioned is
"video poker." There are plenty of video poker games in which the
royal cycle is less than 25,000/(69.3%) (approximately 36,000) hands.
You failed to distinguish between the MEAN and the MEDIAN, and made
the flawed observation that the 50/50 point occurs at half of the
value of the MEAN.
Is he saying that the mean is 1 cycle? If so, he's still assuming
that the cycle in the original situation is just under 50,000 hands
and not just under 36,000 hands. On what basis is he assuming that?
--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:
On the answer to Steve's question, you wrote:
Please be kind enough to publish a retraction as a service to your
readers.Sincerely,
L. Wluiki
Your 69.3% figure is correct. Natural log of 2 is the power _e_ must
be raised to get 2, where e is the natural number 2.71828.....
From windows' scientific calculator, ln 2 = 0.693147.......
I stand by my post.
L.Wluiki
*********************************************************************
--- In vpFREE@yahoogroups.com, Tom Robertson <thomasrrobertson@e...>
wrote:
>--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:
>
>(snip) take a look at this article by John
>> Robison in the Casino City Times, dated today --
>http://casinocitytimes.com/
>>
>> Wanted to post the link here in case there are some of you who
do
>not
>> subscribe.
>>
>> Cheers.
>>
>> Bill Velek********************************************************************
*
>
>Following your link, I came across another article by John
Robison
>which contained a common error, a misconception that is not
immnuned
>in 'gambling gurus', not even one with a master's in computer
>science (see also Peter Griffin's _Extra Stuff_):
>
>re: your 15 January 2004 Casino City Times Article
>
>
>Mr. Robison:
>
>On the answer to Steve's question, you wrote:
>
>"Twenty-five thousand spins isn't really that many. If you were
>playing video poker, you'd have only just passed the 50/50 point
of
>having hit a royal flush."
>
>You were wrong. The 50/50 point of having hit a royal flush
occurs
>at (natural log 2)*Royal Flush Cycle.
I may be misunderstanding something. Isn't the natural log of 2
the
power to which 10 must be raised to get 2? That's .30103. But the
point in a cycle at which there is a 50% chance of hitting it
occurs
about 69.3% of the way into it.
>For Jacks or Better, this point occurs at 27,997 hands;
>for full pay Deuces Wild, this point occurs at 31,387 hands;
>for 10/7 Double Bonus, this point occurs at 33,304 hands.Assuming these figures are right, which all seem to agree with my
69.3% figure, how does his conclusion that 25,000 spins can't be
just
past the half-way point follow? The only criterion mentioned is
"video poker." There are plenty of video poker games in which the
royal cycle is less than 25,000/(69.3%) (approximately 36,000)
hands.
>You failed to distinguish between the MEAN and the MEDIAN, and
made
>the flawed observation that the 50/50 point occurs at half of the
>value of the MEAN.Is he saying that the mean is 1 cycle? If so, he's still assuming
that the cycle in the original situation is just under 50,000 hands
and not just under 36,000 hands. On what basis is he assuming
that?
>Please be kind enough to publish a retraction as a service to
your
>readers.
>
>Sincerely,
>
>L. Wluiki
--- In vpFREE@yahoogroups.com, Tom Robertson <thomasrrobertson@e...>
wrote:
(snip)
The only criterion mentioned is
"video poker." There are plenty of video poker games in which the
royal cycle is less than 25,000/(69.3%) (approximately 36,000)
hands.
(snip)
*********************************************************************
I just checked every game in WinPoker and failed to find one of your
"plenty of video poker games in which the royal cycle is less than
25,000/(69.3%) (approximately 36,000) hands".
Robison was addressing a slots player, not some one who would adjust
strategy on a sufficiently high progressive royal flush jackpot. To
say that one has a 50/50 chance in hitting a royal flush in less
than 25,000 hands of video poker _is wrong_.
I stand by my post.
L.Wluiki
>
>(snip) take a look at this article by John
>
>> Robison in the Casino City Times, dated today --
>
>http://casinocitytimes.com/
>
>> Wanted to post the link here in case there are some of you who do
>
>not
>
>> subscribe.
>>
>> Cheers.
>>
>> Bill Velek
>
>*********************************************************************
>
>Following your link, I came across another article by John Robison
>which contained a common error, a misconception that is not immnuned
>in 'gambling gurus', not even one with a master's in computer
>science (see also Peter Griffin's _Extra Stuff_):
>
>re: your 15 January 2004 Casino City Times Article
>
>
>Mr. Robison:
>
>
>"Twenty-five thousand spins isn't really that many. If you were
>playing video poker, you'd have only just passed the 50/50 point of
>having hit a royal flush."
>
>You were wrong. The 50/50 point of having hit a royal flush occurs
>at (natural log 2)*Royal Flush Cycle.I may be misunderstanding something. Isn't the natural log of 2 the
power to which 10 must be raised to get 2? That's .30103.
No, the "natural log" is to base e rather than base 10.
But the
point in a cycle at which there is a 50% chance of hitting it occurs
about 69.3% of the way into it.
Correct. Note that ln(2) = 0.693147....
>For Jacks or Better, this point occurs at 27,997 hands;
>for full pay Deuces Wild, this point occurs at 31,387 hands;
>for 10/7 Double Bonus, this point occurs at 33,304 hands.Assuming these figures are right, which all seem to agree with my
69.3% figure, how does his conclusion that 25,000 spins can't be just
past the half-way point follow? The only criterion mentioned is
"video poker." There are plenty of video poker games in which the
royal cycle is less than 25,000/(69.3%) (approximately 36,000) hands.
Good point. Unless the article specifed both 50,000 and 25,000, it
isn't correct to assume that the figure of 25,000 was based on a
mean rather than a median.
>You failed to distinguish between the MEAN and the MEDIAN, and made
>the flawed observation that the 50/50 point occurs at half of the
>value of the MEAN.Is he saying that the mean is 1 cycle? If so, he's still assuming
that the cycle in the original situation is just under 50,000 hands
and not just under 36,000 hands. On what basis is he assuming that?
The cycle is the mean number of hands required to hit a royal.
The median is the number of hands required to give the player
a 50% chance of hitting a royal during that many hands of play.
The point is that cycle/2 is not the same as a 50/50 chance of
hitting the royal.
On Friday 23 January 2004 06:17 am, Tom Robertson wrote:
>--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:
>On the answer to Steve's question, you wrote:
--- In vpFREE@yahoogroups.com, Tom Robertson <thomasrrobertson@e...>
wrote:
(snip)>The only criterion mentioned is
> "video poker." There are plenty of video poker games in which the
> royal cycle is less than 25,000/(69.3%) (approximately 36,000)
>hands.(snip)
*********************************************************************
I just checked every game in WinPoker and failed to find one of your
"plenty of video poker games in which the royal cycle is less than
25,000/(69.3%) (approximately 36,000) hands".
8/5 JoB with 1000:1 royal payoff comes close with a cycle of 36443.
If one plays to minimize the average cost of playing for a royal, the
cycle drops to 32573, and this come with only a small decrease in
EV.
Robison was addressing a slots player, not some one who would adjust
strategy on a sufficiently high progressive royal flush jackpot. To
say that one has a 50/50 chance in hitting a royal flush in less
than 25,000 hands of video poker _is wrong_.
Well, that depends. A slots player who didn't take the time to learn
max-EV strategy would probably tend to try for the royal too often,
and could easily adopt a flawed strategy that would yield a median
smaller than 25,000 hands. Playing to maximize the probability
of hitting a royal (a very neg-EV strategy) can cut the cycle about
in half. Nudging the cycle down to 36,000 hands would likely be
possible for most VP games. As mentioned above, playing to minimize
the average cost of losses between royals can bring the cycle well
below 36,000. So, I think an unqualified "wrong" is much too strong
of a claim.
Now I wonder what the optimal strategy would be for maximizing
EV on 9/6 JoB when the player is constrained to have a royal
cycle of 36,000 hands or less 
Actually, I hope to eventually
develop direct methods for solving this kind of "constrained
optimization" problem.
On Friday 23 January 2004 06:24 am, lwluiki wrote:
Mr. Robison:
I agree with the facts presented by Steve Jacobs at vpFREE:
From: Steve Jacobs <jacobs@
Date: Fri Jan 23, 2004 6:10 am
Subject: Re: [vpFREE] Re: Guru Goof? (was Apparently some Indian
Casinos _ARE_ regulated.....
On Friday 23 January 2004 06:24 am, lwluiki wrote:
--- In vpFREE at yahoo gruops, Tom Robertson
<thomasrrobertson@e...>
wrote:
(snip)>The only criterion mentioned is
> "video poker." There are plenty of video poker games in which the
> royal cycle is less than 25,000/(69.3%) (approximately 36,000)
>hands.(snip)
*********************************************************************
I just checked every game in WinPoker and failed to find one of
your
"plenty of video poker games in which the royal cycle is less than
25,000/(69.3%) (approximately 36,000) hands".
8/5 JoB with 1000:1 royal payoff comes close with a cycle of 36443.
If one plays to minimize the average cost of playing for a royal, the
cycle drops to 32573, and this come with only a small decrease in
EV.
Robison was addressing a slots player, not some one who would
adjust
strategy on a sufficiently high progressive royal flush jackpot. To
say that one has a 50/50 chance in hitting a royal flush in less
than 25,000 hands of video poker _is wrong_.
Well, that depends. A slots player who didn't take the time to learn
max-EV strategy would probably tend to try for the royal too often,
and could easily adopt a flawed strategy that would yield a median
smaller than 25,000 hands. Playing to maximize the probability
of hitting a royal (a very neg-EV strategy) can cut the cycle about
in half. Nudging the cycle down to 36,000 hands would likely be
possible for most VP games. As mentioned above, playing to minimize
the average cost of losses between royals can bring the cycle well
below 36,000. So, I think an unqualified "wrong" is much too strong
of a claim.
Now I wonder what the optimal strategy would be for maximizing
EV on 9/6 JoB when the player is constrained to have a royal
cycle of 36,000 hands or less Actually, I hope to eventually
develop direct methods for solving this kind of "constrained
optimization" problem.
I did make "a much too strong of a claim".
It is I who is in need of a retraction.
I apologize, and will post this message to vpFREE.
Sincerely,
L.Wluiki
(snip)
8/5 JoB with 1000:1 royal payoff comes close with a cycle of 36443.
If one plays to minimize the average cost of playing for a royal,
the
cycle drops to 32573,
(snip)
As mentioned above, playing to minimize
--- In vpFREE@yahoogroups.com, Steve Jacobs <jacobs@x> wrote:
the average cost of losses between royals can bring the cycle well
below 36,000.
********************************************************************
Can you explain that a little? One would think that playing to
minimize the average cost of losses between royals would increase
the length of the royal cycle.
L.Wluiki
lwluiki wrote:
snip
... (see also Peter Griffin's _Extra Stuff_) ...
Where can that be found?
Thanks.
Bill Velek
lwluiki wrote:
snip
> ... (see also Peter Griffin's _Extra Stuff_) ...
Where can that be found?
Thanks.
Bill Velek
--- In vpFREE@yahoogroups.com, Bill Velek <billvelek@a...> wrote:
*********************************************************************
Extra Stuff - Gambling Ramblings, by Peter Griffin. Huntington
Press, 1991, with a preface by Anthony Curtis. It should be
available at Great Stuff For Gamblers, The Las Vegas Advisor website.
I cannot praise this collection of gambling wisdom more, highly
entertaining too. If I can only recommend one book, this is IT.
It's hard to believe he's no longer with us. He was my teacher, my
friend, and for a too brief of a time, my colleague.
L.Wluiki
Mr. Robison:
"Twenty-five thousand spins isn't really that many. If you were
playing video poker, you'd have only just passed the 50/50 point
of
having hit a royal flush."
You were wrong. The 50/50 point of having hit a royal flush occurs
at (natural log 2)*Royal Flush Cycle.For Jacks or Better, this point occurs at 27,997 hands;
for full pay Deuces Wild, this point occurs at 31,387 hands;
for 10/7 Double Bonus, this point occurs at 33,304 hands.You failed to distinguish between the MEAN and the MEDIAN, and
made
the flawed observation that the 50/50 point occurs at half of the
value of the MEAN.Please be kind enough to publish a retraction as a service to your
readers.Sincerely,
L. Wluiki
To calculate the probability, I use the following equation: .5 =
[(the royal flush cycle - 1) / the royal flush cycle]^N and solve
for N. Here, the royal flush cycle is dependent on the strategy
employeed.
Cheers.
--- In vpFREE@yahoogroups.com, "lwluiki" <lwluiki@a...> wrote:
On the answer to Steve's question, you wrote:
If Cy is the cycle length for a royal, and L is the average loss of
a single hand (conditioned on not being a royal payoff) then the
average cost between royals is:
C = (Cy - 1) * L
To minimize, you can't just minimize L, and you can't just minimize
Cy, it is the overall product that matters. The optimal strategy for
this objective can be found by pretending that the royal payoff is
just large enough to give a breakeven game. When the royal
payoff is sized such that the max-EV strategy with that "virtual"
payoff gives a breakeven game, then the corresponding playing
strategy minimizes the average cost between royals. This is quite
non-intuitive, but true. When the breakeven game is found, the
virtual payoff is exactly equal to the cost of playing between royals,
because in a breakeven game royals occur just often enough to
"pay" their own cost. In fact, a little thought shows that this property
holds for all payoffs -- breakeven means that each payoff occurs
just often enough to pay its own cost.
If you compute the optimal strategy and then compare the
result to the normal max-EV strategy, you find:
A) If the game is favorable to the player, the min_cost_royal
strategy will have a longer royal cycle than the max-EV
strategy. This is because we pretend the royal is worth less
than it really is, until we reach a breakeven game. Thus we
try for royals less often.
B) If the game is unfavorable to the player, the min_cost_royal
strategy will have a shorter royal cycle than the max-EV
strategy. Here we pretend the royal is worth more than it
really is, until we find the payoff that gives a breakeven game.
So, we play more agressively and try for royals more often.
The min_cost_royal can also be viewed in the following way:
As mentioned above, the non-royal payoffs occur just often
to pay their own way. This means the overall gain or loss can
be treated as if it is all due to the royal payoff being too large/small
to make the game breakeven. This is a special property that only
holds for this strategy, in fact it could be claimed that this is the
defining characteristic of this game. In a sense, all the plays
can be divided into two cases -- plays where we try for a royal,
and plays where we bide our time in order to survive until our
next opportunity to try for a royal. If we looked only at the cases
where we "bide our time," we would see a fair game which favors
neither the casino nor the player. It is possible to develop a
mathematical model that demonstrates this, but I'm not going
to take the time right now to wade through that.
Here is another thing that I find interesting. The min_cost_royal
strategy tends to be very close to the strategy which minimizes
risk (or RoR). I have developed a framework for thinking about VP
which makes this almost obvious, but it takes a lot of effort to
explain, so I won't attempt that here. But in both cases, the
optimal strategy plays less aggressively (compared to max-EV)
when the game is favorable and more agressively when the
game is unfavorable. This difference between min_cost_royal
and min-risk is that min_cost_royal concentrates all of the
"unfairness" into the cases when we try for a royal, while
min-risk assigns most of the unfairness to royals and distributes
the remainder of the unfairness to the other payoffs.
On Friday 23 January 2004 08:32 am, lwluiki wrote:
--- In vpFREE@yahoogroups.com, Steve Jacobs <jacobs@x> wrote:
(snip)
8/5 JoB with 1000:1 royal payoff comes close with a cycle of 36443.> If one plays to minimize the average cost of playing for a royal,
the
> cycle drops to 32573,
(snip)
As mentioned above, playing to minimize> the average cost of losses between royals can bring the cycle well
> below 36,000.********************************************************************
Can you explain that a little? One would think that playing to
minimize the average cost of losses between royals would increase
the length of the royal cycle.