Lets say you go to the casino every day and play 2000 hands of
deuces wild. However you have an odd machine. The paytables are
not shown and when you hit a winning hand the credits alotted to
that hand are not show either. All you know is when you have played
your 2000 hands and how many credits you are up or down for the day.
How many days would you have to play before you could tell which of
the many variations of deuceswild have you been playing?
DWK
none - I dont play stupid games at bad places like your example
----- Original Message -----
From: "deuceswild1000" <deuceswild1000@yahoo.com>
To: <vpFREE@yahoogroups.com>
Sent: Friday, August 08, 2003 1:19 PM
Subject: [vpFREE] Which game is it?
Lets say you go to the casino every day and play 2000 hands of
deuces wild. However you have an odd machine. The paytables are
not shown and when you hit a winning hand the credits alotted to
that hand are not show either. All you know is when you have played
your 2000 hands and how many credits you are up or down for the day.How many days would you have to play before you could tell which of
the many variations of deuceswild have you been playing?DWK
vpFREE Links: http://tinyurl.com/a7ga
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Bravo
--- tomflush <tomflush@nyc.rr.com> wrote:
none - I dont play stupid games at bad places like
your example----- Original Message -----
From: "deuceswild1000" <deuceswild1000@yahoo.com>
To: <vpFREE@yahoogroups.com>
Sent: Friday, August 08, 2003 1:19 PM
Subject: [vpFREE] Which game is it?> Lets say you go to the casino every day and play
2000 hands of
> deuces wild. However you have an odd machine.
The paytables are
> not shown and when you hit a winning hand the
credits alotted to
> that hand are not show either. All you know is
when you have played
> your 2000 hands and how many credits you are up or
down for the day.
>
> How many days would you have to play before you
could tell which of
> the many variations of deuceswild have you been
playing?
>
> DWK
>
>
>
> vpFREE Links: http://tinyurl.com/a7ga
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
>
>
__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com
wrote:
Lets say you go to the casino every day and play 2000 hands of
deuces wild. However you have an odd machine. The paytables are
not shown and when you hit a winning hand the credits alotted to
that hand are not show either. All you know is when you have played
your 2000 hands and how many credits you are up or down for the day.How many days would you have to play before you could tell which of
the many variations of deuceswild have you been playing?DWK
As others pointed out in a less than subtle way, this is a purely
theoretical scenario which no one (in their right mind) would actually
play. But there's nothing wrong with talking about the theory.
The answer, unfortunately, is many millions of hands. It takes
roughly 1 million hands of VP just to ensure a 95% chance of ended up
within +/-1% of the EV. So even if you're playing 100.7% FPDW, you
could end up at 99.7% return after 1 million hands, which would
(mis-)lead you to believe you were playing NSUD.
Of course, that's the edge of the probabilities. There's a much
greater chance that many hands of FPDW (played optimally, of course)
would end up closer to 100.7%. But there's no way to tell with near
100% certainty which version you're playing based on actual results
unless many millions of hands were played. And even then, there's
always that tiny chance of someone's results being way out on an
extreme end. That's the risk that we all just have to live with.
VPAnimal
--- In vpFREE@yahoogroups.com, "deuceswild1000" <deuceswild1000@y...>
> How many days would you have to play before you could tell which of
> the many variations of deuceswild have you been playing?
You could probably know after one day if you kept track of every winning hand during that session and, using all known deuces wild paytables, calculated what your final credits should be. Chances are that only one would match up exactly.
-a
This was a theoretical question. It makes as much sense to me as
discussing how much water evaporates in Las Vegas.
Of course the answer is a million hands or more.
As you can see some people do not realize how much overlap there is
in the various games.
DWK
--- In vpFREE@yahoogroups.com, "tomflush" <tomflush@n...> wrote:
none - I dont play stupid games at bad places like your example
Your theoretical problem turns out to be "ill-formed" because as stated, it
has too many degrees of freedom.
You say the paytables are not shown, but you
do not say that it is "Full Pay" deuces wild. Thus it could be any version ,
even ones that are not well known. ( For example royal pays
10,000-1 ). Lets assume its one of the 20 most popular version of deuces (
NSUD, FP DEUCES, AC Deuces, airport Deuces, double pay deuces, triple pay
deuces, Rio deuces, deuces dlux, deuces delux with 4700 RF, Illinois
deuces,....etc ). Every one of these games have an EV1.....EV20. So far, so
good, however, some of the games are within .1 of each other - but the EV's
can even be the same at this point. We now must decide what strategy to play
the unknown game. A good place to start is using Full Pay Deuces strategy
assuming that this is the game we are up against. Because of the proximities
of the EV's, and you would have to have a cross reference chart to see the
payback of FP deuces strategy vs ( EV1...EV20). Some of the paybacks of
running strategy X against game Y may be very precise and this riddle is
solvable IFF none of these EV delta's are exactly the same. You have to
consider the smallest fractional difference of [EVRX - EVRY] and base your
samples on that. You also have to factor in "perfect play". You must assume
you make 0 mistakes , as players do periodically shift EV due to physical
factors such as eye strain, key stickiness, etc. The rest is statistics and
how close you want to approximate an answer. It most likely would take
millions of hands to achieve a 99.99% confidence level.
Now one final note - Not all machines in the real world have paytables
clearly displayed, but all real world machines tell you how many coins you
WON - even old coin droppers which don't display credits. There is a machine
at the Tropicana that says only "Jacks or Better" - no pay schedule visible.
It turns out to be 9/6 Jacks but you really have to play long enough to get
a full house and a flush to know for sure!
regards...Tom
----- Original Message -----
From: "deuceswild1000" <deuceswild1000@yahoo.com>
To: <vpFREE@yahoogroups.com>
Sent: Friday, August 08, 2003 1:19 PM
Subject: [vpFREE] Which game is it?
Lets say you go to the casino every day and play 2000 hands of
deuces wild. However you have an odd machine. The paytables are
not shown and when you hit a winning hand the credits alotted to
that hand are not show either. All you know is when you have played
your 2000 hands and how many credits you are up or down for the day.How many days would you have to play before you could tell which of
the many variations of deuceswild have you been playing?DWK
Your theoretical problem turns out to be "ill-formed" because as
stated, it
has too many degrees of freedom.
You say the paytables are not shown, but you
do not say that it is "Full Pay" deuces wild. Thus it could be any
version ,
even ones that are not well known. ( For example royal pays
10,000-1 ). Lets assume its one of the 20 most popular version of
deuces (
NSUD, FP DEUCES, AC Deuces, airport Deuces, double pay deuces,
triple pay
deuces, Rio deuces, deuces dlux, deuces delux with 4700 RF,
Illinois
deuces,....etc ). Every one of these games have an EV1.....EV20.
So far, so
good, however, some of the games are within .1 of each other - but
the EV's
can even be the same at this point. We now must decide what
strategy to play
the unknown game. A good place to start is using Full Pay Deuces
strategy
assuming that this is the game we are up against. Because of the
proximities
of the EV's, and you would have to have a cross reference chart to
see the
payback of FP deuces strategy vs ( EV1...EV20). Some of the
paybacks of
running strategy X against game Y may be very precise and this
riddle is
solvable IFF none of these EV delta's are exactly the same. You
have to
consider the smallest fractional difference of [EVRX - EVRY] and
base your
samples on that. You also have to factor in "perfect play". You
must assume
you make 0 mistakes , as players do periodically shift EV due to
physical
factors such as eye strain, key stickiness, etc. The rest is
statistics and
how close you want to approximate an answer. It most likely would
take
millions of hands to achieve a 99.99% confidence level.
Now one final note - Not all machines in the real world have
paytables
clearly displayed, but all real world machines tell you how many
coins you
WON - even old coin droppers which don't display credits. There is
a machine
at the Tropicana that says only "Jacks or Better" - no pay
schedule visible.
It turns out to be 9/6 Jacks but you really have to play long
enough to get
a full house and a flush to know for sure!
regards...Tom
Tom, of course your answer is procedurely and statisitcallly
correct. I concede that my hypothetical game could not be
effectively played to perfect play with out paytable knowledge and
thus not allow one to arrive at an answer. The whole point was to
drive a discussion regarding the inability to determine a difference
of a percent or two in EV without playing well over a million
hands. John Robison wrote an article on this very subject.
If you think that is a worthwhile subject for newcomers to this
group (and lets face it, most do not go back and read posts of a
year or so ago) then I would appreciate you helping to explain it to
a non-statistically oriented person.
If you do not feel that it is a subject worthy of further discussion
then just let a sleeping dog lie.
Thanks,
DWK
--- In vpFREE@yahoogroups.com, "tomflush" <tomflush@n...> wrote: