vpFREE2 Forums

Some VERY Disturbing Statistics

#1

First of all, I have to warn you -- I'm both a math guy and a
compulsive scorekeeper so if you're bored by numbers, you may want to
stop reading right now. In fact, my wife claims that the only reason
I play VP is so that I'll have an excuse to create a spreadsheet.
That, however, is only partially true.

2008, what a frustrating year for VP! My wife and I play Full Pay
Deuces Wild at Red Rock every day. The year started off great. In
the first 3 months of the year, we played 486k hands and had a 2.22%
profit-- nearly triple the expected profit margin of .76%. This was
going to be a very good year ... we thought. Then the bottom dropped
out. In the last 9 months of 2008, we played 934k hands with a loss
of .52%. That may not sound like much but it's more than 1-1/4%
below expected with well over $1 million run through the machine (you
do the math).

What, I wondered, is the probability of showing that much of a loss
for that many hands? I was able to use the bankroll calculation
feature of Video Poker for Winners to determine that the probability
is .62%. In fact, there's only about a 7% probability of showing any
loss at all in that many hands.

So, what happened? Why such poor results for a 9-month period? The
answer, of course: royal flushes (or lack thereof). As I said, we
got off to a great start for the first 3 months of the year. In
January, we had 7 royals in 2.9 RCs (royal cycles = 45,282 hands).
In February we had 7 more royals in 3.8 RCs. And, in March we had 5
royals in 4.1 RCs. We knew we were seriously overachieving with 19
royals in 10.8 RCs but it was fun while it lasted.

What are the odds, you may ask, of getting at least 5 royals in 3
consecutive months playing the number of hands we played each month?
The answer turns out to be 2.07% so that was a very lucky stretch.

But then, we underachieved for the first time in April when we had
only 2 royals in 3.5 RCs. We were out of town for most of the month
of May but managed to hit 1 royal in only .5 RCs. The real shocker
came in June when we had 0 royals in 2.4 RCs. July wasn't very good
either with 2 royals in 3.4 RCs. Then another really bad month in
August with 1 royal in 3.0 RCs. September wasn't too bad with 2
royals in 2.2 RCs. Another bad month in October with 1 royal in 2.1
RCs. November was OK with 2 royals in 1.8 RCs. We had hopes of
ending the year well but had another 0 royal month in December
playing 1.8 RCs.

So, from April through December, just 11 royals in 20.7 RCs. That
was bad enough but the thing that really struck me was the
consistency of the poor results -- we never had more than 2 royals in
any of the 9 months. The probability of getting 2 or fewer royals in
9 consecutive months playing the number of hands we played each month
is .61%. Note that this is virtually the same number as the
probability of losing at the percentage rate we experienced which is
not surprising since we're looking at 2 different ways to express
essentially the same (poor) result. In contrast, if we expand to 3
or fewer royals for the 9 months, the probability jumps to 8.56%.

Finally, one more disturbing statistic. Looking back at the royals
that we did get, only one of them resulted from holding 4 and drawing
one. No wonder we were underachieving on royals -- one royal when
drawing to 4-to-a-royal in 934k hands! I calculate that you should
be dealt 4-to-a-royal (without the deuce) once every 3,022 hands on
average. So that means we had one success in a little over 300 draws.

By the way, we are currently at 125k hands (almost 3 RCs) since our
last royal so our results are actually getting worse (if that's
possible).

Back around September or so, having hit just 3 royals in the
preceding 3 months, my wife asked me if I was sure they hadn't "done
something" to the machines. I laughed and told her that they
couldn't do that because it would be illegal. I'm here to tell you
that I'm not laughing any more. I don't know about you but when
we're getting results that are way less than 1% probable and
approaching 3 standard deviations below the mean, I'm having trouble
attributing it to "a little run of bad luck."

#2

Similarly, I just got my win loss from one of the local places. I put $1,700,000 through primarily 4 different $1 machines in 2008. From Jan to July I had $132,000 in W2Gs. You do the math, but this is a higher return than expected.

With an increase in play from Aug to Dec, I got a total of $24,000 in W2Gs on the same 4 machines!

Duh, ya think?

First of all, I have to warn you -- I'm both a math guy and a compulsive scorekeeper so if you're bored by numbers, you may want to stop reading right now. In fact, my wife claims that the only reason I play VP is so that I'll have an excuse to create a spreadsheet. That, however, is only partially true.2008, what a frustrating year for VP! My wife and I play Full Pay Deuces Wild at Red Rock every day. The year started off great. In the first 3 months of the year, we played 486k hands and had a 2.22% profit-- nearly triple the expected profit margin of .76%. This was going to be a very good year ... we thought. Then the bottom dropped out. In the last 9 months of 2008, we played 934k hands with a loss of .52%. That may not sound like much but it's more than 1-1/4% below expected with well over $1 million run through the machine (you do the math).What, I wondered, is the probability of showing that much of a loss for that many hands? I was able to use the bankroll calculation feature of Video Poker for Winners to determine that the probability is .62%. In fact, there's only about a 7% probability of showing any loss at all in that many hands.So, what happened? Why such poor results for a 9-month period? The answer, of course: royal flushes (or lack thereof). As I said, we got off to a great start for the first 3 months of the year. In January, we had 7 royals in 2.9 RCs (royal cycles = 45,282 hands). In February we had 7 more royals in 3.8 RCs. And, in March we had 5 royals in 4.1 RCs. We knew we were seriously overachieving with 19 royals in 10.8 RCs but it was fun while it lasted.What are the odds, you may ask, of getting at least 5 royals in 3 consecutive months playing the number of hands we played each month? The answer turns out to be 2.07% so that was a very lucky stretch.But then, we underachieved for the first time in April when we had only 2 royals in 3.5 RCs. We were out of town for most of the month of May but managed to hit 1 royal in only .5 RCs. The real shocker came in June when we had 0 royals in 2.4 RCs. July wasn't very good either with 2 royals in 3.4 RCs. Then another really bad month in August with 1 royal in 3.0 RCs. September wasn't too bad with 2 royals in 2.2 RCs. Another bad month in October with 1 royal in 2.1 RCs. November was OK with 2 royals in 1.8 RCs. We had hopes of ending the year well but had another 0 royal month in December playing 1.8 RCs.So, from April through December, just 11 royals in 20.7 RCs. That was bad enough but the thing that really struck me was the consistency of the poor results -- we never had more than 2 royals in any of the 9 months. The probability of getting 2 or fewer royals in 9 consecutive months playing the number of hands we played each month is .61%. Note that this is virtually the same number as the probability of losing at the percentage rate we experienced which is not surprising since we're looking at 2 different ways to express essentially the same (poor) result. In contrast, if we expand to 3 or fewer royals for the 9 months, the probability jumps to 8.56%.Finally, one more disturbing statistic. Looking back at the royals that we did get, only one of them resulted from holding 4 and drawing one. No wonder we were underachieving on royals -- one royal when drawing to 4-to-a-royal in 934k hands! I calculate that you should be dealt 4-to-a-royal (without the deuce) once every 3,022 hands on average. So that means we had one success in a little over 300 draws.By the way, we are currently at 125k hands (almost 3 RCs) since our last royal so our results are actually getting worse (if that's possible).Back around September or so, having hit just 3 royals in the preceding 3 months, my wife asked me if I was sure they hadn't "done something" to the machines. I laughed and told her that they couldn't do that because it would be illegal. I'm here to tell you that I'm not laughing any more. I don't know about you but when we're getting results that are way less than 1% probable and approaching 3 standard deviations below the mean, I'm having trouble attributing it to "a little run of bad luck."

···

To: vpFREE@yahoogroups.comFrom: alpine205@cox.netDate: Thu, 8 Jan 2009 00:17:05 +0000Subject: [vpFREE] Some VERY Disturbing Statistics

_________________________________________________________________
Windows LiveTM Hotmail®: Chat. Store. Share. Do more with mail.
http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_hm_justgotbetter_explore_012009

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

#3

The problem with your analysis is that your separation of the (good)
first part of the year from the (bad) second part of the year is
arbitrary. That makes it difficult if not impossible to draw any
conclusion from the downturn in your results.

Am I correct in assuming you did not think they were cheating in your
favor for the first part of the year? ;>) Those results were just
about as "disturbing" as your subsequent slump. In one case you were
1.14% below ev, in the other you were 1.46% ABOVE ev.

But anyway, look at your whole year! Using your figs, I get that you
won at the rate of 0.42% for all of 2008 and showed a profit of about
$7400. There's nothing disturbing about that figure. There's about
a 20% chance of winning $7400 or worse in the number of hands you
played.

If you isolate streaks, good OR bad, you will find stats that can
satisfy almost any hypothesis. For that matter, if you play long
enough you will find yourself 3 standard deviations behind EV sooner
or later.

The only way to objectively test whether things are not legit is to
set up your test ahead of time. Something like: "I'm going to play
300K hands; if I'm ever $5000 behind, I will conclude they are
cheating, because there should be only a 2% chance of that happening
by luck alone." With a fixed starting point and fixed stopping
point, the results have more meaning.

--Dunbar

First of all, I have to warn you -- I'm both a math guy and a
compulsive scorekeeper so if you're bored by numbers, you may want

to

stop reading right now. In fact, my wife claims that the only

reason

I play VP is so that I'll have an excuse to create a spreadsheet.
That, however, is only partially true.

2008, what a frustrating year for VP! My wife and I play Full Pay
Deuces Wild at Red Rock every day. The year started off great. In
the first 3 months of the year, we played 486k hands and had a

2.22%

profit-- nearly triple the expected profit margin of .76%. This

was

going to be a very good year ... we thought. Then the bottom

dropped

out. In the last 9 months of 2008, we played 934k hands with a

loss

of .52%. That may not sound like much but it's more than 1-1/4%
below expected with well over $1 million run through the machine

(you

do the math).

What, I wondered, is the probability of showing that much of a loss
for that many hands? I was able to use the bankroll calculation
feature of Video Poker for Winners to determine that the

probability

is .62%. In fact, there's only about a 7% probability of showing

any

loss at all in that many hands.

So, what happened? Why such poor results for a 9-month period?

The

answer, of course: royal flushes (or lack thereof). As I said, we
got off to a great start for the first 3 months of the year. In
January, we had 7 royals in 2.9 RCs (royal cycles = 45,282 hands).
In February we had 7 more royals in 3.8 RCs. And, in March we had

5

royals in 4.1 RCs. We knew we were seriously overachieving with 19
royals in 10.8 RCs but it was fun while it lasted.

What are the odds, you may ask, of getting at least 5 royals in 3
consecutive months playing the number of hands we played each

month?

The answer turns out to be 2.07% so that was a very lucky stretch.

But then, we underachieved for the first time in April when we had
only 2 royals in 3.5 RCs. We were out of town for most of the

month

of May but managed to hit 1 royal in only .5 RCs. The real

shocker

came in June when we had 0 royals in 2.4 RCs. July wasn't very

good

either with 2 royals in 3.4 RCs. Then another really bad month in
August with 1 royal in 3.0 RCs. September wasn't too bad with 2
royals in 2.2 RCs. Another bad month in October with 1 royal in

2.1

RCs. November was OK with 2 royals in 1.8 RCs. We had hopes of
ending the year well but had another 0 royal month in December
playing 1.8 RCs.

So, from April through December, just 11 royals in 20.7 RCs. That
was bad enough but the thing that really struck me was the
consistency of the poor results -- we never had more than 2 royals

in

any of the 9 months. The probability of getting 2 or fewer royals

in

9 consecutive months playing the number of hands we played each

month

is .61%. Note that this is virtually the same number as the
probability of losing at the percentage rate we experienced which

is

not surprising since we're looking at 2 different ways to express
essentially the same (poor) result. In contrast, if we expand to 3
or fewer royals for the 9 months, the probability jumps to 8.56%.

Finally, one more disturbing statistic. Looking back at the royals
that we did get, only one of them resulted from holding 4 and

drawing

one. No wonder we were underachieving on royals -- one royal when
drawing to 4-to-a-royal in 934k hands! I calculate that you should
be dealt 4-to-a-royal (without the deuce) once every 3,022 hands on
average. So that means we had one success in a little over 300

draws.

By the way, we are currently at 125k hands (almost 3 RCs) since our
last royal so our results are actually getting worse (if that's
possible).

Back around September or so, having hit just 3 royals in the
preceding 3 months, my wife asked me if I was sure they

hadn't "done

something" to the machines. I laughed and told her that they
couldn't do that because it would be illegal. I'm here to tell you
that I'm not laughing any more. I don't know about you but when
we're getting results that are way less than 1% probable and
approaching 3 standard deviations below the mean, I'm having

trouble

···

--- In vpFREE@yahoogroups.com, "alpine205" <alpine205@...> wrote:

attributing it to "a little run of bad luck."

#4

Dunbar pretty much said it all. I would only add that hitting
30 royals in about 31 royal cycles over the course of a year sounds
more or less like par for the course.

The problem with your analysis is that your separation of the

(good)

first part of the year from the (bad) second part of the year is
arbitrary. That makes it difficult if not impossible to draw any
conclusion from the downturn in your results.

Am I correct in assuming you did not think they were cheating in

your

favor for the first part of the year? ;>) Those results were just
about as "disturbing" as your subsequent slump. In one case you

were

1.14% below ev, in the other you were 1.46% ABOVE ev.

But anyway, look at your whole year! Using your figs, I get that

you

won at the rate of 0.42% for all of 2008 and showed a profit of

about

$7400. There's nothing disturbing about that figure. There's

about

a 20% chance of winning $7400 or worse in the number of hands you
played.

If you isolate streaks, good OR bad, you will find stats that can
satisfy almost any hypothesis. For that matter, if you play long
enough you will find yourself 3 standard deviations behind EV

sooner

or later.

The only way to objectively test whether things are not legit is

to

set up your test ahead of time. Something like: "I'm going to

play

300K hands; if I'm ever $5000 behind, I will conclude they are
cheating, because there should be only a 2% chance of that

happening

by luck alone." With a fixed starting point and fixed stopping
point, the results have more meaning.

--Dunbar

>
> First of all, I have to warn you -- I'm both a math guy and a
> compulsive scorekeeper so if you're bored by numbers, you may

want

to
> stop reading right now. In fact, my wife claims that the only
reason
> I play VP is so that I'll have an excuse to create a

spreadsheet.

> That, however, is only partially true.
>
> 2008, what a frustrating year for VP! My wife and I play Full

Pay

> Deuces Wild at Red Rock every day. The year started off great.

In

> the first 3 months of the year, we played 486k hands and had a
2.22%
> profit-- nearly triple the expected profit margin of .76%.

This

was
> going to be a very good year ... we thought. Then the bottom
dropped
> out. In the last 9 months of 2008, we played 934k hands with a
loss
> of .52%. That may not sound like much but it's more than 1-1/4%
> below expected with well over $1 million run through the machine
(you
> do the math).
>
> What, I wondered, is the probability of showing that much of a

loss

> for that many hands? I was able to use the bankroll calculation
> feature of Video Poker for Winners to determine that the
probability
> is .62%. In fact, there's only about a 7% probability of

showing

any
> loss at all in that many hands.
>
> So, what happened? Why such poor results for a 9-month period?
The
> answer, of course: royal flushes (or lack thereof). As I said,

we

> got off to a great start for the first 3 months of the year. In
> January, we had 7 royals in 2.9 RCs (royal cycles = 45,282

hands).

> In February we had 7 more royals in 3.8 RCs. And, in March we

had

5
> royals in 4.1 RCs. We knew we were seriously overachieving with

19

> royals in 10.8 RCs but it was fun while it lasted.
>
> What are the odds, you may ask, of getting at least 5 royals in

3

> consecutive months playing the number of hands we played each
month?
> The answer turns out to be 2.07% so that was a very lucky

stretch.

>
> But then, we underachieved for the first time in April when we

had

> only 2 royals in 3.5 RCs. We were out of town for most of the
month
> of May but managed to hit 1 royal in only .5 RCs. The real
shocker
> came in June when we had 0 royals in 2.4 RCs. July wasn't very
good
> either with 2 royals in 3.4 RCs. Then another really bad month

in

> August with 1 royal in 3.0 RCs. September wasn't too bad with 2
> royals in 2.2 RCs. Another bad month in October with 1 royal in
2.1
> RCs. November was OK with 2 royals in 1.8 RCs. We had hopes of
> ending the year well but had another 0 royal month in December
> playing 1.8 RCs.
>
> So, from April through December, just 11 royals in 20.7 RCs.

That

> was bad enough but the thing that really struck me was the
> consistency of the poor results -- we never had more than 2

royals

in
> any of the 9 months. The probability of getting 2 or fewer

royals

in
> 9 consecutive months playing the number of hands we played each
month
> is .61%. Note that this is virtually the same number as the
> probability of losing at the percentage rate we experienced

which

is
> not surprising since we're looking at 2 different ways to

express

> essentially the same (poor) result. In contrast, if we expand

to 3

> or fewer royals for the 9 months, the probability jumps to 8.56%.
>
> Finally, one more disturbing statistic. Looking back at the

royals

> that we did get, only one of them resulted from holding 4 and
drawing
> one. No wonder we were underachieving on royals -- one royal

when

> drawing to 4-to-a-royal in 934k hands! I calculate that you

should

> be dealt 4-to-a-royal (without the deuce) once every 3,022 hands

on

> average. So that means we had one success in a little over 300
draws.
>
> By the way, we are currently at 125k hands (almost 3 RCs) since

our

> last royal so our results are actually getting worse (if that's
> possible).
>
> Back around September or so, having hit just 3 royals in the
> preceding 3 months, my wife asked me if I was sure they
hadn't "done
> something" to the machines. I laughed and told her that they
> couldn't do that because it would be illegal. I'm here to tell

you

···

--- In vpFREE@yahoogroups.com, "dunbar_dra" <h_dunbar@...> wrote:

--- In vpFREE@yahoogroups.com, "alpine205" <alpine205@> wrote:
> that I'm not laughing any more. I don't know about you but when
> we're getting results that are way less than 1% probable and
> approaching 3 standard deviations below the mean, I'm having
trouble
> attributing it to "a little run of bad luck."
>

#5

Thank you for your comments. I understand what you're saying but I
think you've missed the point. You're saying that it only matters how
you end up, it doesn't matter how you got there. I think it matters
very much how you got there.

2008 is history. I'm looking forward to 2009 now. What can I
reasonably expect? A modest profit like last year or a continuation of
the last 9 months. I don't know, but I believe that I have good reason
to voice some concern.

#6

If I am understanding you correctly, you are voicing concern over a 9-month
period where the chances you would achieve your results were 0.6%. This
means 6 out of every 1000 people would have this happen to them on perfectly
random machines. Why do you think this is indicative of a problem? 6 out
of 1000 is not all that uncommon. It sucks that you ended up on the bad
tail of the probability curve, but it happens. The odds would have to be
dramatically worse before I would start to think there is even a slight
chance that anything funny was going on.

Most people have trouble accepting unlikely events that go against them.
Rarely do we hear them struggle with unlikely events that go in their favor.

···

On Thu, Jan 8, 2009 at 8:11 AM, alpine205 <alpine205@cox.net> wrote:

  Thank you for your comments. I understand what you're saying but I
think you've missed the point. You're saying that it only matters how
you end up, it doesn't matter how you got there. I think it matters
very much how you got there.

2008 is history. I'm looking forward to 2009 now. What can I
reasonably expect? A modest profit like last year or a continuation of
the last 9 months. I don't know, but I believe that I have good reason
to voice some concern.

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

#7

alpine205 wrote:

Thank you for your comments. I understand what you're saying but I
think you've missed the point. You're saying that it only matters
how you end up, it doesn't matter how you got there. I think it
matters very much how you got there.

2008 is history. I'm looking forward to 2009 now. What can I
reasonably expect? A modest profit like last year or a
continuation of the last 9 months. I don't know, but I believe
that I have good reason to voice some concern.

I'll spin what Dunbar had to say a little differently ...

I think what should be taken most to heart is his suggestion that any
valid test of "fairness" can only be established by specifically
defining the test in advance -- the specific measure (return, number
of royals, etc.) and the duration of the test.

Now, I'll admit that it seems rather arbitrary to say that a run that
has less than 1% probability that's been observed in the past is
pretty much insignificant, but observing it in the future is
significant. An analogy might help.

Someone walks into your workplace, charred to a crisp. He cries out,
"The gods have cursed me ... I'm doomed." Turns out he was just
struck by lighting, for the second time in his life.

There's a distinct possibility he's right. But the truth is, shit
happens ... some of it quite improbable. For the population as a
whole, it's definitely known to occur. My call is that this poor sod
just happens to be the one to suffer it.

The point is that when something very unfortunate happens, you're
definitely going to hear about it. It just isn't a strong indication
that something or someone conspired against him/her.

···

------

There's something else at work here: It's not unusual for someone to
distinguish their results one year from another. But if it happened
that you hadn't begun your record keeping until July, things might
have seemed far less unusual. It would be expected you would have
arrived at a different conclusion about your experience, and be less
concerned about the results you're likely to experience going forward.
The same might be said if you had run your measure for the last two
years. The fact that you observed something very anomalous looking
back for a specific timeframe makes it far less exceptional.

------

Finally, video poker is phenomenally streaky. That relates directly
to the disparate frequency of various hands.

If your results can be explained predominantly by your experience with
royal hits, it's far less exception if it should be explained by vast
differences in the hits on more frequently occurring hands.

This doesn't dismiss your results in being unexceptional. It's
intended to illustrate that the significance of your results when
stated as an aggregate return, even if measured prospectively, can
vary tremendously.

------

Do you have reason for concern? Perhaps. More so if you ask around
those who you play with and find strong evidence that others have
predominantly suffered lackluster results. (Records would be the best
measure; memories are notably skewed.)

But that concern gains you nothing more than reason to construct a
test of a bias in machine return, and to carefully measure. Of
course, we know that no casino would consistently gaffe the same
machines for an extended period. Hell, if I were to choose machines
to gaffe, I'd definitely go for slot machines -- which of those
players are going to be so sensitive to their results to notice 2%
swiped away from a 7%-9% house hold.

My advice, bottom line: If you have the least concern about machine
fairness, don't play there. It's a certainty that any continued
anomaly is going to be good for a real head case.

- H.

#8

In a related issue, let's say your first guess is that a certain
machine has a 99% chance of a top cycle time of x but a 1% chance that
the top cycle is actually 2x. You then play 9 cycles but only get 3
top hits. Using the Poisson distribution, the probability of 3 hits in
9 cycles is 1.5% and the probability of 3 hits in 4.5 cycles (if the
cycle is really 2x) is 16.9%. Using Bayes' theorem, the new adjusted
guess is that the machine has a 90% chance of a top cycle time of x
and a 10% chance that the top cycle is actually 2x.

Poisson probability = exp(-cycles) x cycles^hits / hits!

Bayes theorem: p(B|A) = p(A|B)p(B) / (p(A|B)p(B) + p(A|/B)p(/B))

B = top cycle time is x, /B = top cycle time is 2x
p(B) = 99%, p(/B) = 1%
A = 3 hits
p(A|B) = 1.5%
p(A|/B) = 16.9%

#9

No matter how experienced players are there will be times when the length of losing streaks will make them start talking to themselves. I've been there!! We had the worse 8-month losing spell we EVER suffered in 23 year during the middle of 2008. However, we ended up having one of our better years when we closed the books on December 31st.

You must look at the long-term - and it is longer than most people can accept - and you must not get off-track by short term streaks, no matter how long they seem to last.

And the problem here is that the longer you play the more likely it is that you will hit these extremely long losing streaks. That is why we always hold on to our long-term figures. Most of us like to look at a calendar year - but the long term can be much longer than that, no matter how long you have been playing.

Volatility is the name of the game - perhaps ones of the worse - and best - aspects of the game - and the hardest for players to cope with!

···

________________
Jean $�ott, Frugal Gambler
http://queenofcomps.com/
You can read my blog at
http://lasvegasadvisor.com/blogs/jscott/

#10

If you play N0 hands, taking into account an error rate, you can
expect to come out ahead 84% of the time. At double N0 your
expectation goes to 98%, at triple N0 your expectation goes to 99.9%.
N0 = variance/advantage^2 hands. For a fair FPDW with no error rate
the N0 is about 450,000 hands. With an error rate of 0.25%, N0 jumps
to a little over a million hands.

Regarding your concern, I would not assume that a game is fair with
100% probability. At some point you should consider giving up, if your
results continue to be unfavorable. You could document and submit your
results to gaming and they might or might not take action. I wouldn't
put a lot of hope on gaming, but a complaint may start an
investigation that may uncover something, that's how American Coin was
brought to light.

···

--- In vpFREE@yahoogroups.com, "alpine205" <alpine205@...> wrote:

2008 is history. I'm looking forward to 2009 now. What can I
reasonably expect? A modest profit like last year or a continuation of
the last 9 months. I don't know, but I believe that I have good reason
to voice some concern.

#11

It doesn't work on reel slots. People tend to play a certain amount
and then walk, so tighter slots just decreases play time, the net hold
remains the same and can even decrease if players notice the machines
are tighter. Video poker and blackjack on the other hand has much more
potential since most players assume a fair deal and tend to attribute
bad results to just bad luck or player errors. Players judge tightness
by the paytable, but that only works when the game is fair. A loose
paytable and an unfair deal has enormous earning potential for the
casino with very little regulatory downside risk. In the case of
American Coin, the programmer got shot before he could testify and the
owners got slapped with a million dollar fine or something like that -
how many millions do you think they netted? How much do they make now
in consulting fees to non-Nevada casinos? The people involved in the
Venetian rigged drawing all got jobs out of state, of course their
talents are in demand. Is Ron Harris really having that much trouble
finding a job?
http://en.wikipedia.org/wiki/Ronald_Dale_Harris

···

--- In vpFREE@yahoogroups.com, "Harry Porter" <harry.porter@...> wrote:

Hell, if I were to choose machines
to gaffe, I'd definitely go for slot machines -- which of those
players are going to be so sensitive to their results to notice 2%
swiped away from a 7%-9% house hold.

#12

--- In vpFREE@yahoogroups.com, "Harry Porter" <harry.porter@...>
wrote:
...

My advice, bottom line: If you have the least concern about machine
fairness, don't play there. It's a certainty that any continued
anomaly is going to be good for a real head case.

- H.

I have assumed that machines are not gaffed, because what casino
would risk its licence to make a few dollars more? However in the
present climate, desperate casino management might consider such a
course when faced with loss of employment, debt default or
bankruptcy.

Has anyone experience in calling in the Gaming Comission to test a
machine? A freind mentioned that he did that several years ago using
a nonplaying proxy so that casino would not know it was him. [The
machine in question tested ok.] I see nothing wrong in calling the
Gaming Comission; after all, before casinos pay large royals, I have
seen them remove the mother board from the machine to check that it
was not tampered with and was properly functioning.

David

#13

Talk about streaks....Jenn and I play S/L JOB in the $1-2 denomination. Our play is slower than most on this board because we do double after wins so we average about 300 hands per hour. We make 2 trips/month to Atlantic City averaging 10 hours play per trip. 6000 hands /month minimum @ $5-10 per hand. (72000 hands per year in AC) We also log 4 vegas trips per year in addition to those plays where we average the same hand count per hour for 40 hours per trip.) another 12000 hands per year. So we play roughly 85K hands per year total. Our last royal was at Bally's in AC in January of 2004, 5 years ago this week. We have played over 400K hands without ever seeing a dealt royal or even drawing a royal. Up until 2004, the previous 4 years we averaged 1-2 royals per year on the same amount of play albeit on quarters at that time. I am anxiously waiting for the math to catch up to the reality.
Luckily, we finished the year within 1K of breakeven without ever seeing the monster hand.Remember ..... life is only what you make it. Never settle for second best. Bill

···

_________________________________________________________________
Windows Live™ Hotmail®: Chat. Store. Share. Do more with mail.
http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_hm_justgotbetter_explore_012009

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

#14

Ok....we are convinced the Casino will not purposely "fix" machines and thus risk their license. We probably all agree with that. We recently played at casino X and experinced a poor hit rate after approximately 5000 hands on various machines. We had had other similar experiences at the same casino. I wrote an e-mail to the casino manager and told them our last trip was in fact our last trip because I did not have confidence in their VP machines. He called me within 1 hour of the e-mail. He stated, (paraphrased) [ well I will admit that some games in some casinos are set at a higher hold rate than others, while still other games have a lower hold rate. So I guess it just depends on which machines you happen to playing on. ] By hold rate he meant the amount of percentage the casino makes off of each machine. He went on to say (paraphrased) [ the state allows us to set the hold rate between a certain percentage range.
That percentage may vary at anytime as long as it stays within the range] After the phone call I sent an e-mail to my casino host in response to his asking us to reconsider. I replied with a lengthy response about the hold rate being way to high on the VP machines and until it changes we would not be back. I asked him to provide me the record of percentage of payback overall for the casino and his response taken directly from his e-mail to me is as follows:

“Thank you again we do not report the percentages of hold but are held to a certain percentage range by law. I don't know the range off hand but I will check with Operations tomorrow to get the range. You can probably look it up on line some where?”

Question: does any of this mean anything or were they just talking about paytables and that lower percentage paytables equal higher hold rates or does it mean the hit rates of paying hands are lowered while the paytable stays the same?

Even VP with lower paytables should have about the same hit or non-hit rate as machines with good paytables. We have experienced very high no hit rates at this particular casino while playing good paytables. We have found by practicing at home that an average no hit rate is approximately between 52% and 54% after playing 5000 hands.

I have assumed that machines are not gaffed, because what casino
would risk its licence to make a few dollars more? However in the
present climate, desperate casino management might consider such a
course when faced with loss of employment, debt default or
bankruptcy.

Has anyone experience in calling in the Gaming Comission to test a
machine? A freind mentioned that he did that several years ago using
a nonplaying proxy so that casino would not know it was him. [The
machine in question tested ok.] I see nothing wrong in calling the
Gaming Comission; after all, before casinos pay large royals, I have
seen them remove the mother board from the machine to check that it
was not tampered with and was properly functioning.

David

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

#15

--- In vpFREE@yahoogroups.com, "nightoftheiguana2000"
<nightoftheiguana2000@...> wrote:

Regarding your concern, I would not assume that a game is fair with
100% probability. At some point you should consider giving up, if your
results continue to be unfavorable. You could document and submit your
results to gaming and they might or might not take action. I wouldn't
put a lot of hope on gaming, but a complaint may start an
investigation that may uncover something, that's how American Coin was
brought to light.

I read where a financial analyst wrote a letter years ago to the SEC
stating categorically that Madoff couldn't possibly be honestly making
the profits he claimed. The SEC did nothing and look what
happened!!!! Boards like this one help VPers because its a good way to
share our individual experiences.

#16

I'm no where near enough experienced with VP but I know enough about
advantage play with BJ. One of the things you learn quickly is that small
(miniscule) errors become big problems over time. One mistake and a tip to
the waitress and you can kiss your EV for that hour goodbye.
With the amount of coin you're cycling through, have you looked at all
aspects of your game and pay tables? Even an tiny error would affect your
bottom line with your signifcant coin-in. You also mention a partner. BJ
teams peer review each other on a constant basis too.

···

On Thu, Jan 8, 2009 at 3:34 PM, brumar_lv <brumar_lv@yahoo.com> wrote:

  --- In vpFREE@yahoogroups.com <vpFREE%40yahoogroups.com>,
"nightoftheiguana2000"
<nightoftheiguana2000@...> wrote:
> Regarding your concern, I would not assume that a game is fair with
> 100% probability. At some point you should consider giving up, if your
> results continue to be unfavorable. You could document and submit your
> results to gaming and they might or might not take action. I wouldn't
> put a lot of hope on gaming, but a complaint may start an
> investigation that may uncover something, that's how American Coin was
> brought to light.
>>>>>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>>>>>
I read where a financial analyst wrote a letter years ago to the SEC
stating categorically that Madoff couldn't possibly be honestly making
the profits he claimed. The SEC did nothing and look what
happened!!!! Boards like this one help VPers because its a good way to
share our individual experiences.

--
Thanks,

Dennis
______________________________
Read about my weight-loss journey:
http://www.healthstewards.com/member_detail.php?id=76&selected=1

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

#17

No matter how experienced players are there will be times when the

length

of losing streaks will make them start talking to themselves. I've

been

there!! We had the worse 8-month losing spell we EVER suffered in

23 year

during the middle of 2008. However, we ended up having one of our

better

years when we closed the books on December 31st.

I dunno, Jean. Flopping $100K at the Palms last December surely
balanced out any bad luck you could have had. And that was on hand #1, no?

I had one of my best years ever, too, but it was about as wacky as it
could get. I think I could put my year into this one day, $2 triple
play, $180K coin in, $23K in W2-Gs including a royal, win $3K on the
game. I lost 13% to the drop on the 2d half of the play. Let's just
say it was a wild card game with moderate variance/covariance. FOUR
dealt royals (unfortunately, only one on a multi-line, but that one
was for $50K). Way too many days where I lost 8% playing single line
Jacks. Two days where I won over $5K playing $5 Jacks, with no royal.
You look at my results, you'd think I was playing Triple Double.

2004 was an utter disaster for me, I remember one royal in nine
cycles, and that was flopping a royal for $1175 at Fiesta on the 101%
Joker; I almost broke the monitor on that one, and no tax form for
seven months-I still made $40K on the year.

What I've learned from this is you can't fight the Universe. You'll
always lose. But, yeah, it's hard not to take a losing streak
personally. At least 2009 has gotten off to a good start so far. But
I'm not seeing where the EV is coming from this year. Many casinos
have dramatically cut back what they spend on promotions. More and
more people know how to play. But going back to that Universe thing,
out of nowhere, the situations are finding me, and I received one very
large, $5K, break already. My advice, if you're not going to break the
glass or the montior, is to think positively lolol.

···

--- In vpFREE@yahoogroups.com, "Jean Scott" <queenofcomps@...> wrote:

#18

Sounds to me like they were just equating lower video poker paytables
to higher hold rates.

Game manufacturers usually provide a field return for each game -
about 2% lower than the expected return playing correct strategy. For
example, the field return for 9/6 JoB would be 97.54% instead of
99.54%. So the casino's expected hold would be 2.46% (100% - 97.54%).

"Hit rates of paying hands" usually refers to slot machines not video
poker.

···

--- In vpFREE@yahoogroups.com, Kathy <kc42223@...> wrote:

Ok....we are convinced the Casino will not purposely "fix" machines
and thus risk their license. We probably all agree with that. We
recently played at casino X and experinced a poor hit rate after
approximately 5000 hands on various machines. We had had other
similar experiences at the same casino. I wrote an e-mail to the
casino manager and told them our last trip was in fact our last
trip because I did not have confidence in their VP machines. He
called me within 1 hour of the e-mail. He stated, (paraphrased) [
well I will admit that some games in some casinos are set at
a higher hold rate than others, while still other games have
a lower hold rate. So I guess it just depends on which machines
you happen to playing on. ] By hold rate he meant the amount of
percentage the casino makes off of each machine. He went on to say
(paraphrased) [ the state allows us to set the hold rate between a
certain percentage range.
That percentage may vary at anytime as long as it stays within the
range] After the phone call I sent an e-mail to my casino host in
response to his asking us to reconsider. I replied with a lengthy
response about the hold rate being way to high on the VP machines
and until it changes we would not be back. I asked him to provide
me the record of percentage of payback overall for the casino and
his response taken directly from his e-mail to me is as follows:

"Thank you again we do not report the percentages of hold but are
held to a certain percentage range by law. I don't know the range
off hand but I will check with Operations tomorrow to get the
range. You can probably look it up on line some where?"

Question: does any of this mean anything or were they just talking
about paytables and that lower percentage paytables equal higher
hold rates or does it mean the hit rates of paying hands are
lowered while the paytable stays the same?

#19

<<I dunno, Jean. Flopping $100K at the Palms last December surely
balanced out any bad luck you could have had.>>

That was in 2007 - book closed on that year. 8-month losing spell was in 2008. What have you done for me lately, Mr. Casino Owner? :slight_smile: (If we go back far enough - 23 years - we are about 1 mil ahead and so probably should never complain about losing spells that are "only" 8 months long!)

And, yes, Brad got that Hundred Play dealt royal on his first hand of the session

···

________________
Jean $�ott, Frugal Gambler
http://queenofcomps.com/
You can read my blog at
http://lasvegasadvisor.com/blogs/jscott/

#20

Hit rates (cycle times) refers to video poker also, for example for a
typical fair deal machine, the royal flush hit rate is about once
every 45,000 hands on average, depending on strategy. In the case of
the American Coin machines, it was impossible to hit a full coin
royal, which was a mistake and probably why they were caught as
eventually people figured that out. It's much harder to figure out if
the machines simply have extended hit rates, so that they still hit,
just much less frequently than expected. Statistically, it's harder to
detect an altered hit rate than no hit rate at all. If you have a
statistics background, ask yourself the question: if this machine
actually has a doubled top jackpot hit rate, how would I figure that
out? Programming wise, it's relatively easy to detect when the shuffle
will produce a royal and to alter half of those so they do not, or 2/3
etc. How would you detect that statistically. It can be done, it's
just not easy, and it costs money and time. And good luck trying to
convince gaming of your findings.

···

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

"Hit rates of paying hands" usually refers to slot machines not video
poker.