J. J. Lodder wrote:

> Yes, I would want to see the other side of that coin.
> It is also called the Monte Carlo fallacy, based on an historical event.
> Some evening, more than a hundred years ago, Black came up 27 successive
> times. (or some such number) It was a genuine certified roulette wheel.
> Fortumes were lost that night by people massively betting on Red,
> and going on betting on Red to ever larger amounts.
> It seems to be a true and well-documented story, but I haven't verified,

I have a book written by an American, but it's translated to Danish. The
title is (translated back) "Mrs. Fortuna". It mentions this story, and
if I remember correctly it says that one of the Marx Brothers went away
with a nice win.

The book is not easily accessible, so I can't check it.

I seem to remember that the number was 28, but don't hang me if it's
wrong.

--
Bertel, Denmark

On Friday, December 8, 2023 at 12:20:02 PM UTC-7, Bertel Lund Hansen wrote:
> J. J. Lodder wrote:
....

> > It is also called the Monte Carlo fallacy, based on an historical event..
> > Some evening, more than a hundred years ago, Black came up 27 successive
> > times. (or some such number) It was a genuine certified roulette wheel.
> > Fortumes were lost that night by people massively betting on Red,
> > and going on betting on Red to ever larger amounts.
> > It seems to be a true and well-documented story, but I haven't verified,
>
> I have a book written by an American, but it's translated to Danish. The
> title is (translated back) "Mrs. Fortuna". It mentions this story, and
> if I remember correctly it says that one of the Marx Brothers went away
> with a nice win.
....

"Dame Fortune" is or was a common English name for that demigoddess.

--
Jerry Friedman

Bertel Lund Hansen <gadekryds@lundhansen.dk> wrote:

> J. J. Lodder wrote:
>
> > Yes, I would want to see the other side of that coin.
> > It is also called the Monte Carlo fallacy, based on an historical event.
> > Some evening, more than a hundred years ago, Black came up 27 successive
> > times. (or some such number) It was a genuine certified roulette wheel.
> > Fortumes were lost that night by people massively betting on Red,
> > and going on betting on Red to ever larger amounts.
> > It seems to be a true and well-documented story, but I haven't verified,
>
> I have a book written by an American, but it's translated to Danish. The
> title is (translated back) "Mrs. Fortuna". It mentions this story, and
> if I remember correctly it says that one of the Marx Brothers went away
> with a nice win.
>
> The book is not easily accessible, so I can't check it.
>
> I seem to remember that the number was 28, but don't hang me if it's
> wrong.

Aw, now you made me look it up. No problem, wikipedia say so.
<https://en.wikipedia.org/wiki/Gambler%27s_fallacy#Monte_Carlo_Casino>

I must be clairvoyant:
Wikip says 28, but they refer to a source that says 26,

Jan

Jerry Friedman <jerry.friedman99@gmail.com> wrote:

> On Friday, December 8, 2023 at 12:20:02?PM UTC-7, Bertel Lund Hansen wrote:
> > J. J. Lodder wrote:
> ...
>
> > > It is also called the Monte Carlo fallacy, based on an historical event.
> > > Some evening, more than a hundred years ago, Black came up 27 successive
> > > times. (or some such number) It was a genuine certified roulette wheel.
> > > Fortumes were lost that night by people massively betting on Red,
> > > and going on betting on Red to ever larger amounts.
> > > It seems to be a true and well-documented story, but I haven't verified,
> >
> > I have a book written by an American, but it's translated to Danish. The
> > title is (translated back) "Mrs. Fortuna". It mentions this story, and
> > if I remember correctly it says that one of the Marx Brothers went away
> > with a nice win.
> ...
>
> "Dame Fortune" is or was a common English name for that demigoddess.

Not very English, to begin with.
She, and the Wheel of Fortune she turns, were imported on Tarot cards.
(from Italy, where the Roman goddess hadn't been forgotten)

Jan

On Friday, December 8, 2023 at 1:47:20 PM UTC-7, J. J. Lodder wrote:
> Jerry Friedman <jerry.fr...@gmail.com> wrote:
> > On Friday, December 8, 2023 at 12:20:02?PM UTC-7, Bertel Lund Hansen wrote:
> > > J. J. Lodder wrote:
> > ...
> >
> > > > It is also called the Monte Carlo fallacy, based on an historical event.
> > > > Some evening, more than a hundred years ago, Black came up 27 successive
> > > > times. (or some such number) It was a genuine certified roulette wheel.
> > > > Fortumes were lost that night by people massively betting on Red,
> > > > and going on betting on Red to ever larger amounts.
> > > > It seems to be a true and well-documented story, but I haven't verified,
> > >
> > > I have a book written by an American, but it's translated to Danish. The
> > > title is (translated back) "Mrs. Fortuna". It mentions this story, and
> > > if I remember correctly it says that one of the Marx Brothers went away
> > > with a nice win.
> > ...
> >
> > "Dame Fortune" is or was a common English name for that demigoddess.
>
> Not very English, to begin with.

As the etymology of "Dame Fortune" suggests. If you call her Lady Luck,
there is room for doubt.

https://www.youtube.com/watch?v=MfiKk4wxiVM

> She, and the Wheel of Fortune she turns, were imported on Tarot cards.
> (from Italy, where the Roman goddess hadn't been forgotten)

As she would have it, one of my students yesterday was wearing a T-shirt
depicting that card.

--
Jerry Friedman

On 2023-12-08 17:42:36 +0000, Athel Cornish-Bowden said:

> On 2023-12-08 17:17:01 +0000, J. J. Lodder said:
>
>> Athel Cornish-Bowden <me@yahoo.com> wrote:
>>
>>> On 2023-12-07 19:17:52 +0000, lar3ryca said:
>>>
>>>> On 2023-12-07 12:24, Rich Ulrich wrote:
>>>>> On Wed, 6 Dec 2023 09:10:11 +1100, Peter Moylan
>>>>> <peter@pmoylan.org.invalid> wrote:
>>>>>
>>>>>> On 05/12/23 23:20, Bertel Lund Hansen wrote:
>>>>>>> J. J. Lodder wrote:
>>>>>>>
>>>>>>>>> Interesting. I believe probability theory was invented because
>>>>>>>>> people had an economic need to understand something about gambling,
>>>>>>>>> though I doubt it helped.
>>>>>>>>
>>>>>>>> Certainly, by Pascal, among others.
>>>>>>>> Some people were addicted to crazy bets,
>>>>>>>> like: bet you I can throw 50 point with one throw of 10 dice.
>>>>>>>> Pascal showed how to calculate systematically
>>>>>>>> how much such a bet is worth.
>>>>>>>
>>>>>>> I think that thinking about probability is much older than that. Cesar
>>>>>>> talked about throwing dice when he crossed the Rubicon, and leaders of
>>>>>>> an army must have considered the chance of success before starting some
>>>>>>> planned action.
>>>>>>
>>>>>> The concept of "chance of success" has been around for a long time, but
>>>>>> the estimates were based on guesswork. (And throwing dice amounts to an
>>>>>> admission that "I don't have a clue what my chances are".) What changed
>>>>>> when Cardano and others came on the scene was the creation of a theory
>>>>>> with a solid mathematical footing. Modern probability theory allows
>>>>>> well-grounded calculations to be done.
>>>>>>
>>>>>> The difference is important because some of the results turn out to be
>>>>>> counterintuitive. That is, estimates of probability based on gut
>>>>>> feelings rather than mathematics can be wildly wrong.
>>>>>
>>>>> The most misleading of the 'gut feelings' may be the illusion
>>>>> that when two alternatives are on hand, their chances are equal.
>>>>>
>>>>> Drawing to an inside straight? - 'amateur mistake' in poker, which
>>>>> is to say, it happens a lot. Experience is a teacher. But even if you
>>>>> are playing games of chance, there are combinations that you will
>>>>> seldom see.
>>>>>
>>>>> Outside of games of chance, there re LOTs of occasions where
>>>>> 'chances are equal' is a tougher illusion to dispell. Is this
>>>>> politician lying? -Well, he MIGHT be telling the truth.
>>>>
>>>> A common gambler's mistake is thinking that just because something with
>>>> an equal chance (flipping a coin, for example) has come up one way 10
>>>> times, that the chances are greater than 50-50 for the other way to
>>>> come up next.
>>>
>>> If anything it's the other way round, because 10/10 suggests a biassed coin.
>>
>> Yes, I would want to see the other side of that coin.
>> It is also called the Monte Carlo fallacy, based on an historical event.
>> Some evening, more than a hundred years ago, Black came up 27 successive
>> times. (or some such number) It was a genuine certified roulette wheel.
>> Fortumes were lost that night by people massively betting on Red,
>> and going on betting on Red to ever larger amounts.
>> It seems to be a true and well-documented story, but I haven't verified,
>
> 27 is an interesting number in this context. In the 1st volume of the
> Advanced Thery of Statistics (a book I no longer have, so I can't
> check), Kendall & Stuart warn that one cannot assume that a large
> enough sample will always converge on the population mean, because
> there may be biasses that make over- or under- estimation more likely.
> The example they give is of farmers' predictions of crop yields in good
> or bad years for agriculture. They found that in 26 out of 27 years of
> predictions the UK up to the 1930s, the actual yields proved to be
> higher than the predicted yields. They explained this result in terms
> of habitual pessimism on the part of farmers.

It turns out that a PDF file of the book is available, with Table 8.4
on p. 190 at

https://archive.org/details/in.ernet.dli.2015.57860/page/n204/mode/1up?view=theater

My memory was not 100% correct in all details, but near enough:
certainly the major concluson was correct.

At the very beginning of the book, before the Preface, there is a
quotation from The Handbook of Hymen, by O. Henry, that has nothing to
do with sampling bias, but which has always amused me, ending as
follows:

"Go on, Mr. Pratt", says Mrs. Sampson, "Them ideas is so original and
soothing. I think statistics are just as lovely as they can be."

I have long agreed with Mrs. Sampson, but I think we're probably in a
tiny minority of humanity.

--
Athel -- French and British, living in Marseilles for 36 years; mainly
in England until 1987.

On Fri, 8 Dec 2023 21:11:48 +0100, nospam@de-ster.demon.nl (J. J.
Lodder) wrote:

>Bertel Lund Hansen <gadekryds@lundhansen.dk> wrote:
>
>> J. J. Lodder wrote:
>>
>> > Yes, I would want to see the other side of that coin.
>> > It is also called the Monte Carlo fallacy, based on an historical event.
>> > Some evening, more than a hundred years ago, Black came up 27 successive
>> > times. (or some such number) It was a genuine certified roulette wheel.
>> > Fortumes were lost that night by people massively betting on Red,
>> > and going on betting on Red to ever larger amounts.
>> > It seems to be a true and well-documented story, but I haven't verified,
>>
>> I have a book written by an American, but it's translated to Danish. The
>> title is (translated back) "Mrs. Fortuna". It mentions this story, and
>> if I remember correctly it says that one of the Marx Brothers went away
>> with a nice win.
>>
>> The book is not easily accessible, so I can't check it.
>>
>> I seem to remember that the number was 28, but don't hang me if it's
>> wrong.
>
>Aw, now you made me look it up. No problem, wikipedia say so.
><https://en.wikipedia.org/wiki/Gambler%27s_fallacy#Monte_Carlo_Casino>
>
>I must be clairvoyant:
>Wikip says 28, but they refer to a source that says 26,
>

I clicked on the link. Same article I quoted from.

Check your vision? My look at Wikip says 26 today, the same
as it did when I posted '26' on the 7th. The edit history shows
no changes in December.

--
Rich Ulrich

On 2023-12-09 03:04, Athel Cornish-Bowden wrote:
> On 2023-12-08 17:42:36 +0000, Athel Cornish-Bowden said:
>
>> On 2023-12-08 17:17:01 +0000, J. J. Lodder said:
>>
>>> Athel Cornish-Bowden <me@yahoo.com> wrote:
>>>
>>>> On 2023-12-07 19:17:52 +0000, lar3ryca said:
>>>>
>>>>> On 2023-12-07 12:24, Rich Ulrich wrote:
>>>>>> On Wed, 6 Dec 2023 09:10:11 +1100, Peter Moylan
>>>>>> <peter@pmoylan.org.invalid> wrote:
>>>>>>
>>>>>>> On 05/12/23 23:20, Bertel Lund Hansen wrote:
>>>>>>>> J. J. Lodder wrote:
>>>>>>>>
>>>>>>>>>> Interesting.  I believe probability theory was invented because
>>>>>>>>>> people had an economic need to understand something about
>>>>>>>>>> gambling,
>>>>>>>>>> though I doubt it helped.
>>>>>>>>>
>>>>>>>>> Certainly, by Pascal, among others.
>>>>>>>>> Some people were addicted to crazy bets,
>>>>>>>>> like: bet you I can throw 50 point with one throw of 10 dice.
>>>>>>>>> Pascal showed how to calculate systematically
>>>>>>>>> how much such a bet is worth.
>>>>>>>>
>>>>>>>> I think that thinking about probability is much older than that.
>>>>>>>> Cesar
>>>>>>>> talked about throwing dice when he crossed the Rubicon, and
>>>>>>>> leaders of
>>>>>>>> an army must have considered the chance of success before
>>>>>>>> starting some
>>>>>>>> planned action.
>>>>>>>
>>>>>>> The concept of "chance of success" has been around for a long
>>>>>>> time, but
>>>>>>> the estimates were based on guesswork. (And throwing dice amounts
>>>>>>> to an
>>>>>>> admission that "I don't have a clue what my chances are".) What
>>>>>>> changed
>>>>>>> when Cardano and others came on the scene was the creation of a
>>>>>>> theory
>>>>>>> with a solid mathematical footing. Modern probability theory allows
>>>>>>> well-grounded calculations to be done.
>>>>>>>
>>>>>>> The difference is important because some of the results turn out
>>>>>>> to be
>>>>>>> counterintuitive. That is, estimates of probability based on gut
>>>>>>> feelings rather than mathematics can be wildly wrong.
>>>>>>
>>>>>> The most misleading of the 'gut feelings' may be the illusion
>>>>>> that when two alternatives are on hand, their chances are equal.
>>>>>>
>>>>>> Drawing to an inside straight? - 'amateur mistake' in poker, which
>>>>>> is to say, it happens a lot. Experience is a teacher.  But even if
>>>>>> you
>>>>>> are playing games of chance, there are combinations that you will
>>>>>> seldom see.
>>>>>>
>>>>>> Outside of games of chance, there re LOTs of occasions where
>>>>>> 'chances are equal' is a tougher illusion to dispell. Is this
>>>>>> politician lying? -Well, he MIGHT be telling the truth.
>>>>>
>>>>> A common gambler's mistake is thinking that just because something
>>>>> with
>>>>> an equal chance (flipping a coin, for example) has come up one way 10
>>>>> times, that the chances are greater than 50-50 for the other way to
>>>>> come up next.
>>>>
>>>> If anything it's the other way round, because 10/10 suggests a
>>>> biassed coin.
>>>
>>> Yes, I would want to see the other side of that coin.
>>> It is also called the Monte Carlo fallacy, based on an historical event.
>>> Some evening, more than a hundred years ago, Black came up 27 successive
>>> times. (or some such number) It was a genuine certified roulette wheel.
>>> Fortumes were lost that night by people massively betting on Red,
>>> and going on betting on Red to ever larger amounts.
>>> It seems to be a true and well-documented story, but I haven't verified,
>>
>> 27 is an interesting number in this context. In the 1st volume of the
>> Advanced Thery of Statistics (a book I no longer have, so I can't
>> check), Kendall & Stuart warn that one cannot assume that a large
>> enough sample will always converge on the population mean, because
>> there may be biasses that make over- or under- estimation more likely.
>> The example they give is of farmers' predictions of crop yields in
>> good or bad years for agriculture. They found that in 26 out of 27
>> years of predictions the UK up to the 1930s, the actual yields proved
>> to be higher than the predicted yields. They explained this result in
>> terms of habitual pessimism on the part of farmers.
>
> It turns out that a PDF file of the book is available, with Table 8.4 on
> p. 190 at
>
> https://archive.org/details/in.ernet.dli.2015.57860/page/n204/mode/1up?view=theater
>
> My memory was not 100% correct in all details, but near enough:
> certainly the major concluson was correct.
>
> At the very beginning of the book, before the Preface, there is a
> quotation from The Handbook of Hymen, by O. Henry, that has nothing to
> do with sampling bias, but which has always amused me, ending as follows:

Did it have anything to do with virginity?

> "Go on, Mr. Pratt", says Mrs. Sampson, "Them ideas is so original and
> soothing. I think statistics are just as lovely as they can be."
>
> I have long agreed with Mrs. Sampson, but I think we're probably in a
> tiny minority of humanity.
>
>

--
To the person who invented Zero: Thanks for nothing.

On 2023-12-09 19:19:00 +0000, lar3ryca said:

> On 2023-12-09 03:04, Athel Cornish-Bowden wrote:
>> On 2023-12-08 17:42:36 +0000, Athel Cornish-Bowden said:
>>
>>> On 2023-12-08 17:17:01 +0000, J. J. Lodder said:
>>>
>>>> Athel Cornish-Bowden <me@yahoo.com> wrote:
>>>>
>>>>> On 2023-12-07 19:17:52 +0000, lar3ryca said:
>>>>>
>>>>>> On 2023-12-07 12:24, Rich Ulrich wrote:
>>>>>>> On Wed, 6 Dec 2023 09:10:11 +1100, Peter Moylan
>>>>>>> <peter@pmoylan.org.invalid> wrote:
>>>>>>>
>>>>>>>> On 05/12/23 23:20, Bertel Lund Hansen wrote:
>>>>>>>>> J. J. Lodder wrote:
>>>>>>>>>
>>>>>>>>>>> Interesting.  I believe probability theory was invented because
>>>>>>>>>>> people had an economic need to understand something about gambling,
>>>>>>>>>>> though I doubt it helped.
>>>>>>>>>>
>>>>>>>>>> Certainly, by Pascal, among others.
>>>>>>>>>> Some people were addicted to crazy bets,
>>>>>>>>>> like: bet you I can throw 50 point with one throw of 10 dice.
>>>>>>>>>> Pascal showed how to calculate systematically
>>>>>>>>>> how much such a bet is worth.
>>>>>>>>>
>>>>>>>>> I think that thinking about probability is much older than that. Cesar
>>>>>>>>> talked about throwing dice when he crossed the Rubicon, and leaders of
>>>>>>>>> an army must have considered the chance of success before starting some
>>>>>>>>> planned action.
>>>>>>>>
>>>>>>>> The concept of "chance of success" has been around for a long time, but
>>>>>>>> the estimates were based on guesswork. (And throwing dice amounts to an
>>>>>>>> admission that "I don't have a clue what my chances are".) What changed
>>>>>>>> when Cardano and others came on the scene was the creation of a theory
>>>>>>>> with a solid mathematical footing. Modern probability theory allows
>>>>>>>> well-grounded calculations to be done.
>>>>>>>>
>>>>>>>> The difference is important because some of the results turn out to be
>>>>>>>> counterintuitive. That is, estimates of probability based on gut
>>>>>>>> feelings rather than mathematics can be wildly wrong.
>>>>>>>
>>>>>>> The most misleading of the 'gut feelings' may be the illusion
>>>>>>> that when two alternatives are on hand, their chances are equal.
>>>>>>>
>>>>>>> Drawing to an inside straight? - 'amateur mistake' in poker, which
>>>>>>> is to say, it happens a lot. Experience is a teacher.  But even if you
>>>>>>> are playing games of chance, there are combinations that you will
>>>>>>> seldom see.
>>>>>>>
>>>>>>> Outside of games of chance, there re LOTs of occasions where
>>>>>>> 'chances are equal' is a tougher illusion to dispell. Is this
>>>>>>> politician lying? -Well, he MIGHT be telling the truth.
>>>>>>
>>>>>> A common gambler's mistake is thinking that just because something with
>>>>>> an equal chance (flipping a coin, for example) has come up one way 10
>>>>>> times, that the chances are greater than 50-50 for the other way to
>>>>>> come up next.
>>>>>
>>>>> If anything it's the other way round, because 10/10 suggests a biassed coin.
>>>>
>>>> Yes, I would want to see the other side of that coin.
>>>> It is also called the Monte Carlo fallacy, based on an historical event.
>>>> Some evening, more than a hundred years ago, Black came up 27 successive
>>>> times. (or some such number) It was a genuine certified roulette wheel.
>>>> Fortumes were lost that night by people massively betting on Red,
>>>> and going on betting on Red to ever larger amounts.
>>>> It seems to be a true and well-documented story, but I haven't verified,
>>>
>>> 27 is an interesting number in this context. In the 1st volume of the
>>> Advanced Thery of Statistics (a book I no longer have, so I can't
>>> check), Kendall & Stuart warn that one cannot assume that a large
>>> enough sample will always converge on the population mean, because
>>> there may be biasses that make over- or under- estimation more likely.
>>> The example they give is of farmers' predictions of crop yields in good
>>> or bad years for agriculture. They found that in 26 out of 27 years of
>>> predictions the UK up to the 1930s, the actual yields proved to be
>>> higher than the predicted yields. They explained this result in terms
>>> of habitual pessimism on the part of farmers.
>>
>> It turns out that a PDF file of the book is available, with Table 8.4
>> on p. 190 at
>>
>> https://archive.org/details/in.ernet.dli.2015.57860/page/n204/mode/1up?view=theater
>>
>>
>> My memory was not 100% correct in all details, but near enough:
>> certainly the major concluson was correct.
>>
>> At the very beginning of the book, before the Preface, there is a
>> quotation from The Handbook of Hymen, by O. Henry, that has nothing to
>> do with sampling bias, but which has always amused me, ending as
>> follows:
>
> Did it have anything to do with virginity?

Apparently not, but I haven't read it.
>
>> "Go on, Mr. Pratt", says Mrs. Sampson, "Them ideas is so original and
>> soothing. I think statistics are just as lovely as they can be."
>>
>> I have long agreed with Mrs. Sampson, but I think we're probably in a
>> tiny minority of humanity.

--
Athel -- French and British, living in Marseilles for 36 years; mainly
in England until 1987.