Method to beat the lottery
Andrew P. Bajorinas
I see many people in this group looking for a "system" to win the
lottery. Are they serious? This is not a flame, but a serious question.
Since the drawing is a random drawing what system could exist? Belief that
some system exists seems illogical to me. So I guess my question is:
What reasons are there to believe any system exists?
------------------------------------------------------------------------
| Andrew P. Bajorinas Bajo...@Perkin-Elmer.com |
| -----------------------------------------------------------------------|
| The above opinions are my own. | I will never let my cat |
| My employer thinks I am working. 8^) | use my net access again! |
------------------------------------------------------------------------
Terry Thero
: I see many people in this group looking for a "system" to win the
: Since the drawing is a random drawing what system could exist? Belief that
: some system exists seems illogical to me. So I guess my question is:
: What reasons are there to believe any system exists?
Pretty sound logic to me Andrew :-) If a system did exist, rather than
advertise or sell it, you would hear about the person going around to win
all the various state and power ball lotteries. As near as I can tell, all
the lotteries are PURELY random.
You bets your bucks, you takes your chances. Take care, and good luck :-)
CU
twthero
_____________________________________________
| Terry W. Thero: e-mail: |
| te...@col.hp.com |
|____________________________________________|
| The Only Easy Day Was Yesterday |
|____________________________________________|
Thomas Doehne
> I see many people in this group looking for a "system" to win the
>lottery. Are they serious? This is not a flame, but a serious question.
>Since the drawing is a random drawing what system could exist? Belief that
>some system exists seems illogical to me. So I guess my question is:
>
> What reasons are there to believe any system exists?
I believe I have one. First, I only buy tickets when the expected
value of a ticket gives a return above 5%. (To calculate
expected value, take the after-tax cash flow from winning,
discount the cash flow by an appropriate interest rate -- 5%
in this case -- and multiply by the probability that you
will win that cash flow from a single ticket. If the expected
value is positive, the return is above 5 %. Make sure you
add in the expected value of the lesser prizes, as they
make a definite difference in the value of a ticket.)
Second, I use a computer program to pick number sets. The
program I use filters out commonly chosen number sets, so
that I probably won't split the pot.
I end up not playing the lottery very often -- usually the
pots are too low. Generally, the pot has to be above about
17 million for a positive expected value (for the Michigan
lotto).
--
Tom Doehne
doe...@cse.ogi.edu
Ken Tetterington
This is not true at all. People beat the lottery every day. It may not
come down to working out a method to beat a random ticket, but rather it
comes down to a matter of using intuition and probabilities to choose
numbers.
Don't tell me about odds, as it's like the national budget. You can jimmy
the odds this way and that. What are the odds of 1 coin being trown face
up? What are the odds of throwing face up 1200 times in a row. Each throw
has the same odds, but the cumulative odds are not the same. I do not
want to hear about odds.
Tell me what the odds are of a person choosing their same numbers for 1
1/2 years and then failing to buy that ticket. Low and behold the day he
forgets to play his tickets, he gets a match. IT'S NOT THE ODDS THAT
DETERMINE THE NEXT WINNER, IT'S SOMETHING FAR BIGGER THAN THAT.
The thing is that if you trust the odds, you will likely not win for the
next 13,983,816 draws. Don't try to convince me with odds, as they do not
apply beyond the class room..
Ken..
John Griffin
I see you're still in search of your very first clue. I will answer your
question about the 1200 coin flips. If heads has come up 1199 times in a
row using a fair coin, the probability of 1200 in a row is 1/2. You have
heard that before, right? I won't bother with your 'odds' nonsense; it's
far too complex for you, as was proven in earlier discussions.
Andrew P. Bajorinas
doe...@church.cse.ogi.edu (Thomas Doehne) wrote:
]
]
]I believe I have one. First, I only buy tickets when the expected
]value of a ticket gives a return above 5%. (To calculate
]expected value, take the after-tax cash flow from winning,
]discount the cash flow by an appropriate interest rate -- 5%
]in this case -- and multiply by the probability that you
]will win that cash flow from a single ticket. If the expected
]value is positive, the return is above 5 %. Make sure you
]add in the expected value of the lesser prizes, as they
]make a definite difference in the value of a ticket.)
]Second, I use a computer program to pick number sets. The
]program I use filters out commonly chosen number sets, so
]that I probably won't split the pot.
]
]I end up not playing the lottery very often -- usually the
]pots are too low. Generally, the pot has to be above about
]17 million for a positive expected value (for the Michigan
]lotto).
]
Wow, what an impressive (to me) post. Clearly you have a good
strategy for maximizing the benefit. It is clear you have done
your homework.
I guess, however, what I really was asking was more concerned
with people who have "systems" to get the winning numbers. People
who keep track of previous winners and other such things.
Thanks for the reply.
Andy Bajorinas
Normand Veilleux
>From: doe...@church.cse.ogi.edu (Thomas Doehne)
[A winning system]
>I believe I have one. First, I only buy tickets when the expected
>value of a ticket gives a return above 5%. (To calculate
>expected value, take the after-tax cash flow from winning,
>discount the cash flow by an appropriate interest rate -- 5%
>in this case -- and multiply by the probability that you
>will win that cash flow from a single ticket. If the expected
>value is positive, the return is above 5 %. Make sure you
>add in the expected value of the lesser prizes, as they
>make a definite difference in the value of a ticket.)
>Second, I use a computer program to pick number sets. The
>program I use filters out commonly chosen number sets, so
>that I probably won't split the pot.
>
>I end up not playing the lottery very often -- usually the
>pots are too low. Generally, the pot has to be above about
>17 million for a positive expected value (for the Michigan
>lotto).
Well, I guess that means that your system can't be used in
Canada! I believe the record high for the Canadian Lotto 6/49
is 14 or 15 million.
I really wonder why people are so paranoid about splitting the
jackpot with someone else. I mean, wouldn't you prefer splitting
a $3 million dollar jackpot 3 ways rather than not winning at
all? You wouldn't have to twist my arm! I'd take the split
any day! :-)
JAMES SARNELLI
(Andrew P. Bajorinas) writes:
>
>
>
> I see many people in this group looking for a "system" to win the
>lottery. Are they serious? This is not a flame, but a serious
question.
>Since the drawing is a random drawing what system could exist? Belief
that
>some system exists seems illogical to me. So I guess my question is:
>
> What reasons are there to believe any system exists?
>
>
>
>| Andrew P. Bajorinas
Bajo...@Perkin-Elmer.com |
>|
-----------------------------------------------------------------------
-----------------------------------------------------------------------
Andrew
Using a system refers to taking a logical approach to playing the
lottery/lotto games. Winning the lottery has metaphorically been
compared to getting struck by lightening. Although getting struck by
lightening is improbable there are numerous actions that you can take
to increase your probability of participating in such an event. The
same is true of the lottery. In the next few days I will begin posting
some logical methods of lottery/lotto play. Try out the methods and
confirm for yourself that playing systematically increases your
probability of matching numbers drawn.
MORE TO FOLLOW
JAM...@AOL.COM
fycnrdths,'mntnthrck,whrshldb.
Tetterington) wrote:
|This is not true at all. People beat the lottery every day. It may not
|come down to working out a method to beat a random ticket, but rather it
|comes down to a matter of using intuition and probabilities to choose
|numbers.
|
|Don't tell me about odds, as it's like the national budget. You can jimmy
|the odds this way and that. What are the odds of 1 coin being trown face
|up? What are the odds of throwing face up 1200 times in a row. Each throw
|has the same odds, but the cumulative odds are not the same. I do not
|want to hear about odds.
|
|Tell me what the odds are of a person choosing their same numbers for 1
|1/2 years and then failing to buy that ticket. Low and behold the day he
|forgets to play his tickets, he gets a match. IT'S NOT THE ODDS THAT
|DETERMINE THE NEXT WINNER, IT'S SOMETHING FAR BIGGER THAN THAT.
|
|The thing is that if you trust the odds, you will likely not win for the
|next 13,983,816 draws. Don't try to convince me with odds, as they do not
|apply beyond the class room..
|
|Ken..
Ken,
This moronic post just goes to show that you either do not know enough, or
refuse to learn enough (the latter is the most probable) to even discuss
probability intelligently, nevermind trying to disprove Terry's statements
or even create a winning system.
I really suggest that you and James go out and learn something before you
spout off like an authority. Although I doubt you will, because if you
did learn some basic probability/statistics, you would be forced to drop
your lotto fantasies.
The truly ironic fact is that you need probability/statistics to prove
that your systems work. Without them, there is no way that you can prove
that your system(s) do any better than random choices.
ray
----------
Ray DeGennaro
dege...@bmsrs.usc.edu
----------
Did you know that 'gullible' is not in Webster's Dictionary?
John Griffin
: In <3r9tej$h00...@bajoriap.perkin-elmer.com> Bajo...@Perkin-Elmer.com
: (Andrew P. Bajorinas) writes:
: >
: >
: >
: > I see many people in this group looking for a "system" to win the
: >lottery. Are they serious? This is not a flame, but a serious
: question.
: >Since the drawing is a random drawing what system could exist? Belief
: that
: >some system exists seems illogical to me. So I guess my question is:
: >
: > What reasons are there to believe any system exists?
: -----------------------------------------------------------------------
: Andrew
: Using a system refers to taking a logical approach to playing the
: lottery/lotto games. Winning the lottery has metaphorically been
: compared to getting struck by lightening. Although getting struck by
: lightening is improbable there are numerous actions that you can take
: to increase your probability of participating in such an event. The
: same is true of the lottery. In the next few days I will begin posting
: some logical methods of lottery/lotto play. Try out the methods and
: confirm for yourself that playing systematically increases your
: probability of matching numbers drawn.
: MORE TO FOLLOW
: JAM...@AOL.COM
Andrew
You will "laugh yer ass off" when Sarnelli starts posting his nonsense.
Don't take him seriously - he has been very, very ill. His last sentence
above shows that he hasn't recovered yet.
Ron Meisenheimer
>
>In <3r9tej$h00...@bajoriap.perkin-elmer.com> Bajo...@Perkin-Elmer.com
>(Andrew P. Bajorinas) writes:
>>
>> I see many people in this group looking for a "system" to win the
>>lottery. Are they serious? This is not a flame, but a serious
>question.
>>Since the drawing is a random drawing what system could exist? Belief
>that
>>some system exists seems illogical to me. So I guess my question is:
>>
>> What reasons are there to believe any system exists?
>Andrew
>
> Using a system refers to taking a logical approach to playing the
>lottery/lotto games. Winning the lottery has metaphorically been
>compared to getting struck by lightening. Although getting struck by
>lightening is improbable there are numerous actions that you can take
>to increase your probability of participating in such an event. The
>same is true of the lottery.
Only by buying more tickets.
>In the next few days I will begin posting some logical methods of
>lottery/lotto play. Try out the methods and
>confirm for yourself that playing systematically increases your
>probability of matching numbers drawn.
Not true, as has been pointed out to you many, many times. If you
buy n different tickets using *any* system, and I buy n different
tickets using no system whatever, your chance of winning is
identically equal to my own, a trivial result in elementary
probability.
Why don't you turn over a new leaf, James, with the start of this
new newsgroup? I recommend beginning with an earnest study of algebra,
followed by elementary probability and statistics. You really have a
golden opportunity here not to embarrass yourself before a whole
new audience.
Regards,
Ron Meisenheimer (ro...@ns.net)
fycnrdths,'mntnthrck,whrshldb.
| Using a system refers to taking a logical approach to playing the
|lottery/lotto games. Winning the lottery has metaphorically been
|compared to getting struck by lightening. Although getting struck by
|lightening is improbable there are numerous actions that you can take
|to increase your probability of participating in such an event. The
|same is true of the lottery. In the next few days I will begin posting
|some logical methods of lottery/lotto play. Try out the methods and
|confirm for yourself that playing systematically increases your
|probability of matching numbers drawn.
|
|MORE TO FOLLOW
|JAM...@AOL.COM
Yes James, please do post something this time. In rec.gambling, you never
got beyond step two. Actually, you never got to step two, you were proven
wrong over, and over, and over again and resorted to name calling and
whining. Please try to stick to the issues this time. And if, other
people start throwing mud first (which I doubt), be a little more mature
than them and still stick to the topic. I promise you I will stick to the
topic until you start throwing stones, after that, I can't promise
anything.
David K. Bryant
>Winning the lottery has metaphorically been
>compared to getting struck by lightening. Although getting struck by
>lightening is improbable there are numerous actions that you can take
>to increase your probability of participating in such an event.
>
>The same is true of the lottery.
Surefire way of increasing your probability of getting "struck" by the lottery:
do while not winner
Buy a ticket.
Wait for the draw.
In the next few days I will begin posting
>some logical methods of lottery/lotto play. Try out the methods and
>confirm for yourself that playing systematically increases your
>probability of matching numbers drawn.
I can hardly wait.
Daniel Efron
SARNELLI) wrote:
> Using a system refers to taking a logical approach to playing the
> compared to getting struck by lightening. Although getting struck by
> lightening is improbable there are numerous actions that you can take
> to increase your probability of participating in such an event. The
> some logical methods of lottery/lotto play. Try out the methods and
> confirm for yourself that playing systematically increases your
> probability of matching numbers drawn.
>
> JAM...@AOL.COM
Being struck by lightning and winning the lottery are not metaphorically
comparable. First, the chance of being struck by lightning is much better
than the chance of winning most lotteries. Second, being struck by
lightning involves well proven physical interactions (thunderstorm, being
the highest object in an area, etc.).
There is absolutely nothing you can do to increase your chances of winning
a random lottery. No modern US lottery has ever been proven to be
usefully non-random except by illicit tampering.
In order to be able to even address the issue of increasing your chances
of winning the lottery, you have to proven that the lottery is usefully
non-random. Fortunately, there are some very basic statistical tests to
prove that a lottery is not behaving like random independent variables.
Unfortunately, no modern US lottery has ever been proven to be usefully
non-random except by illicit tampering (oops, I said that already).
--
Dan Efron
Stanley T.H. Chow
Normand Veilleux <nvei...@emr1.emr.ca> wrote:
>>I end up not playing the lottery very often -- usually the
>>pots are too low. Generally, the pot has to be above about
>>17 million for a positive expected value (for the Michigan
>>lotto).
>
>Well, I guess that means that your system can't be used in
>Canada! I believe the record high for the Canadian Lotto 6/49
>is 14 or 15 million.
Except that Canadian lotteries paid the whole thing in cash,
tax-free. An extreme case of a US lottery payout is like:
$1 million per year for 29 years
$29 million in the 30th year
Depending on the interest rate, this has "Net Present Value"
of anywhere from 10.5 Million (at 10% interest) to 20 Million
(at 5% interest).
Figure in the tax of roughly 50%, the US$50 million is worth
net, after-tax about 5 to 10 million. Add in current exchange
rate, this becomes C$7 to 14 million.
In other words, the Canadian 6/49 has better expected value.
(And is simply due to better payout ratio).
>I really wonder why people are so paranoid about splitting the
>jackpot with someone else. I mean, wouldn't you prefer splitting
>a $3 million dollar jackpot 3 ways rather than not winning at
>all? You wouldn't have to twist my arm! I'd take the split
>any day! :-)
I don't think its paranoia. People sensibly prefer to avoid
splitting, especially when it does not cost anything.
--
Stanley Chow; sc...@bnr.ca, stanley....@nt.com; (613) 763-2831
Bell Northern Research Ltd., PO Box 3511 Station C, Ottawa, Ontario
Me? Represent other people? Don't make them laugh so hard.
Kevin Martin Leuthold
>In article <3rij6l$7...@ixnews5.ix.netcom.com>, jam...@ix.netcom.com (JAMES SARNELLI) says:
>>
>>In <3r9tej$h00...@bajoriap.perkin-elmer.com> Bajo...@Perkin-Elmer.com
>>(Andrew P. Bajorinas) writes:
>>>
>>> I see many people in this group looking for a "system" to win the
>>>lottery. Are they serious? This is not a flame, but a serious
>>question.
>>>Since the drawing is a random drawing what system could exist? Belief
>>that
>>>some system exists seems illogical to me. So I guess my question is:
>>>
>>> What reasons are there to believe any system exists?
>>Andrew
>>
>> Using a system refers to taking a logical approach to playing the
>>lottery/lotto games. Winning the lottery has metaphorically been
>>compared to getting struck by lightening. Although getting struck by
>>lightening is improbable there are numerous actions that you can take
>>to increase your probability of participating in such an event. The
>>same is true of the lottery.
> Only by buying more tickets.
>>In the next few days I will begin posting some logical methods of
>>lottery/lotto play. Try out the methods and
>>confirm for yourself that playing systematically increases your
>>probability of matching numbers drawn.
> Not true, as has been pointed out to you many, many times. If you
> buy n different tickets using *any* system, and I buy n different
> tickets using no system whatever, your chance of winning is
> identically equal to my own, a trivial result in elementary
> probability.
I think it depends on what you mean by "winning". In some lotteries,
you win money if you match only some of the numbers. If you include
that in your definition of winning, then it seems that you can
benefit from being systematic.
Let's say there's a lottery where you pick 3 numbers out of the
numbers {1,2,3,4,5}. And if 2 or 3 out of 3 numbers are picked,
you win a prize. Now let's say i pick the following 2 tickets:
1 2 3
1 4 5
and you pick the following 2 tickets:
1 2 3
1 2 4
It's easy to check that no matter what combination comes up,
I will win a prize, but you will not win if the combination
3 4 5 comes up.
Now, I doubt that in a real lottery, you could actually guarantee
that you would win money this way, but it seems that you could
come up with a system that would be guaranteed to recoup more
of your money than another system via the smaller prizes, and
therefore be a better system. (by the way, I have just started
following this group. Is this what people mean when they refer
to "wheeling")
--
Kevin Leuthold
Thomas Doehne
>In article <3reb0m$m...@reuter.cse.ogi.edu>,
> doe...@church.cse.ogi.edu (Thomas Doehne) wrote:
lottery omitted, for brevity.]
>I guess, however, what I really was asking was more concerned
>with people who have "systems" to get the winning numbers. People
>who keep track of previous winners and other such things.
A key part of the system is to filter out numbers that others
choose. So tracking which combinations 'won' is important --
those are combinations that people choose, and therefore should
be avoided if you want to split the pot. Also, when the pots
are large and a lot of people are playing, combinations that
no one hits are ones that you want to generate -- again, so
that you don't split the pot if you win.
Somehow, I don't think this is what you had in mind, tho....
--
Tom Doehne
doe...@cse.ogi.edu
Ken Tetterington
: I see you're still in search of your very first clue. I will answer your
: question about the 1200 coin flips. If heads has come up 1199 times in a
: row using a fair coin, the probability of 1200 in a row is 1/2. You have
: heard that before, right? I won't bother with your 'odds' nonsense; it's
: far too complex for you, as was proven in earlier discussions.
For all of your math statistics, you are telling me that to flip a coin
heads up , twice in a row, out of 2 flips, is 1/2...IDIOT.
You are trying to apply the same odds on one event as over many
consecutive events.
I realy am sorry, for coming down hard on this guy, but the facts are
that even though the odds can be calculated, for a random draw, the facts
are that the random odds _MAY_ not apply to the winning ticket. Read
'Syncronicity' (If I got the spelling right). One of the best "SYSTEMS" I
ever put into a Lottery program was the SUBLIMINAL messages. The message
could be changed in the program. The messages ranged from "I WILL WIN" to
"MY HAIR WON'T FALL OUT". The fact is that I don't think most knew people it
was even there..:-) In the Canadian 6-49, 3 or more grand prize winners
have said that they "HAD A DREAM THE NIGHT BEFORE IN WHICH THEY SAW THE
WINNING NUMBERS". 'LUCK MAG. MAY, 1995, published by Canada Lotteries Corp'
The reason the math guys will not win any grand prize is because they are
looking at the odds, and the odds say they can not win. I uderstand odds
and statistics, but do you understand the way life works. Untill you do,
you will be stuck to the odds.
OUT OF 1 MILLION MATH GUYS, LIGHTNING WILL STRIKE DOWN 1. :-)
Ken...welcome to the next level..
Paul Barnett
>: I see you're still in search of your very first clue. I will answer your
>: question about the 1200 coin flips. If heads has come up 1199 times in a
>: row using a fair coin, the probability of 1200 in a row is 1/2. You have
>: heard that before, right? I won't bother with your 'odds' nonsense; it's
>: far too complex for you, as was proven in earlier discussions.
>For all of your math statistics, you are telling me that to flip a coin
>heads up , twice in a row, out of 2 flips, is 1/2...IDIOT.
No, that isn't what he said. I'll paraphrase it in a way that should
clarify it:
"Even if heads has come up 1199 times in a row using a fair coin, the
probability of heads coming up on the 1200th flip is 1/2."
The probability of a coin landing heads up in two consecutive flips is
1/4, as that is one of the 4 possible outcomes.
>You are trying to apply the same odds on one event as over many
>consecutive events.
No, he is stating that the odds for an event are calculated
independently of the outcome of the unrelated events preceding it.
Stanley T.H. Chow
Kevin Martin Leuthold <leut...@symcom.math.uiuc.edu> wrote:
>I think it depends on what you mean by "winning". In some lotteries,
>you win money if you match only some of the numbers. If you include
>that in your definition of winning, then it seems that you can
>benefit from being systematic.
But you have to include *everything*.
>
>Let's say there's a lottery where you pick 3 numbers out of the
>numbers {1,2,3,4,5}. And if 2 or 3 out of 3 numbers are picked,
>you win a prize. Now let's say i pick the following 2 tickets:
>
>1 2 3
>1 4 5
>
>and you pick the following 2 tickets:
>
>1 2 3
>1 2 4
>
>It's easy to check that no matter what combination comes up,
>I will win a prize, but you will not win if the combination
>3 4 5 comes up.
What if the lucky combination is 1 2 3? Then both sets win
the jackpot, but only second set wins 2 out of 3. So that
evens everything out. (not yet)
You have to go through each of the possible winning combos
(120 of them if I am not mistaken), determine the winnings
for each of the sets, then sum. The only sensible comparision
is between the total sums. I am pretty sure that you will
find exactly the same number of prizes of each type.
>(by the way, I have just started
>following this group. Is this what people mean when they refer
>to "wheeling")
Yes. And it is a useful concept since it reduces variance.
Thomas Doehne
>
>: I see you're still in search of your very first clue. I will answer your
>: question about the 1200 coin flips. If heads has come up 1199 times in a
>: row using a fair coin, the probability of 1200 in a row is 1/2. You have
>: heard that before, right? I won't bother with your 'odds' nonsense; it's
>: far too complex for you, as was proven in earlier discussions.
>For all of your math statistics, you are telling me that to flip a coin
>heads up , twice in a row, out of 2 flips, is 1/2...IDIOT.
>You are trying to apply the same odds on one event as over many
>consecutive events.
If you read it carefully, he's applying a conditional probability:
"If heads come up 1199 times in a row" means *given* than heads have
already come up 1199 times in a row, the probability that the next toss
is heads (giving 1200 in a row) is 1/2. The probability that the
second of two tosses is heads, *given* that the first toss is heads,
is 1/2. Basic probability, about the probabilities of independent
events.
>I realy am sorry, for coming down hard on this guy, but the facts are
>that even though the odds can be calculated, for a random draw, the facts
>are that the random odds _MAY_ not apply to the winning ticket. Read
>'Syncronicity' (If I got the spelling right). One of the best "SYSTEMS" I
>ever put into a Lottery program was the SUBLIMINAL messages. The message
>could be changed in the program. The messages ranged from "I WILL WIN" to
>"MY HAIR WON'T FALL OUT". The fact is that I don't think most knew people it
>was even there..:-) In the Canadian 6-49, 3 or more grand prize winners
>have said that they "HAD A DREAM THE NIGHT BEFORE IN WHICH THEY SAW THE
>WINNING NUMBERS". 'LUCK MAG. MAY, 1995, published by Canada Lotteries Corp'
>
>The reason the math guys will not win any grand prize is because they are
>looking at the odds, and the odds say they can not win. I uderstand odds
>and statistics, but do you understand the way life works. Untill you do,
>you will be stuck to the odds.
My comment on this is: 'a mystic and his money are soon parted'.
--
Tom Doehne
doe...@cse.ogi.edu
Thomas Sippel - Dau
In article <3rh41t$c...@big.aa.net>, hilb...@big.aa.net (John Griffin) writes:
- In article <3rga7g$d...@news.sas.ab.ca>,
- Ken Tetterington <par...@sas.ab.ca> wrote:
-
- I see you're still in search of your very first clue. I will answer your
- question about the 1200 coin flips. If heads has come up 1199 times in a
- row using a fair coin, the probability of 1200 in a row is 1/2. You have
- heard that before, right? I won't bother with your 'odds' nonsense
Well I don't like talking about "odds" either, but in your case, what is the
probability that you have a fair coin ? And with what confidence would you
say that the coin is "fair" or "unfair" (double headed, say) ?
Thomas
--
*** This is the operative statement, all previous statements are inoperative.
* email: cma...@ic.ac.uk (Thomas Sippel - Dau) (uk.ac.ic on Janet)
* voice: +44 171 594 6904 (day) fax: +44 171 594 6958
* snail: Imperial College of Science, Technology and Medicine
Thomas Sippel - Dau
- In article <3r9tej$h00...@bajoriap.perkin-elmer.com>
- Bajo...@Perkin-Elmer.com (Andrew P. Bajorinas) writes:
-
-
- > I see many people in this group looking for a "system" to win the
- >lottery. Are they serious? This is not a flame, but a serious question.
- >Since the drawing is a random drawing what system could exist? Belief that
- >some system exists seems illogical to me. So I guess my question is:
- >
- > What reasons are there to believe any system exists?
There are systems, but not to win, but to avoid wasting money.
Say you take (in 6 of 49) the numbers 1, 2, 3, 4, and all combinations
of the remaining ones. Thus you make 22*45 or 990 bets. Filling in
990 ickets is a bore, and you may well make a mistake, such as leaving
out (... 23, 25) and playing (... 22, 24) twice. If (1, 2, 3, 4, 22, 24)
comes up, you split the maximum with yourself, which is a silly thing
to do if there is no cap on a single row win, and loathsome if there
are no other winners.
If the other combination came up, you would really start kicking yourself.
Thus some lotteries have "system playing tickets", where you state the
rules and pay the cash, and the computer generates all the bets for you.
Ron Meisenheimer
>
>ro...@ns.net (Ron Meisenheimer) writes:
>
>>In article <3rij6l$7...@ixnews5.ix.netcom.com>, jam...@ix.netcom.com (JAMES SARNELLI) says:
>>>
>>>(Andrew P. Bajorinas) writes:
>>>>
>>>> I see many people in this group looking for a "system" to win the
>>>question.
>>>that
>>>>
>
>>>
>>> Using a system refers to taking a logical approach to playing the
>>>lottery/lotto games. Winning the lottery has metaphorically been
>>>compared to getting struck by lightening. Although getting struck by
>>>lightening is improbable there are numerous actions that you can take
>>>to increase your probability of participating in such an event. The
>>>same is true of the lottery.
>
>> Only by buying more tickets.
>
>>>In the next few days I will begin posting some logical methods of
>>>lottery/lotto play. Try out the methods and
>>>confirm for yourself that playing systematically increases your
>>>probability of matching numbers drawn.
>
>> Not true, as has been pointed out to you many, many times. If you
>> buy n different tickets using *any* system, and I buy n different
>> tickets using no system whatever, your chance of winning is
>> identically equal to my own, a trivial result in elementary
>> probability.
>
>you win money if you match only some of the numbers. If you include
>that in your definition of winning, then it seems that you can
>benefit from being systematic.
>
years ago Pythagoras proved his famous theorem about right-triangles:
The square of the hypotenuse is equal to the sum of the squares of
the other two sides. It was *proven*. Do you understand?
Pythagoras proved it irrefutably. There is no need for us to
entertain *any* claim that there exists a right-triangle that
violates the Pythagorean Theorem. We can dismiss any such claim
out-of-hand. We don't have to give it a moment's thought. No
such triangle exists and it is a monumental waste of time and
effort to try to find one. Those who insist otherwise are merely
delusional.
Kevin, I in no way mean this to be an attack upon you. Failing
to have studied probability is no sin. The sin is in obstinately
insisting that you have knowledge where you have none, as Sarnelli,
Ken T., et. al., have done. You seem to be a very reasonable
person who has come here to learn something, and I cannot attack
you for that. There are many, many things I do not know.
However, those of us who are familiar with probability *know*,
because it has been *proven* irrefutably, as surely as the
Pythagorean Theorem has, that no "system" can improve your
probability of winning the jackpot or any portion of it, or,
for that matter, any lesser prize (unless you purchase more
tickets, or are involved in a rigged lottery). We do not have
to consider *any* proposed exception. And so, out of no
disrespect to you, I will not even consider your example.
It cannot be right and no right-triangle violates Pythogoras's
ancient law!
Regards,
Ron Meisenheimer (ro...@ns.net)
Kevin Martin Leuthold
OK, I admit I was wrong, as Stanley Chow pointed out, in claiming
that my "system" would be "better" than another "system". The expected
value of each "system" would be the same, and I overlooked this.
However, I stick to my original claim that of the two "systems"
I outlined in my first post, the first has a probability of 1
of winning a prize, and the second has a probability of less than
1 of winning a prize.
So if you claim that
> no "system" can improve your
> probability of winning the jackpot or any portion of it, or,
> for that matter, any lesser prize (unless you purchase more
> tickets, or are involved in a rigged lottery)
I don't believe you, but if you change that statement to:
no "system" can improve your
*expected winnings* (unless you purchase more
tickets, or are involved in a rigged lottery)
then I find that believable.
Since you have obviously studied more probability than I have,
however, you must be right and I must be delusional. So could
you please consider my example and tell me what's wrong with it.
And please keep the explanation very simple, with lots of
references to the Pythagorean theorem, so you don't lose me.
--
Kevin Leuthold
John Griffin
Ken Tetterington <par...@sas.ab.ca> wrote:
>John Griffin (hilb...@big.aa.net) wrote:
>
>: I see you're still in search of your very first clue. I will answer your
>: question about the 1200 coin flips. If heads has come up 1199 times in a
>: row using a fair coin, the probability of 1200 in a row is 1/2. You have
>: heard that before, right? I won't bother with your 'odds' nonsense; it's
>: far too complex for you, as was proven in earlier discussions.
>
>For all of your math statistics, you are telling me that to flip a coin
>heads up , twice in a row, out of 2 flips, is 1/2...IDIOT.
Enough said. This statement completely defines Ken Tetterington.
What I pointed out to you was one of the simplest of concepts.
Try again, Ken, I know this is really, really difficult: If I flipped
a coin a minute ago, and it came up heads, the probability of heads on
the next flip is 1/2! Amazing, isn't it? Not to confuse you, but the
probability is still 1/2 if it came up tails before. Now, of course
you will dispute that, because you just *know* that the probability
of the sequence TH is only 1/4, right?
You really should stay away from discussions involving rational thought.
>
>You are trying to apply the same odds on one event as over many
>consecutive events.
>
>I realy am sorry, for coming down hard on this guy, but the facts are
>that even though the odds can be calculated, for a random draw, the facts
>are that the random odds _MAY_ not apply to the winning ticket. Read
>'Syncronicity' (If I got the spelling right). One of the best "SYSTEMS" I
>ever put into a Lottery program was the SUBLIMINAL messages. The message
>could be changed in the program. The messages ranged from "I WILL WIN" to
>"MY HAIR WON'T FALL OUT". The fact is that I don't think most knew people it
>was even there..:-) In the Canadian 6-49, 3 or more grand prize winners
>have said that they "HAD A DREAM THE NIGHT BEFORE IN WHICH THEY SAW THE
>WINNING NUMBERS". 'LUCK MAG. MAY, 1995, published by Canada Lotteries Corp'
That is totally fucking hilarious, and of course you didn't get the
spelling right. I agree that the mindless bullshit is the best thing
ever put into a "lottery program".
>
>The reason the math guys will not win any grand prize is because they are
>looking at the odds, and the odds say they can not win. I uderstand odds
>and statistics, but do you understand the way life works. Untill you do,
>you will be stuck to the odds.
Let me get this straight, because it doesn't make any sense on the
surface. You say that if I buy a lottery ticket, I can't win because
I understand that it isn't likely? That's pretty far-fetched, but
it isn't nearly as far-fetched as your claim that you understand "odds
and statistics".
>
>OUT OF 1 MILLION MATH GUYS, LIGHTNING WILL STRIKE DOWN 1. :-)
After which all the others are completely safe from lightning, right?
>
>Ken...welcome to the next level..
No thanks; I'll stay up on this one.
JAMES SARNELLI
>
>JAMES SARNELLI (jam...@ix.netcom.com) wrote:
>: In <3r9tej$h00...@bajoriap.perkin-elmer.com>
Bajo...@Perkin-Elmer.com
>: (Andrew P. Bajorinas) writes:
>: >
>: >
>: >
>: > I see many people in this group looking for a "system" to win
the
>: >lottery. Are they serious? This is not a flame, but a serious
>: question.
>: >Since the drawing is a random drawing what system could exist?
Belief
>: that
>: >some system exists seems illogical to me. So I guess my question
is:
>: >
>: > What reasons are there to believe any system exists?
>
-----------------------------------------------------------------------
>: Andrew
>
>: Using a system refers to taking a logical approach to playing the
>: lottery/lotto games. Winning the lottery has metaphorically been
>: compared to getting struck by lightening. Although getting struck by
>: lightening is improbable there are numerous actions that you can
take
>: to increase your probability of participating in such an event. The
posting
>: some logical methods of lottery/lotto play. Try out the methods and
>: confirm for yourself that playing systematically increases your
>: probability of matching numbers drawn.
>
>: JAM...@AOL.COM
>
>Andrew
>
> You will "laugh yer ass off" when Sarnelli starts posting his
nonsense.
>Don't take him seriously - he has been very, very ill. His last
sentence
>above shows that he hasn't recovered yet.
Hillbilly, I think you have over dosed on hog jowls or something. ?You
can now take pride in spewing your obcscenities in 2 news groups.
MORE TO FOLLOW
JAM...@AOL.COM
John Griffin
Thomas Sippel - Dau <vul...@imperial.ac.uk> wrote:
>
>In article <3rh41t$c...@big.aa.net>, hilb...@big.aa.net (John Griffin) writes:
>- In article <3rga7g$d...@news.sas.ab.ca>,
>- Ken Tetterington <par...@sas.ab.ca> wrote:
>-
>- I see you're still in search of your very first clue. I will answer your
>- question about the 1200 coin flips. If heads has come up 1199 times in a
>- row using a fair coin, the probability of 1200 in a row is 1/2. You have
>- heard that before, right? I won't bother with your 'odds' nonsense
>
>Well I don't like talking about "odds" either, but in your case, what is the
>probability that you have a fair coin ? And with what confidence would you
>say that the coin is "fair" or "unfair" (double headed, say) ?
This is fun! I'll answer your question too, by god. Given what I said:
"If heads has come up 1199 times in a row using a fair coin...",
the probability that I have a fair coin is 1.0. In an infinite number
of trials, 1199 consecutive heads will come up sometime.
John Griffin
>- Bajo...@Perkin-Elmer.com (Andrew P. Bajorinas) writes:
>-
>-
>- > I see many people in this group looking for a "system" to win the
>- >lottery. Are they serious? This is not a flame, but a serious question.
>- >Since the drawing is a random drawing what system could exist? Belief that
>- >some system exists seems illogical to me. So I guess my question is:
>- >
>- > What reasons are there to believe any system exists?
>
>There are systems, but not to win, but to avoid wasting money.
>
>Say you take (in 6 of 49) the numbers 1, 2, 3, 4, and all combinations
>of the remaining ones. Thus you make 22*45 or 990 bets. Filling in
>990 ickets is a bore, and you may well make a mistake, such as leaving
>out (... 23, 25) and playing (... 22, 24) twice. If (1, 2, 3, 4, 22, 24)
>comes up, you split the maximum with yourself, which is a silly thing
>to do if there is no cap on a single row win, and loathsome if there
>are no other winners.
Was this paragraph supposed to relate to the "avoid wasting money"
comment? Your 990 tickets are *precisely* the same as any other
990 tickets.
>
>If the other combination came up, you would really start kicking yourself.
>Thus some lotteries have "system playing tickets", where you state the
>rules and pay the cash, and the computer generates all the bets for you.
This should be illegal. They are suggesting to people that a profitable
system might exist, conning suckers into buying more tickets than they
would normally buy. This just preys on retards and daydreamers.
Normand Veilleux
>From: par...@sas.ab.ca (Ken Tetterington)
>
>John Griffin (hilb...@big.aa.net) wrote:
>
>: I see you're still in search of your very first clue. I will answer your
>: question about the 1200 coin flips. If heads has come up 1199 times in a
>: row using a fair coin, the probability of 1200 in a row is 1/2. You have
>: heard that before, right? I won't bother with your 'odds' nonsense; it's
>: far too complex for you, as was proven in earlier discussions.
>
>For all of your math statistics, you are telling me that to flip a coin
>heads up , twice in a row, out of 2 flips, is 1/2...IDIOT.
ROTFL!
Great minds think alike and fools seldom differently.
Perhaps a small example will show you who the real fool is:
Let's list all possible events for 1 flip:
H
T
So, heads has 1 chance in 2 of coming up on the first flip.
Now, lets list all possible events for 2 flips:
HH
HT
TH
TT
Assuming, we have already witnessed the first flip and we know
it was H, then we can eliminate from the list of possibilities
all those that had a T on the first flip since we know this
did not happen and we can't go back and change it eihter. So,
we are left with
HH
HT
Notice that there is 1 chance in 2 that heads comes up on the
second flip after it came up on the first flip. Pure, simple,
logic, that any 5 year old can understand.
Now, lets get ambitious and list all possibilities for 3 flips:
HHH
HHT
HTH
HTT
THH
THT
TTH
TTT
Assuming, we have already witnessed the 2 first flips and we know
it was HH, then we can eliminate from the list of possibilities
all those that had HT, TH, or TT on the first 2 flips since we know
this did not happen and we can't go back and change it eihter. So,
we are left with
HHH
HHT
Notice again that there is 1 chance in 2 that heads comes up on
the third flip after it came up on the first 2 flips. Isn't that
amazing! We keep adding and adding and adding flips and heads
keeps having and having and having the same odds for every flip!
There are 8 possible combinations if you don't any of the flips,
but as soon as you have seen the first two flips there are only
2 combinations left that are possible. Try it with TT if you
like and see what is the chance that H comes up next.
Let's flip Ken, heads I win, tails you lose! :-)
If the Lotto program you wrote is based on your misunderstanding
of something as simple as coin flips, then it's not worth much,
to say the least!
Normand Veilleux
>From: sc...@bnr.ca (Stanley T.H. Chow)
>
>>Kevin Martin Leuthold <leut...@symcom.math.uiuc.edu> wrote:
>>
>>Let's say there's a lottery where you pick 3 numbers out of the
>>numbers {1,2,3,4,5}. And if 2 or 3 out of 3 numbers are picked,
>>you win a prize. Now let's say i pick the following 2 tickets:
>>
>>1 2 3
>>1 4 5
>>
>>and you pick the following 2 tickets:
>>
>>1 2 3
>>1 2 4
>>
>>It's easy to check that no matter what combination comes up,
>>I will win a prize, but you will not win if the combination
>>3 4 5 comes up.
>
>What if the lucky combination is 1 2 3? Then both sets win
>the jackpot, but only second set wins 2 out of 3. So that
>evens everything out. (not yet)
>
>You have to go through each of the possible winning combos
>(120 of them if I am not mistaken), determine the winnings
>for each of the sets, then sum.
Aaaaargh!! Watch how use that word! It's been a nightmare
recently and then some! :-)
>The only sensible comparision
>is between the total sums. I am pretty sure that you will
>find exactly the same number of prizes of each type.
Actually there are 5*4*3/3/2/1 = 10 possibilities. So, let's
analyse it right here and now. First, let's call 123, 145
set # 1 and 123, 124 set # 2. Now let's list all 10 possible
combinations and see what each one would win for each set if
it were drawn.
| set # 1 | set # 2 |
|---------|---------|---------|---------|
draw | # of 2 | # of 3 | # of 2 | # of 3 |
| matches | matches | matches | matches |
-----|---------|---------|---------|---------|
123 | 0 | 1 | 1 | 1 |
124 | 2 | 0 | 1 | 1 |
125 | 2 | 0 | 2 | 0 |
134 | 2 | 0 | 2 | 0 |
135 | 2 | 0 | 1 | 0 |
145 | 0 | 1 | 1 | 0 |
234 | 1 | 0 | 2 | 0 |
235 | 1 | 0 | 1 | 0 |
245 | 1 | 0 | 1 | 0 |
345 | 1 | 0 | 0 | 0 |
-----|---------|---------|---------|---------|
Total| 12 | 2 | 12 | 2 |
So, on average, each set wins 1.2 times a 2 number match and
.2 times a 3 number match. Hence, ON AVERAGE the two sets are
identical since their average expected gain is the same. BTW,
if you were to compare any other set of 2 tickets you would
find the exact same result.
Note that set # 1 does indeed "cover" all possibilities since
no matter which one is drawn it always wins something. Hence,
set # 1 is indeed a "wheel" with a minimum win-guarantee. Set
# 2 does not have a minimum win-guarantee since it does not win
anything when 345 is drawn. Yet, amazingly enough it has the
same average expected gain!
The same logic applies to the Lotto 6/49. Any set of N tickets,
no matter how the tickets are selected, has exactly the same
average expected gain. Using sums can't change that; playing
HOT numbers can't change that, purchasing SETGEN can't change
that and using "wheels" can't change that either, assuming of course
that all possibilities are equally probable (which is what is meant
when we say the draw is random). The only way these "tools" would
have a chance at increasing the average expected gain is IF THE
COMBINATIONS WERE NOT EQUALLY PROBABLE. And so far, no one, I repeat
no one has ever been able to prove mathematically that is the case.
BTW, if this is ever proven right (which is EXTREMELY unlikely) it
would also mean that these same "tools" could, in certain instances
reduce your average expected gain. So, even if the situation that
all the pseudo Lotto experts like Ken T., James S. and Lon V. believe
exists, would be proven to be true, then it would still not be a
WIN-WIN situation. You would WIN some and you would LOSE some.
Wouldn't it be a bummer though if so and so's SUPER DUPER LOTTO
program would keep making predictions that would keep reducing
your gains! What kind of ADVANTAGE would that be? Thank God that,
as long as every combination is equally probable, everyone is
protected from such silly programs.
Ron Meisenheimer
Blast! Why couldn't the first person to question me be stupid, beligerent,
and unreasonable? I'm ready for *him!* Ah, well . . . such is life [whine].
You're example does indeed guarantee you at least some return on
each bet you make, while my picks leave me flat if 3-4-5 come up. Let
J stand for the jackpot and B stand for the booby-prize (matching two
of three numbers). If you buy the two tickets 1-2-3 and 1-4-5, while I buy
the two tickets 1-2-3 and 1-2-4 in a 3/5 lottery, there are 10 possibilities
as below:
Winning Numbers You Win I Win
---------------------------------------------------------
1 2 3 J J + B
1 2 4 B + B J + B
1 2 5 B + B B + B
1 3 4 B + B B + B
1 3 5 B + B B
1 4 5 J B
2 3 4 B B + B
2 3 5 B B
2 4 5 B B
3 4 5 B -
---------------------------------------------------------
#J's=2 #J's=2
#B's=12 #B's=12
Now, if I've done this tedious exercise correctly, there are two ways for
you to win the jackpot, same as me. There are 12 possibilities for you
to win a booby-prize, same as me. While, there is an appearance that
your way is "better" than mine, because I cannot recoup any of my
losses if 3-4-5 come up, my way is "better" when I win the jackpot, for
you can never win as much as J + B. Any advantage your system has
in the 3-4-5 case is precisely and exactly cancelled by the advantage
my non-system has in other cases.
The expectation of the two approaches is exactly the same. The number
of jackpots and the number of booby-prizes are exactly the same with
either your wheel or my non-system. There is no basis for choosing
one system over the other, and neither one guarantees that you can
"recoup [btw, considerably more than just recoupment is the goal, no?]
more of your money" than the other. Guaranteeing that you "win" at
least something on each and every draw is no better than abandoning
that goal and accepting a complete loss in some cases, while
guaranteeing a slightly higher return in others.
It is not the raw number of recoupments per draw that determines
whether or not you are a winner. You can get back a pittance on each
and every draw and still come out badly behind the eight ball. The *only*
thing that counts is whether or not your expectation is positive or negative.
In all state-run lotteries it is negative.
Now, before someone says, "Yeah, but if you get the booby-prize, it's five
dollars in your pocket and the bet only cost you one dollar," take a closer
look at the table. In some cases, Kevin's wheel only returned one booby-
prize and he had to buy *two* tickets to get it. In a real lottery, you would
have to spend considerably more than two dollars to guarantee at least
one five-dollar booby-prize on each of 18 million (California) combinations.
Well, back to my bulldog stance. I will consider no more "systems." The
situation is every bit as hopeless as trying to find a right-triangle in plane
geometry that violates the Pythagorean Theorem. Our time is too valuable
to waste on such pipedreams.
Any help?
Regards,
Ron Meisenheimer (ro...@ns.net)
P.S. Kevin, no, no, no, you are *not* delusional. You are one of the few
naysayers who seems sincerely interested in an explanation, and your
proposed counterexample was quite clever.
Thomas Sippel - Dau
In article <3rl5jm$c...@big.aa.net>, hilb...@big.aa.net (John Griffin) writes:
- In article <3rkjkh$m...@oban.cc.ic.ac.uk>, Thomas Sippel - Dau <vul...@imperial.ac.uk> wrote:
- >Well I don't like talking about "odds" either, but in your case, what is the
- >probability that you have a fair coin ? And with what confidence would you
- >say that the coin is "fair" or "unfair" (double headed, say) ?
-
- This is fun! I'll answer your question too, by god. Given what I said:
- "If heads has come up 1199 times in a row using a fair coin...",
- the probability that I have a fair coin is 1.0. In an infinite number
- of trials, 1199 consecutive heads will come up sometime.
And this is exactly where probability and statistics diverge. You *assumed*
you had a fair coin, so there is no noton of probability. In a real experiment
you would have to *prove* that the actual coin you are using is "fair".
With statistics, you watch the coin being flipped, and decide after a given
number of tosses if you accept it as fair. With your particular coin, most
people would long have "accepted" the coin as a double-headed one, and thus
rejected your claim that it was "fair", and they could have done that with
an extreely high level of confidence.
Thomas Doehne
>In article <3rk9q1$3...@vixen.cso.uiuc.edu>, leut...@symcom.math.uiuc.edu
(Kevin Martin Leuthold) says:
>>
>>I think it depends on what you mean by "winning". In some lotteries,
>>you win money if you match only some of the numbers. If you include
>>that in your definition of winning, then it seems that you can
>>benefit from being systematic.
[Ron talks about the Pythagorean theorem, and mathematical
knowledge, as prelude to the following claim:]
> However, those of us who are familiar with probability *know*,
> because it has been *proven* irrefutably, as surely as the
> Pythagorean Theorem has, that no "system" can improve your
> probability of winning the jackpot or any portion of it, or,
> for that matter, any lesser prize (unless you purchase more
> tickets, or are involved in a rigged lottery). We do not have
> to consider *any* proposed exception. And so, out of no
> disrespect to you, I will not even consider your example.
> It cannot be right and no right-triangle violates Pythogoras's
> ancient law!
Two points, one just for fun, the other serious. First, is it
true that the Pythagorean theorems hold in some of the funkier
spaces, like hyperbolic (that's the saddle-shaped one, yes?).
Second, while it is true that no system can improve your
expected payout (except by avoiding pot splitting), you
can adopt a system that will the number of payouts you
can expect to get for buying some N tickets (while lowering
the amount you will get if it pays out). Also, it is
possible to guarantee that you will win something, but
your N has to be very large...and will require buying
tickets that will split the pot if they win.
I'm not claiming that you can increase the amount you'd
expect to win (you can't), but you can alter the chance that
you'd win something.
To make an analogy, if you play the stock market, you can
invest in blue chip stocks, or in penny stocks. If you
invest in blue chip stocks, you have a much higher chance
of having a positive return on your investment (but they
aren't likely to be spectacular returns). If you invest in penny
stocks (and manage to avoid the frauds), you're not very
likely to get a positive return, but if the investment does
happen to pay off, your return is sometimes splendid. Something
that doesn't happen with the plodding blue chips.... So you're
accepting a higher degree of risk (no payout) in exchange for
a spectacular payout if you do hit. And in the lotto case,
you can do the opposite, to some extent.
Make sense?
--
Tom Doehne
doe...@cse.ogi.edu
Thomas Doehne
>In article <3rkkgg$n...@oban.cc.ic.ac.uk>,
>>- Bajo...@Perkin-Elmer.com (Andrew P. Bajorinas) writes:
>>- > What reasons are there to believe any system exists?
>>There are systems, but not to win, but to avoid wasting money.
>>Say you take (in 6 of 49) the numbers 1, 2, 3, 4, and all combinations
>>of the remaining ones. Thus you make 22*45 or 990 bets. Filling in
>>990 ickets is a bore, and you may well make a mistake, such as leaving
>>out (... 23, 25) and playing (... 22, 24) twice. If (1, 2, 3, 4, 22, 24)
>>comes up, you split the maximum with yourself, which is a silly thing
>>to do if there is no cap on a single row win, and loathsome if there
>>are no other winners.
>
>Was this paragraph supposed to relate to the "avoid wasting money"
>comment? Your 990 tickets are *precisely* the same as any other
>990 tickets.
Not necessarily so. First, we have to assume that the lotto
player is trying to minimize risk. That is, that the player
is trying to get the greatest chance at *some* payout, by
sacrificing the chance at a higher payout on some combinations.
(This is the idea behind portfolio theory -- you spread your
money over a large number of chances, so that you won't win
or lose big.)
Now, take a lottery that pays out if you get a match on three
numbers, in addition to payouts if you match more than that.
If you want to minimize risk, you want to try to buy tickets
that cover as many of the three-number combinations with as
few tickets as possible. (Is this what wheeling tries to
do?) Note that this *does not* change the expected return
of the bet. What it does is lower the variability of the
return. [It's too late for the math, and I'd have to go
back to my probability books, but here's the gist: there
are X many combinations in a 46-choose-6 pool. There are
many, many fewer in a 46-choose-3 (call that number Y). If
I remember correctly, 46-choose-6 has somewhere around
10-12 million combinations. You could conceivably buy them
all with enough money (and I think someone tried with a
lottery in Virginia/West Virginia a couple of years ago).
It takes a lot less money to cover all the 3-number combos.
--
Tom Doehne
doe...@cse.ogi.edu
Thomas Doehne
>
>>From: sc...@bnr.ca (Stanley T.H. Chow)
>>
>>>Kevin Martin Leuthold <leut...@symcom.math.uiuc.edu> wrote:
>Note that set # 1 does indeed "cover" all possibilities since
>no matter which one is drawn it always wins something. Hence,
>set # 1 is indeed a "wheel" with a minimum win-guarantee. Set
># 2 does not have a minimum win-guarantee since it does not win
>anything when 345 is drawn. Yet, amazingly enough it has the
>same average expected gain!
>The same logic applies to the Lotto 6/49. Any set of N tickets,
>no matter how the tickets are selected, has exactly the same
>average expected gain.
[Bunch of stuff about various systems deleted for irrelevance
to my point]
Expected gain is unchanged. Risk, however, is lowered. Playing
a set that covers all possibilities is less risky, in that the
variance between the possible returns is lower. The average
is the same, but the variance of the payoffs is lower. Lowering
the variance is useful if you're mildly risk averse.
If you want the thrill of winning, you're more likely to win if
you spread your bets. You'll win less each time (as you
illustrated), but win more often. Why not? The expected
value of the return doesn't change.
If you're more than mildly risk averse, you shouldn't be
playing lotteries anyway. :-)
--
Tom Doehne
doe...@cse.ogi.edu
Newsgroups: rec.gambling.lottery
Subject: Re: Method to beat the lottery
Summary:
Expires:
References: <3rk9q1$3...@vixen.cso.uiuc.edu> <3rkhuc$f...@bmerhc5e.bnr.ca> <nveilleu.13...@emr1.emr.ca>
Sender:
Followup-To:
Distribution:
Organization: Oregon Grad. Inst. Computer Science and Eng., Beaverton
Keywords:
Peter Sullivan
>
>In article <3rk9q1$3...@vixen.cso.uiuc.edu>, leut...@symcom.math.uiuc.edu
(Kevin Martin Leuthold) says:
>>
>>ro...@ns.net (Ron Meisenheimer) writes:
>>
>>>>(Andrew P. Bajorinas) writes:
>>>>>
>>>>> I see many people in this group looking for a "system" to win the
>>>>question.
>>>>that
>>>>>
>>
>>>>
>>>> Using a system refers to taking a logical approach to playing the
>>>>lottery/lotto games. Winning the lottery has metaphorically been
>>>>compared to getting struck by lightening. Although getting struck by
>>>>lightening is improbable there are numerous actions that you can take
>>>>to increase your probability of participating in such an event. The
>>>>same is true of the lottery.
>>
>>
>>>>lottery/lotto play. Try out the methods and
>>>>confirm for yourself that playing systematically increases your
>>>>probability of matching numbers drawn.
>>
>>> buy n different tickets using *any* system, and I buy n different
>>> tickets using no system whatever, your chance of winning is
>>> identically equal to my own, a trivial result in elementary
>>> probability.
>>
>>you win money if you match only some of the numbers. If you include
>>that in your definition of winning, then it seems that you can
>>benefit from being systematic.
>>
> years ago Pythagoras proved his famous theorem about right-triangles:
> The square of the hypotenuse is equal to the sum of the squares of
> the other two sides. It was *proven*. Do you understand?
> Pythagoras proved it irrefutably. There is no need for us to
> entertain *any* claim that there exists a right-triangle that
> violates the Pythagorean Theorem. We can dismiss any such claim
> out-of-hand. We don't have to give it a moment's thought. No
> such triangle exists and it is a monumental waste of time and
> effort to try to find one. Those who insist otherwise are merely
> delusional.
>
> Kevin, I in no way mean this to be an attack upon you. Failing
> to have studied probability is no sin. The sin is in obstinately
> insisting that you have knowledge where you have none, as Sarnelli,
> Ken T., et. al., have done. You seem to be a very reasonable
> person who has come here to learn something, and I cannot attack
> you for that. There are many, many things I do not know.
>
> because it has been *proven* irrefutably, as surely as the
> Pythagorean Theorem has, that no "system" can improve your
> probability of winning the jackpot or any portion of it, or,
> for that matter, any lesser prize (unless you purchase more
> tickets, or are involved in a rigged lottery). We do not have
> to consider *any* proposed exception. And so, out of no
> disrespect to you, I will not even consider your example.
> It cannot be right and no right-triangle violates Pythogoras's
> ancient law!
>
>
>Ron Meisenheimer (ro...@ns.net)
>
Ron (& anyone else who has been following this thread):
I agree that there is no proven method for artificially increasing your
probability of matching numbers drawn (as James Sarnelli promises). The
large jackpot games such as powerball and others attract people with the
possibility that they can win millions. Although they do deliver to a few
fortunate individuals every once in a while, the expected gain is simply too
low to bother. If you're going to throw money away, throw it someplace where
it at least might do some good (rather than lining government coffers).
However, if you must play the lottery, do your homework. You can
substantially increase your expected gain by playing games that offer lower
prizes but better expected returns. Also, some of these games are fixed
payout games whereby winners do not share a prize pool. Games such as
5-minute Keno aren't too bad (beware the replayability trap though).
Good Luck,
Peter Sullivan (psul...@gtech.com)
Daniel Efron
(Thomas Doehne) wrote:
> Second, while it is true that no system can improve your
> expected payout (except by avoiding pot splitting), you
> can adopt a system that will the number of payouts you
> can expect to get for buying some N tickets (while lowering
> the amount you will get if it pays out). Also, it is
> possible to guarantee that you will win something, but
> your N has to be very large...and will require buying
> tickets that will split the pot if they win.
>
> I'm not claiming that you can increase the amount you'd
> expect to win (you can't), but you can alter the chance that
> you'd win something.
You can decrease variance, but you don't increase the chance of winning
something over simply buying N distinct tickets. If the lottery behaves
as random, indepedent events, buying all N combinations of one draw is
statistically the same as buying N tickets over many draws as far as
winning tickets are concerned.
Buying all N combinations of the lottery does guarantee an exact set of
payoffs (and you will lose approximately 50% of your money), but N distinct
tickets in a number of draws will produce the same number of winners within
an expected standard deviation. As N gets larger (buying all tickets in
many drawings, for instance), the number of prizes won from both methods
will continue to converge.
You have not increased your chance of winning, you have merely decreased
your variance.
--
Dan Efron
sha...@vm.cc.purdue.edu
leut...@symcom.math.uiuc.edu (Kevin Martin Leuthold) writes:
<James Sarnelli's bad lightning analogy and empty promise to post his
methods deleted>
>
>I think it depends on what you mean by "winning". In some lotteries,
>you win money if you match only some of the numbers. If you include
>that in your definition of winning, then it seems that you can
>benefit from being systematic.
>
>numbers {1,2,3,4,5}. And if 2 or 3 out of 3 numbers are picked,
>you win a prize. Now let's say i pick the following 2 tickets:
>
>1 2 3
>1 4 5
>
>and you pick the following 2 tickets:
>
>1 2 3
>1 2 4
>
>It's easy to check that no matter what combination comes up,
>I will win a prize, but you will not win if the combination
>3 4 5 comes up.
>
1 2 3 or 1 2 4 pop up - (or even 1 2 5). In those cases (you) only
win a single prize. Check all your combinations - sum up all the
prizes won, and you'll see that they come out *exactly* the same.
>Now, I doubt that in a real lottery, you could actually guarantee
>that you would win money this way, but it seems that you could
>come up with a system that would be guaranteed to recoup more
>of your money than another system via the smaller prizes, and
>following this group. Is this what people mean when they refer
>to "wheeling")
that required you to buy the same number of tickets.
And yeah, that's a form of wheeling.
Mike Sharkey
wasted space so this damn program will send
sha...@vm.cc.purdue.edu
>I outlined in my first post, the first has a probability of 1
>of winning a prize, and the second has a probability of less than
>1 of winning a prize.
be a good thing. But it does so at the cost of reducing the likelihood
of getting *two* winners at the same time - which isn't a good thing.
The NPV of either 'system' is the same.
Mike Sharkey
Thomas Sippel - Dau
In article <defron-1406...@x90-41.upress.umn.edu>, def...@maroon.tc.umn.edu (Daniel Efron) writes:
-
- You can decrease variance, but you don't increase the chance of winning
- something over simply buying N distinct tickets.
Please, please, language.
The more tickets you buy, the higher your chance of winning something.
The more tickets you buy, the higher the probability of winning a high
prize, (par force), but also the the probability of getting a dud (in
fact this is almost guaranteed).
Once you have a dud, the variance of your winnings (being an average of
the squares of the difference of the winnings and their mean) is dominated by:
1. the highest winning ticket you have (giving a big numerator)
2. the number of tickets you bought (increasing the nominator)
Thus, buy 1, win $1000, variance 0
buy 1, a dud variance 0
buy 2, win $1000 and a dud, variance 125,000
buy 1000, win $100000, $10, 9998 duds variance 10,080,016
If the lottery behaves
- as random, indepedent events, buying all N combinations of one draw is
- statistically the same as buying N tickets over many draws as far as
- winning tickets are concerned.
**** bzzzt, no. Assume 6 from 49 (13,983,816 possibilities)
If you buy all combinations, you are assured of matching 6. If you buy
1 ticket at random in 13983816 draws, the probability of matching 6
at least once is the complement of the probability of not matching 6
in any of the draws, i.e. the complement of:
(13983815/13983816) ** 13983816 = 0.36787943 >=< 1/e
or about 63.212%
- Buying all N combinations of the lottery does guarantee an exact set of
- payoffs (and you will lose approximately 50% of your money), but N distinct
- tickets in a number of draws will produce the same number of winners within
- an expected standard deviation. As N gets larger (buying all tickets in
- many drawings, for instance), the number of prizes won from both methods
- will continue to converge.
Nope, the expectation (i.e. the integral of payout over probability) is
constant at 50% or whatever the lottery pays.
- You have not increased your chance of winning, you have merely decreased
- your variance.
No, tendentially you increase it, but not monotonically so (see above).
Ron Meisenheimer
>
>Two points, one just for fun, the other serious. First, is it
>true that the Pythagorean theorems hold in some of the funkier
>spaces, like hyperbolic (that's the saddle-shaped one, yes?).
Here, I must be careful because I've never studied non-euclidean
geometry. The occassional reference I have stumbled across leads
me to believe that the Pythagorean Theorem does not hold except
in plane geometry, the interior angles of a triangle don't sum to
180 degrees, etc. Yes to your saddle-shape question.
>
>Second, while it is true that no system can improve your
>expected payout (except by avoiding pot splitting), you
>can adopt a system that will the number of payouts you
>can expect to get for buying some N tickets (while lowering
>the amount you will get if it pays out). Also, it is
>possible to guarantee that you will win something, but
>your N has to be very large...and will require buying
>tickets that will split the pot if they win.
> [ . . . ]
>
>Make sense?
>
Makes sense. But as Kevin correctly pointed out earlier, we have
to be careful how we define 'win.' Surely, you have not "won" if
you spend more money than you get back. See my last post (apologies
to the group for that last post. I didn't mean for all the earlier
references to get cut out).
Regards,
Ron Meisenheimer (ro...@ns.net)
Martin Veneroso
>If you're going to throw money away, throw it someplace where
>it at least might do some good (rather than lining government coffers).
I don't know about that -- I rather like the idea of other people paying
my state taxes for me by playing the lottery.
I have to ask, though: is the G-Tech from whose system you are posting
the giant lottery systems contractor?
--
Martin Veneroso | Interested in a mailing list about poker in the Bay Area?
sl...@best.com | Write to ba-poker...@best.com for a subscription.
See my Web page at <http://www.best.com/~slick/> for my geek code.
sha...@vm.cc.purdue.edu
cma...@cc.ic.ac.uk (Thomas Sippel - Dau) writes:
>In article <defron-1406...@x90-41.upress.umn.edu>, def...@maroon.tc.umn.edu (Daniel Efron) writes:
> If the lottery behaves
>- as random, indepedent events, buying all N combinations of one draw is
>- statistically the same as buying N tickets over many draws as far as
>- winning tickets are concerned.
>
>**** bzzzt, no. Assume 6 from 49 (13,983,816 possibilities)
>
>If you buy all combinations, you are assured of matching 6. If you buy
>1 ticket at random in 13983816 draws, the probability of matching 6
>at least once is the complement of the probability of not matching 6
>in any of the draws, i.e. the complement of:
>
> (13983815/13983816) ** 13983816 = 0.36787943 >=< 1/e
>
>or about 63.212%
between buying all at once and strung out over a period of time would
be insignificant.
Kinda like in Engineering where, for small angles, the sine of the angle
is often approximated as the angle itself.
In this case, 'small' could be (somewhat arbitrarily) defined as 10% or less
of the total possible combinations.
<more deleted>
>- You have not increased your chance of winning, you have merely decreased
>- your variance.
>
>No, tendentially you increase it, but not monotonically so (see above).
>
> Thomas
Some of us ain't as swift as we'd like to think, but I'm sure willing to
listen and learn - unlike a certain poster on LOTTO LOGIC.
Actually, about ten minutes of thought just went by and I do believe that
Thomas is right. But, God!, I'd like to see this put into simpler terms.
Somewhere at about college freshman (humanities) level would be great.
It'll still be over Mr. Sarnelli's head, but no sense in insulting the
wider majority of readers by talking down to them.
Mike Sharkey
jac...@xmission.com
>In article <3rl621$c...@big.aa.net> hilb...@big.aa.net (John Griffin) writes:
>>In article <3rkkgg$n...@oban.cc.ic.ac.uk>,
>>Thomas Sippel - Dau <vul...@imperial.ac.uk> wrote:
>>>
>>>- In article <3r9tej$h00...@bajoriap.perkin-elmer.com>
>>>- Bajo...@Perkin-Elmer.com (Andrew P. Bajorinas) writes:
>
>>>- > What reasons are there to believe any system exists?
>
>>>There are systems, but not to win, but to avoid wasting money.
>
>>>Say you take (in 6 of 49) the numbers 1, 2, 3, 4, and all combinations
>>>of the remaining ones. Thus you make 22*45 or 990 bets. Filling in
>>>990 ickets is a bore, and you may well make a mistake, such as leaving
>>>out (... 23, 25) and playing (... 22, 24) twice. If (1, 2, 3, 4, 22, 24)
>>>comes up, you split the maximum with yourself, which is a silly thing
>>>to do if there is no cap on a single row win, and loathsome if there
>>>are no other winners.
>>
>>Was this paragraph supposed to relate to the "avoid wasting money"
>>comment? Your 990 tickets are *precisely* the same as any other
>>990 tickets.
>
> Not necessarily so. First, we have to assume that the lotto
> player is trying to minimize risk.
I don't think we have to assume this. In fact, if we *do* assume
that the lotto player is trying to minimize risk, then the conclusion
we reach is that the lotto player won't buy a ticket, since this
minimizes the risk by reducing it to zero.
--
Steve Jacobs (jac...@xmission.com) \ Warped by OS/2
"Expectation isn't everything..." \
Normand Veilleux
>From: doe...@church.cse.ogi.edu (Thomas Doehne)
>
>>Thomas Sippel - Dau <vul...@imperial.ac.uk> wrote:
>>
>>Was this paragraph supposed to relate to the "avoid wasting money"
>>comment? Your 990 tickets are *precisely* the same as any other
>>990 tickets.
>
>Not necessarily so. First, we have to assume that the lotto
>is trying to get the greatest chance at *some* payout, by
>sacrificing the chance at a higher payout on some combinations.
>(This is the idea behind portfolio theory -- you spread your
>money over a large number of chances, so that you won't win
>or lose big.)
>
>Now, take a lottery that pays out if you get a match on three
>numbers, in addition to payouts if you match more than that.
>If you want to minimize risk, you want to try to buy tickets
>that cover as many of the three-number combinations with as
>few tickets as possible. (Is this what wheeling tries to
>do?)
That is exactly what wheeling does since a "wheel" is indeed
a set of tickets that covers of all possible combinations. That
is why a wheel has a minimum win-guarantee. I realize the term
"partial wheel" is also used in the Lotto world, but I don't use
it. A partial wheel cannot be a cover and therefore it is nothing
more than a "regular" set of tickets!
>Note that this *does not* change the expected return
>of the bet. What it does is lower the variability of the
>return.
Right. Buying a set of 10 identical tickets will win you more
money (when it wins) than buying a wheel of 10 different tickets,
but it has a lower probability of winning a prize. So, one set
wins less often, but wins higher amounts and the other set wins
smaller amounts but more often. Hence, on average both sets of
tickets have the same expected gains.
>[It's too late for the math, and I'd have to go
>back to my probability books, but here's the gist: there
>are X many combinations in a 46-choose-6 pool. There are
>many, many fewer in a 46-choose-3 (call that number Y). If
>I remember correctly, 46-choose-6 has somewhere around
>10-12 million combinations. You could conceivably buy them
>all with enough money (and I think someone tried with a
>lottery in Virginia/West Virginia a couple of years ago).
>It takes a lot less money to cover all the 3-number combos.
In the case of the Lotto 6/49, there are 13,983,816 combinations
and ALL those combinations can be covered by a wheel of 174
tickets. This wheel guarantees a 3 on 6 prize, but averages
six of them.
One thing that is worth mentioning about the 3-number combos
too is that a ticket of 6 numbers covers twenty 3-number combos.
Whereas a ticket only covers one 6-number combo! So, not only
are there less 3-number combos, but a ticket of 6 numbers covers
more of them. This combined effect explains why it takes only
174 tickets to cover about 14 million combinations.
Normand Veilleux
>From: cma...@cc.ic.ac.uk (Thomas Sippel - Dau)
>
>-From: John Griffin
>-
>- This is fun! I'll answer your question too, by god. Given what I said:
>- "If heads has come up 1199 times in a row using a fair coin...",
>- the probability that I have a fair coin is 1.0. In an infinite number
>- of trials, 1199 consecutive heads will come up sometime.
>
>And this is exactly where probability and statistics diverge. You *assumed*
>you had a fair coin, so there is no noton of probability. In a real experiment
>you would have to *prove* that the actual coin you are using is "fair".
>
>With statistics, you watch the coin being flipped, and decide after a given
>number of tosses if you accept it as fair. With your particular coin, most
>people would long have "accepted" the coin as a double-headed one, and thus
>rejected your claim that it was "fair", and they could have done that with
>an extreely high level of confidence.
In all fairness though, statistics can be misleading. To show this,
let's contruct a hypothetical scenario and use 200 consecutive heads
instead of 1200.
Now, I do the flipping and a trustworthy statistician does the
data accumulation and analysis. After 10 trillion flips, we
have 5 trillion - 2,000 heads and therefore 5 trillion + 2,000
tails. The statistician does his thing and concludes that we
are indeed using a fair coin.
All this time you were waiting at the door (sorry for the wait! :-)
We let you in and inform you that we are using a fair coin. I
proceed to flip the coin and the first 200 flips you witness are
all heads. Now, according to your superior knowledge of statistics
you would inform me that the coin is most certainly NOT FAIR.
What's wrong with this picture? Is the coin fair or not?
Now let's assume the trustworthy statistician steps in and assures
you that the coin is truly fair. So, you decide to let me continue
flipping. After 10 trillion flips, we have 5 trillion + 3,500
heads and therefore 5 trillion - 3,500 tails. Now, you redo your
calculations and conclude that the coin is indeed fair.
Now, what's wrong with this picture? You just said a minute ago
(I'm a fast flipper OK! :-) that the coin was NOT FAIR and now you
say that it is fair. Make up your mind! Is it fair or not?
What I'm trying to illustrate here is that using statistics, there
is a chance of rejecting a perfectly fair coin based on certain
samples. Luckily though, the amount of samples that will make you
reject such a coin is fairly small compared to the amount of samples
that would make you accept it as fair. Hence, the PROBABILITY :-)
that you will select a "bad" sample is low. That is why we put a
high level of confidence on the result obtained.
See? It's very simple: probability depends on statistics and
statistics depend on probability! :-) NEXT!
Peter Sullivan
>>it at least might do some good (rather than lining government coffers).
>
>my state taxes for me by playing the lottery.
>
>I have to ask, though: is the G-Tech from whose system you are posting
>the giant lottery systems contractor?
>--
Yep. I currently work for Gtech as a senior software engineer at their
world headquarters in Rhode Island (although, after my comment above, I
wonder if I will be for very much longer). I've been in the lottery
business in one way or another for 10 years or so. I think that qualifies
me to make comment on some of this stuff.
Normand Veilleux
>From: ro...@ns.net (Ron Meisenheimer)
>
> Winning Numbers You Win I Win
>---------------------------------------------------------
> 1 2 4 B + B J + B *
> 1 2 5 B + B B + B
> 1 3 4 B + B B + B
> 1 4 5 J B
> 2 3 4 B B + B
> 2 3 5 B B
> 2 4 5 B B
>---------------------------------------------------------
> #J's=2 #J's=2
> #B's=12 #B's=12
I just denoted with an asterisk the tickets that would win a
higher prize for one of the 2 sets. You can easily see that
60% of the time, both sets have the same payouts. In 20% of
the cases a particular set will have a payout slightly smaller
than the other set and in the last 20%, the set will have a
payout slightly larger than the other set.
It's a bird, it's plane ... No, no! It's a tie! :-)
Neil Readwin
:Tell me what the odds are of a person choosing their same numbers for 1
:1/2 years and then failing to buy that ticket. Low and behold the day he
:forgets to play his tickets, he gets a match. IT'S NOT THE ODDS THAT
:DETERMINE THE NEXT WINNER, IT'S SOMETHING FAR BIGGER THAN THAT.
Are you trying to tell us you're a religious man, Ken?
Ron Meisenheimer
>Ron (& anyone else who has been following this thread):
>
>I agree that there is no proven method for artificially increasing your
>probability of matching numbers drawn (as James Sarnelli promises). The
>large jackpot games such as powerball and others attract people with the
>possibility that they can win millions. Although they do deliver to a few
>fortunate individuals every once in a while, the expected gain is simply too
>low to bother. If you're going to throw money away, throw it someplace where
>it at least might do some good (rather than lining government coffers).
>
>However, if you must play the lottery, do your homework. You can
>substantially increase your expected gain by playing games that offer lower
>prizes but better expected returns. Also, some of these games are fixed
>payout games whereby winners do not share a prize pool. Games such as
>5-minute Keno aren't too bad (beware the replayability trap though).
>
and others in this group are trying to do is to emphasize to those
individuals who are contemplating the purchase of a lotto system, that all
lotto strategies are equivalent for a given game. No need to pay someone
else for his strategy. Yours, whatever it is, is just as good (it's better
to say just as bad, but that's another thread).
Never pay someone else for his method of number selection!
Regards,
Ron Meisenheimer (ro...@ns.net)
Mike H.
On 12 Jun 1995 23:39:01 GMT, in <3rij6l$7...@ixnews5.ix.netcom.com>
jam...@ix.netcom.com (JAMES SARNELLI) wrote.....
> same is true of the lottery. In the next few days I will begin posting
> some logical methods of lottery/lotto play. Try out the methods and
> confirm for yourself that playing systematically increases your
> probability of matching numbers drawn.
>
> MORE TO FOLLOW
A strange sense of Deja Vu envelopes me. Am I entering the Sarnelli Zone ??
----------------------------------------------------------------------------
Mike H. (Software Sans Frontieres) PGP:818C97EB75366540 8A27D2AB8E7482CB
Umail 1.50 from ftp.demon.co.uk:/pub/ibmpc/umail,/pub/ibmpc/windows/umail
----------------------------------------------------------------------------
Normand Veilleux
>From: sha...@vm.cc.purdue.edu
>
>>cma...@cc.ic.ac.uk (Thomas Sippel - Dau) writes:
>>
>>No, tendentially you increase it, but not monotonically so (see above).
>
>
Me too please! :-)
>Some of us ain't as swift as we'd like to think, but I'm sure willing to
>listen and learn - unlike a certain poster on LOTTO LOGIC.
>
>Actually, about ten minutes of thought just went by and I do believe that
>Thomas is right. But, God!, I'd like to see this put into simpler terms.
>Somewhere at about college freshman (humanities) level would be great.
Actually, it's easy to see that not all sets of tickets have the same
odds of winning. Just take the simple example that has been discussed
recently:
Winning Numbers You Win I Win
---------------------------------------------------------
1 2 3 J J + B
1 2 4 B + B J + B
1 2 5 B + B B + B
1 3 4 B + B B + B
1 3 5 B + B B
1 4 5 J B
2 3 4 B B + B
2 3 5 B B
2 4 5 B B
3 4 5 B -
---------------------------------------------------------
#J's=2 #J's=2
#B's=12 #B's=12
The "I win" column represented the set {123, 124} and it had 90%
chance of winning a prize. Whereas the "You Win" column represented
the set {123, 145} and had 100% chance of winning a prize.
Since 90 is not the same as 100, it shows that 2 different sets
don't always have the same odds. Actually, you could have a set
of three tickets that only has 90% chance of winning. Take for
example {123, 124, 125}.
Normand Veilleux
>From: jac...@xmission.com
>
>>Thomas Doehne) writes:
>>
>> Not necessarily so. First, we have to assume that the lotto
>> player is trying to minimize risk.
>
>that the lotto player is trying to minimize risk, then the conclusion
>we reach is that the lotto player won't buy a ticket, since this
>minimizes the risk by reducing it to zero.
Not bad! But, why do you always have to be so logical!? :->
Maybe that's what James can't stand! I guess, it's a very
annoying habit! :-)
Daniel Efron
DE> If the lottery behaves
DE> as random, indepedent events, buying all N combinations of one draw is
DE> statistically the same as buying N tickets over many draws as far as
DE> winning tickets are concerned.
>
> **** bzzzt, no. Assume 6 from 49 (13,983,816 possibilities)
>
> If you buy all combinations, you are assured of matching 6. If you buy
> 1 ticket at random in 13983816 draws, the probability of matching 6
> at least once is the complement of the probability of not matching 6
> in any of the draws, i.e. the complement of:
>
> (13983815/13983816) ** 13983816 = 0.36787943 >=< 1/e
>
> or about 63.212%
I think we definitely have a language problem here. First, "bzzzt,
no" would be rude in the US, but might not be in the UK. Second,
you're talking about the probability of winning once.
Sure, the probability of winning once is greater by buying the entire
combination range of the lottery. But your probability of winning more
than once becomes zero.
The key is that 63.212% of the time you win once *or more* (as you said).
If you weight all of the probabilities of winning n times by multiplying
by n, you'll get the expected number of wins buying one ticket at a time
13983816 times. (n = number of wins, p(n) = probability of getting that
many wins in 13983816 individual draws playing one ticket at a time, of
course this is for a 6/49 lottery)
n p(n) p(n) * n
- ---------- -------------
0 0.36787943 0
1 0.36787945 0.36787945
2 0.18393973 0.36787945
3 0.06131324 0.18393971
4 0.01532831 0.06131323
5 0.00306566 0.0153283
6 0.00051094 0.00306566
7 7.2992E-05 0.00051094
8 9.124E-06 7.2992E-05
9 1.0138E-06 9.124E-06
10 1.0138E-07 1.0138E-06
11 9.2161E-09 1.0138E-07
12 7.6801E-10 9.2161E-09
13 5.9078E-11 7.6801E-10
14 4.2198E-12 5.9078E-11
15 2.8132E-13 4.2198E-12
[rest of the table is not consequential]
As we would expect the total for p(n) is 1. As we would also expect
the total for the weighted p(n) is also 1.
When I said, as N (the number of tickets purchased) gets large, I meant
when you buy many, many times as many tickets over the course of many,
many draws as it would take to purchase all of the tickets in one
entire draw. Eventually, the number of winners from both methods will
be essentially the same.
As far as one draw goes, buying the entire set of numbers does guarantee
a victory. Buying one ticket N times (where N is what it would take
to buy the entire set) for N draws will not guarantee a winner, but it
will frequently yield more than one winner (about 26% of the time), enough
to make up the difference.
> - Buying all N combinations of the lottery does guarantee an exact set of
> - payoffs (and you will lose approximately 50% of your money), but N distinct
> - tickets in a number of draws will produce the same number of winners within
> - an expected standard deviation. As N gets larger (buying all tickets in
> - many drawings, for instance), the number of prizes won from both methods
> - will continue to converge.
>
> Nope, the expectation (i.e. the integral of payout over probability) is
> constant at 50% or whatever the lottery pays.
Again, I guess in the UK this would not be considered rude. I really
don't know what you're saying "nope" to here. You must think that I'm
implying that playing smaller sets of numbers over a larger number of
draws will yield more than 50% return. I am not.
Again, I'll say what I always say. If the lottery behaves as random,
independent variables, nothing you can do will either increase or
decrease your chances of winning a ticket except for buying more tickets.
No modern US lottery has ever been shown to be usefully non-random except
by illicit tampering.
--
Dan Efron
Tony Grange
>
>
>
> I see many people in this group looking for a "system" to win the
>lottery. Are they serious? This is not a flame, but a serious question.
>Since the drawing is a random drawing what system could exist? Belief that
>some system exists seems illogical to me. So I guess my question is:
>
>
Work yourself an advantage: include numbers in your picks above 31.
You're less likely to have to split your millions with others, since
many folks base their "lucky numbers" on birthdays. God knows you wouldn't
want to have to share $30,000,000 with anyone else.
Tony Grange
>
>
>>From: doe...@church.cse.ogi.edu (Thomas Doehne)
>
>[A winning system]
>>
>>I end up not playing the lottery very often -- usually the
>>pots are too low. Generally, the pot has to be above about
>>17 million for a positive expected value (for the Michigan
>>lotto).
>
>
>I really wonder why people are so paranoid about splitting the
>jackpot with someone else. I mean, wouldn't you prefer splitting
>a $3 million dollar jackpot 3 ways rather than not winning at
>all? You wouldn't have to twist my arm! I'd take the split
>any day! :-)
>
----
It's interesting, however, to actually see a pseudo-rational reason
for only buying when the jackpot gets big. It's always seemed so
bizarre to reason that $3 million wasn't enough to win (wait until
it's $60 million); it's still bizarre, but based on a specified system,
rather than "intuition". The system is still screwed - expected value
really means nothing when you are dealing with such long odds and so few
trials.
Tony
John Griffin
David Carr <dc...@its-menz.cc.monash.edu.au> wrote:
>Be careful with the notion that numbers above 31 are necessarily good.
>
>Recent evidence in a draw in Australia showed that the Division 5
>payout (3 numbers + supplementary number) was as high as you would get.
I'm an Australia lottery rookie. Please tell me what that means!
>similarly the payout for matching 4 numbers. The 5 number payout was
>almost average, while the 5+supplementary number and the payout for 6
>numbers wreway below average.
>
>In other words on an individual basis, these numbers are the best, but in
>combination, not necessarily.
Let me guess what you mean here: If you include 31 on your ticket, then
the probability of of drawing each of the higher numbers is decreased?
It almost seems that you were talking about sums, but that meaningless
artifact has been so thoroughly debunked that it's rarely mentioned now.
>
>The single factor that contributes most to the size of the payout (and
>here I am concentrating on payouts for 5 or 6 numbers matching---when I
>win big, I want to sweep the poll) is the
>spread of the numbers on the form (i.e. if you take 6 numbers, there are
>15 possible pairs, and you can calculate the Euclidean distance, or
>City-block distance of all these pairs, and take combinations with high
>average distances). There are various distance measures available.
>
>This is a purely empirical result, backed up by evidence of both the
>German and Australian lotteries. You end up taking combinations, with a
>high proportion of numbers on (or near) the perimeter of the Lotto form.
>There are always a few exceptions(i.e. favoured numbers)---somehow the
>mystical numbers invariably are odd (i.e. 3,5,7), so these may need to be
>avoided.
As I mentioned, I don't know how your lottery works, so forgive me for
asking...
What are mystical numbers?
Don't they have "quick pick" or something like that? This would save
you a hell of a lot of trouble without costing you anything.
David Carr
Recent evidence in a draw in Australia showed that the Division 5
payout (3 numbers + supplementary number) was as high as you would get.
similarly the payout for matching 4 numbers. The 5 number payout was
almost average, while the 5+supplementary number and the payout for 6
numbers wreway below average.
In other words on an individual basis, these numbers are the best, but in
combination, not necessarily.
The single factor that contributes most to the size of the payout (and
here I am concentrating on payouts for 5 or 6 numbers matching---when I
win big, I want to sweep the poll) is the
spread of the numbers on the form (i.e. if you take 6 numbers, there are
15 possible pairs, and you can calculate the Euclidean distance, or
City-block distance of all these pairs, and take combinations with high
average distances). There are various distance measures available.
This is a purely empirical result, backed up by evidence of both the
German and Australian lotteries. You end up taking combinations, with a
high proportion of numbers on (or near) the perimeter of the Lotto form.
There are always a few exceptions(i.e. favoured numbers)---somehow the
mystical numbers invariably are odd (i.e. 3,5,7), so these may need to be
avoided.
Yours,
David Carr
-----------------
Monash University
Accident Research Centre
Clayton, VIC 3168
Australia
Tel. +61/3/99054378
Fax. +61/3/99054363
email dc...@its-menz.cc.monash.edu.au
Dud: ... Here, have a sandwich. My feet are killing me.
Pete: What's that got to do with the sandwich?
Dud: Nothing, I just said it afterwards, that's all.
Pete: Well, you shouldn't say things like that together,
it could confuse a stupid person.
Thomas Doehne
(Tony Grange) writes:
>>>From: doe...@church.cse.ogi.edu (Thomas Doehne)
>>>
>>>I end up not playing the lottery very often -- usually the
>>>pots are too low. Generally, the pot has to be above about
>>>17 million for a positive expected value (for the Michigan
>>>lotto).
>It's interesting, however, to actually see a pseudo-rational reason
>for only buying when the jackpot gets big. It's always seemed so
>bizarre to reason that $3 million wasn't enough to win (wait until
>it's $60 million); it's still bizarre, but based on a specified system,
>rather than "intuition". The system is still screwed - expected value
>really means nothing when you are dealing with such long odds and so few
>trials.
Actually, one can argue that buying lottery tickets based on that
system are a reasonable part of an investment portfolio, where
the portfolio is spread between the range of safer and riskier
investments. Buying a lotto ticket is very risky as an investment,
but (if chosen properly) has a very high return. People put
some of their portfolio out on the risky end because that's
the only way to get rich investing. Lotto tickets are just
way, way out on the risky end. Mind, I'm not seriously
advocating the lottery as part of a balanced investment
portfolio.
BTW, you really should avoid the ad hominems ("pseudo-rational").
--
Tom Doehne
doe...@cse.ogi.edu
Ron Blaylock
: Makes sense. But as Kevin correctly pointed out earlier, we have
: to be careful how we define 'win.' Surely, you have not "won" if
: you spend more money than you get back. See my last post (apologies
: to the group for that last post. I didn't mean for all the earlier
: references to get cut out).
: Regards,
: Ron Meisenheimer (ro...@ns.net)
Actually I disagree with Ron's statement of "win." OF course this is
only my opinion.
Last summer I "won" first place in my bowling league. For that I got a
trophy and $400. The season was 33 weeks. I paid $13 a week which comes to
$429 dollars. Is Ron telling me I didn't win?
I spend $5,000 playing the lottery and I have a ticket which pays out
$100. In my opinion I have won the lottery. I may not have won a large
jackpot or won millions which may have been my objective, but still,
I have won.
My overall opinion is that the definition of "win" can only be determined
by the person who said it. If your definition of "win" is to obtain more
money than you paid, then you cannot possibly win until what you receive is
more than what you paid. My definition is that I "win" whenever a
lottery ticket pays me money, so I win a lot. (Very small amounts at a
time I must admit.)
Ah... Life in America where we can all have our own opinions. Isn't it
GREAT!
Ron (also)
Blaylock
Kevin Bongiovannu
>
>Terry Thero (te...@col.hp.com) wrote:
>
>
>
>This is not true at all. People beat the lottery every day. It may not
>come down to working out a method to beat a random ticket, ...
You are correct. Back in 1982, I knew a person that used "black magic" to
calculate his numbers (he was a hillbilly from W. Virginia). I witness many
times he hitting pick 3 and 4. Any person that buys 20 tickets with the
same number and hits knows how to win.
Ron Meisenheimer
>
>< SNIP a bunch ...>
>
>: Makes sense. But as Kevin correctly pointed out earlier, we have
>: to be careful how we define 'win.' Surely, you have not "won" if
>: you spend more money than you get back. See my last post (apologies
>: to the group for that last post. I didn't mean for all the earlier
>: references to get cut out).
>
>: Regards,
>
>: Ron Meisenheimer (ro...@ns.net)
>
>Actually I disagree with Ron's statement of "win." OF course this is
>only my opinion.
>
>Last summer I "won" first place in my bowling league. For that I got a
>trophy and $400. The season was 33 weeks. I paid $13 a week which comes to
>$429 dollars. Is Ron telling me I didn't win?
>
Congratulations. May you "win" a million such prizes!
>I spend $5,000 playing the lottery and I have a ticket which pays out
>$100. In my opinion I have won the lottery. I may not have won a large
>jackpot or won millions which may have been my objective, but still,
>I have won.
>
>My overall opinion is that the definition of "win" can only be determined
>by the person who said it. If your definition of "win" is to obtain more
>money than you paid, then you cannot possibly win until what you receive is
>more than what you paid. My definition is that I "win" whenever a
>lottery ticket pays me money, so I win a lot. (Very small amounts at a
>time I must admit.)
>
>Ah... Life in America where we can all have our own opinions. Isn't it
>GREAT!
>
Your opinions, particularly on mathematics, are, well . . . BwaHAHAHAHAHAHA!
Regards,
Ron Meisenheimer (ro...@ns.nset)
Normand Veilleux
>From: blay...@ashur.arc.nasa.gov (Ron Blaylock)
>
>< SNIP a bunch ...>
>
>: Makes sense. But as Kevin correctly pointed out earlier, we have
>: to be careful how we define 'win.' Surely, you have not "won" if
>: you spend more money than you get back. See my last post (apologies
>: to the group for that last post. I didn't mean for all the earlier
>: references to get cut out).
>
>only my opinion.
>
>Last summer I "won" first place in my bowling league. For that I got a
>trophy and $400. The season was 33 weeks. I paid $13 a week which comes to
>$429 dollars. Is Ron telling me I didn't win?
On a similar note, here are two posts I sent to rec.gambling in early
December last year:
From: nvei...@emr1.emr.ca (Normand Veilleux)
Subject: Re: RE: Are you playing LOTTO?
Date: Mon, 26 Dec 1994 19:15:15 UNDEFINED
>From: Cli...@cityscape.co.uk (Clique)
>>Norman Veilleux says
>>>From: ETM...@prodigy.com (Garry Kyle)
>>>
>>>G'day Mate
>>>I am of the belief that in a random choice drawing such as most lottos
>>>there is no way to improve odds. Buying more than one ticket does not
>>>increase odds of winning at all.
>>
>>I think I understand what you are trying to say: If you don't
>>buy any tickets, the only thing that can happen is that you
>>won't win. So, that gives you 0% probability of winning ...
>>
>> ... Wow, you're right! We will have to revise all the books of
>>probability in the whole universe! :->
>>
>Hey, lay off poor old Garry. In a sort of way he was right even if he
>didn't know he was right. Buy one ticket and you *might* win. So some
>sort of odds exist for winning. Buy 100,000,000 tickets and you
>certainly lose (i.e. lose overall) as your outlay will exceed the pool
>available to the winners. So there's a certainty of loss. Somewhere in
>between those two points your possibility of winning turns into a
>certainty of losing and presumably the line follows some sort of
>regular curve.
>
>So, yes, buy two tickets rather than one and you decrease the chances of
>winning as you move along the line towards a certainty of losing. I
>think.
Must I point out that the original comment was "Buying more than
one ticket does not increase odds of winning..."
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This is false. Assume a certain lottery has a grand total of 10
tickets and that only one of those tickets will win the only prize
available. If I buy 1 ticket, I have 1 chance in 10 of winning or
10%. Those are my odds. If I buy 2 tickets, then I have 2 chances
in 10 of winning or 20%. My ODDS are now bigger. And my ODDS keep
increasing with every ticket I buy until I reach 100% probability of
winning a prize.
"Winning" is a key word here. We seem to be using a different
definition. From my point of view, the winnings is the value
of the tickets after the draw. That can be 0, but it cannot be
negative. On the other hand, you associate winning with making
a profit. That is a totally different ball game.
My mother played the Lotto 6/49 the other day. She won! $10
dollars that is. She had bought 20 tickets and so had spent
$20. If you were to ask her, "how much did you win at the
lottery the other day?" She would definitely answer, "I won
$10" not "I lost $10."
I went to Las Vegas once, about five years ago, for a full week.
I was attending the Comdex show. Not being a gambler I spent
only 20 dollars total in slot machines. I won some money, but
I put it back in the machines and I never got a win big enough
to cover my costs, so I ended with a $20 loss. The only reason
I'm mentioning this is that while in Vegas I saw this big slot
machine that had a 99% probability of winning and that in the
unlikely event that you did not win, your money would be refunded.
I couldn't believe my eyes! Either I win or I get my money
back: it's impossible to lose, I thought. I got closer to the
machine and saw that it took dollar coins, then I looked at the
payout chart. Three cherries made you win 25 cents! OK, you
win, but with winnings like that you can't make a profit!
That is the same logic you apply to lotteries and that's fine
with me. You can WIN at the lottery, but you won't necessarily
make a PROFIT. The ODDS don't care if you make a profit or not,
they represent the amount of chances you have of getting back
some money. The difference between your cost and your winnings
is your profit (or loss). This has nothing to do with the ODDS
since you could "happen" to make the same profit by playing 2
different games of chance with different ODDS of winning.
Now, just to check you have read this far, I'm willing to bet
you $5 that, if you send me $20 through the mail, I will send
you $100 through the mail. Before you send anything, think
about it: who will make the profit ? You or me ?
HINT: Remember, I'm not a gambler!
From: nvei...@emr1.emr.ca (Normand Veilleux)
Subject: Re: Are you playing LOTTO? Starting a Newsgroup
Date: Mon, 26 Dec 1994 20:07:49 UNDEFINED
>From: ETM...@prodigy.com (Garry Kyle)
>if you buy every possible combination of numbers except one set,and I
>purchase the only remaining set of six numbers, and a machine randomly
>picks the six winning numbers do I have less chance of winning than you?
>Random "chance" is the key word here.
>Any comments are welcomed and appreciated.
Yes, I do. For simplicity, let's assume the lottery we play only
has 10 tickets and that only one of the tickets will win the only
prize available. Same example as my last post.
Now, according to your above comment I would buy 9 tickets and you
would buy 1 ticket. Assuming all tickets have an equal chance of
winning (this is the case in a random draw), then you have 10%
chances of winning and I have 90% chances. The Gods ... I mean
the odds are against you. Since 90% is bigger than 10%, I have
more chances of winning then you.
I'll stress this again, you are asking about the "chances of
winning" not about the "chances of making a profit." Some
people will complain and say that I am forgeting about the
costs of buying the tickets. I'm not forgeting about it, it
is irrelevent to the ODDS of winnings. Whether my 9 tickets
cost $9, or $90 or $9,000 I still have 90% chances of winning
the prize.
BTW, if it was my birthday and my mum bought me a ticket and
my dad bought me a ticket and my uncle bought me a ticket ...
etc., until 9 people buy me a ticket, then my nine tickets
would cost me nothing and guess what ? I would still have the
same ODDS of winning. In this case though, my winnings would
be equal to my profit.
If you want to be sure to make a profit, don't gamble! By
definition gambling means taking the risk of losing everything
you wager, for the possibility of winning something back!
The ODDS of winning merely quantify this "possibility".
sha...@vm.cc.purdue.edu
nvei...@emr1.emr.ca (Normand Veilleux) writes:
<lots deleted>
>
>"Winning" is a key word here. We seem to be using a different
>definition. From my point of view, the winnings is the value
>of the tickets after the draw. That can be 0, but it cannot be
>negative. On the other hand, you associate winning with making
>a profit. That is a totally different ball game.
>
>My mother played the Lotto 6/49 the other day. She won! $10
>dollars that is. She had bought 20 tickets and so had spent
>$20. If you were to ask her, "how much did you win at the
>lottery the other day?" She would definitely answer, "I won
>$10" not "I lost $10."
>
a similar understanding, but just want to verify.
Now I agree that people will tend to focus/state how much *return* they got,
and not how much they *invested* - especially in 'all-or-nothing' type
games, like lotteries, bingo, keno, etc.
But let's say I'm playing a on-going game - like blackjack. I sit down at a
$5 table with $50 in chips. I play for 4 hours, 200 hands, and win 97 hands,
losing 103. So I leave the table with $20. Do I now say;
"I won $20" ? "I won $485" ? or do I say "I lost $30" ?
Is this any different from a person that spends $50 on lottery tickets
and 'wins' $20 in prizes?
But maybe this is what Lon &/or James mean when they say they have a
'winning' system for lottos.
>Now, just to check you have read this far, I'm willing to bet
>you $5 that, if you send me $20 through the mail, I will send
>you $100 through the mail. Before you send anything, think
>about it: who will make the profit ? You or me ?
>HINT: Remember, I'm not a gambler!
>
<more deleted>
>
>I'll stress this again, you are asking about the "chances of
>winning" not about the "chances of making a profit." Some
>people will complain and say that I am forgeting about the
>costs of buying the tickets. I'm not forgeting about it, it
>is irrelevent to the ODDS of winnings. Whether my 9 tickets
>cost $9, or $90 or $9,000 I still have 90% chances of winning
>the prize.
'beat' the lottery. If it's $50, then the first scenario is a winner,
the second and third are losers. And since (most) lotteries return ~50%
the second scenario is most likely. 9 times out of 10 - lose $40.
1 time out of 10 - lose $90. Total loss in 10 games - $450.
And this is, of course, the simple 'buy more tickets' that all lotto
systems end up relying on.
Mike Sharkey
Normand Veilleux
>From: sha...@vm.cc.purdue.edu
>
>>nvei...@emr1.emr.ca (Normand Veilleux) writes:
>>
>>"Winning" is a key word here. We seem to be using a different
>>definition. From my point of view, the winnings is the value
>>of the tickets after the draw. That can be 0, but it cannot be
>>negative. On the other hand, you associate winning with making
>>a profit. That is a totally different ball game.
>>
>>My mother played the Lotto 6/49 the other day. She won! $10
>>dollars that is. She had bought 20 tickets and so had spent
>>$20. If you were to ask her, "how much did you win at the
>>lottery the other day?" She would definitely answer, "I won
>>$10" not "I lost $10."
>>
>
>a similar understanding, but just want to verify.
>Now I agree that people will tend to focus/state how much *return* they got,
>and not how much they *invested* - especially in 'all-or-nothing' type
>games, like lotteries, bingo, keno, etc.
>But let's say I'm playing a on-going game - like blackjack. I sit down at a
>$5 table with $50 in chips. I play for 4 hours, 200 hands, and win 97 hands,
>losing 103. So I leave the table with $20. Do I now say;
>"I won $20" ? "I won $485" ? or do I say "I lost $30" ?
We do have the same understanding. In my book, you lost $30. This
is from a "profit" oriented terminology.
>Is this any different from a person that spends $50 on lottery tickets
>and 'wins' $20 in prizes?
Yes, it's different. Unfortunately, lotteries or draws in general
have "winners." Whether those "winners" are actually making a profit
is irrelevant; they are "winners" of the contest. I know it's hard
for a mathematically inclined person to accept that, but consider
that you are going against a tradition of hundreds of years of "and
the winner is..." when you claim there are no winners in a draw.
To avoid contradicting people by saying to them, "but, you haven't
won" when they just *know* they were declared "winners", I prefer
letting them be winners, but explaining that even though they won
the lottery it cost them more money than they won, so overall it
was not a profitable transaction.
This reminds me of the two fishermen that started a business of
buying fish at $2.00 a pound in one city and reselling it at
$1.95 a pound in another city. Eventually one the guys told his
partner, "I really don't see how we will ever make a profit by
buying at $2.00 a pound and selling at $1.95." His partner
pondered his statement for a while and replied, "You're right. I
think we need a bigger boat!"
So, to apply this to Lotto. If you lose a little on every ticket
you buy, then buy more tickets! You'll "win" more often! :->
My winning system is to buy 174 tickets every draw. I win every
time! What I don't understand though, is why my bank account
balance keeps going down! Maybe I should buy 348 tickets! :->
>>Now, just to check you have read this far, I'm willing to bet
>>you $5 that, if you send me $20 through the mail, I will send
>>you $100 through the mail. Before you send anything, think
>>about it: who will make the profit ? You or me ?
>>HINT: Remember, I'm not a gambler!
>>
>Nah, I'll keep my $15, thank you!
I meant to remove this part of the post since it really applied
only to the original person to whom I replied back in Dec. '94.
Edward Croft
Newsgroups: rec.gambling.lottery
Subject: Re: Method to beat the lottery
Date: 15 Jun 1995 01:25:50 GMT
Organization: NSNet
Message-ID: <3ro26u$7...@falcon.ns.net>
References: <3r9tej$h00...@bajoriap.perkin-elmer.com> <3rij6l$7...@ixnews5.ix.netcom.com> <3rj1o9$q...@falcon.ns.net> <3rk9q1$3pm@vix=
en.cso.uiuc.edu> <3rks <3rmk7u$h...@gateway.gtech.com>
JAMES SARNELLI
>
>
>In article <3rmk7u$h...@gateway.gtech.com>, psul...@gtech.com (Peter
Sullivan) says:
>>Ron (& anyone else who has been following this thread):
>>
>>I agree that there is no proven method for artificially increasing
your
>>probability of matching numbers drawn (as James Sarnelli promises).
lower
>>
>
>Regards,
>
>Ron Meisenheimer (ro...@ns.net)
>
>>>>RON: We are in agreement. I do not believe that you can
artificially increase your probability of winning a prize. However, I
do believe that you can natually increase your probability of winning a
prize relative to other players, by taking the time to study and
understand the way that the lottery/lotto system of number generation
works. It is unfortunate that the math weenie/guru contingent in this
group have less confidence in the theory and laws of probability and
statistics to have practical application than I do. Einstein spent his
life seeking an underlying law that would tie all natural law together.
In spite of his great achievements he never stopped believing that
there was more to discover. Perhaps, my critics would like to answer
these questions, or at least consider them:
1. Is there any such thing as a perfect random number generator?
2. If such a machine could be discovered, would it defy the laws
of physics and continue to generate perfectly random numbers, with no
evidence of mechanical bias?
3. Is there any such thing as bias and is there then anyway to
evaluate it?
4. Would such manifestation of bias be any different than the
mechanical bias of a roulette wheel which was discussed in "BEAT THE
WHEEL"?
5. Rather than remain locked in at the probabilities of matching
one or more numbers on a given ticket, wouldn't it make more sense to
study the theory and actuality of the random number generation process
itself, so as to compare expectations with results?
>>>>>>>Don't these subjects at least warrant discussion? The Bible
states: "There is nothing new under the Sun" but it is also true that
at any time in history man only knows a small fraction of what can be
known ünder the Sun".
MORE TO FOLLOW
JAM...@AOL.COM
jac...@xmission.com
>I do not believe that you can
>artificially increase your probability of winning a prize. However, I
>do believe that you can natually increase your probability of winning a
>prize relative to other players, by taking the time to study and
>understand the way that the lottery/lotto system of number generation
>works.
Then you claim that you are detecting significant degrees of bias
in lotteries? I defy you to show that any statistically significant
bias exists. Of course, you were challenged to show this before and
you never did, so I know you won't/can't do it now.
>It is unfortunate that the math weenie/guru contingent in this
>group have less confidence in the theory and laws of probability and
>statistics to have practical application than I do.
On the contrary, we have great confidence in what probability theory
tells us about the lottery. In fact, we've shown that YOUR OWN PICKS
fell right in line with what you would expect from probability theory,
yet you had deluded yourself into thinking that you had made some huge
discovery because of your own mathematical ineptitude. Now THAT was
incredibly funny, and I'm sure you'll treat us to the same humor in
the months to come.
>Einstein spent his
>life seeking an underlying law that would tie all natural law together.
Two points:
1) He never found it, and nobody else has either.
2) You're no Einstein.
> In spite of his great achievements he never stopped believing that
> there was more to discover.
Nobody has claimed that there is "nothing left to discover" about
the lottery. That doesn't imply that the game is beatable. That
also doesn't imply that YOU have discovered anything of any real
value.
>Perhaps, my critics would like to answer
>these questions, or at least consider them:
Sure, I'll take a stab at them. I'll also take this opportunity to
point out that you rarely (if ever) bother to answer any of the serious
questions that we pose to you.
>
> 1. Is there any such thing as a perfect random number generator?
Take your pick:
1) Yes, such things probably exist at the quantum level. The
Heisenberg Uncertainty Principle has a lot to say about the
fundamental tradeoffs in what you can know about a given
physical system.
2) No, there are no PERFECT random number generators. However,
probability theory can be applied just as easily to imperfect
number generators, so this fact won't help you to invalidate
everything we've been telling you for months. Sorry, this
won't allow you to ignore mathematics for the sake of your
silly morphing nonsense.
You've never shown (or even claimed) that any actual lottery is
SIGNIFICANTLY non-random. Also, the presence of a small amount of
non-randomness does NOT imply that you can increase your expectation
significantly. "Slightly non-random" in no way implies "you can
predict this thing exactly if you just watch it long enough".
> 2. If such a machine could be discovered, would it defy the laws
> of physics and continue to generate perfectly random numbers, with no
> evidence of mechanical bias?
How would this defy any laws of physics? I'm not aware of any law
of physics which states "there can be no perfect random number
generators". Gosh, are you going to embarrass yourself in front
of physicists now?
Physics question for James: (careful James, it is a trick question)
1) If you know the exact position of a particle and it's momentum
(speed and direction of motion), can you predict exactly where
that particle will be at all times in the future?
bonus question:
2) If you know EXACTLY where something is, what can you say about
where it is going?
for extra credit:
3) Use the Heisenberg Uncertainty Principle to design a random
number generator, and talk about how you could adjust your
system to increase or decrease the randomness involved.
> 3. Is there any such thing as bias and is there then anyway to
> evaluate it?
Certainly bias can exist in any physical system. The fact the bias
CAN exist does not imply that bias DOES exist. This also doesn't
imply that the bias (if it exists) is great enough that it can be
exploited to any significant extent.
> 4. Would such manifestation of bias be any different than the
> mechanical bias of a roulette wheel which was discussed in "BEAT THE
> WHEEL"?
Yes. When you play roulette, you get to sit there and watch the
wheel for every play. If someone comes and adjusts the wheel or
replaces it with a new one, you will know about it. Lotteries use
several different sets of equipment, and they won't tell you which
one they are using until they've stopped selling tickets.
>5. Rather than remain locked in at the probabilities of matching
>one or more numbers on a given ticket,
Translation: "I've already proved that I can't deal with all of that..."
>wouldn't it make more sense to
>study the theory and actuality of the random number generation process
>itself, so as to compare expectations with results?
OK James, what is the theory of random number generation for
the lottery? What is the actuality? What are the EXPECTED results?
What are the ACTUAL results? How do the expected results compare
with the actual results? Do they indicate that bias exists? If so,
how much bias is there?
You haven't given us any confidence that you are mathematically
equiped to actually perform such an evaluation. In fact, you've proved
repeatedly that you are completely unqualified to do so. Are you
claiming that this is what you've done? Seems to me that you've
simply ASSUMED such bias must exist, and you've further ASSUMED that
your methods are actually useful in exploiting such biases.
>>>>>>>>Don't these subjects at least warrant discussion? The Bible
>states: "There is nothing new under the Sun" but it is also true that
>at any time in history man only knows a small fraction of what can be
>known ünder the Sun".
Sure James, they warrant discussion. Tell us the extent of the biases
that you've discovered and measured. Show us the analysis you have
preformed to compute the amount of bias present. We're ready and
waiting to examine your results.
Prediction: James hasn't done any such analysis, and couldn't prove
that the lottery is biased if his life depended on it.
Note to newbies: I may seem harsh on James, but it is because he
has wasted countless hours of time supposedly "presenting" his system,
when in fact he has never presented a damned thing that wasn't just
a load of nonsense. I've posed MANY MANY serious questions to James
in order to have a discussion of this topic, and he has uniformly
ignored every single question that I posted. I can guarantee you
several things if you try to discuss this topic with James:
1) If you disagree with him, he will start calling you names.
Yet, he will squeal like a stuck pig if any ad-hominen attacks
should be directed towards him.
2) If you disagree with him, he will say that you don't know what
you are talking about. He will claim that he understands subjects
such as probability theory, while his comments indicate clearly
that he is completely clueless on the topic.
3) If you disagree with him, he will claim that you do not
understand simple English. Yet, instead of using simple
English himself, he will try to obscure what he is saying
in phrases like "randomness morphs into predictability".
4) If you ask him serious questions about his methods, he will
label your post as "cyber-pollution". This is part of his
ongoing ploy to avoid answering serious questions from those
who are capable of showing that he doesn't know what he is
talking about.
5) If you ask him to provide evidence for his claims, he will
ignore you.
6) If you prove that he is wrong, he will never admit it or
retract a single word. He will however whine that you
are being "unfair" to him, and say that you are "attacking"
him.
7) James will embarrass himself repeatedly, yet he won't ever
realize what an embarrassment he is to himself.
Just don't kid yourself into thinking that you'll actually be able
to DISCUSS anything with James. He isn't here to explore ideas, he
is here to deliver the Divine Inspired Truth of the Great James
Sarnelli to anyone with an "open mind" (where "open mind" means
"if you don't agree with James, you don't have one"). Phrases such
are "you have a point", "I was wrong about that", "that was an
error in my calculations" simply are not in the James Sarnelli
Dictionary of Simple English, so don't ever expect to see them.
In short, James Sarnelli was a weasel in a previous life, and he
hasn't changed. Talking to him is a huge waste of time. It can
be rather amusing, if you like to see someone weasel their way out
of admitting their mistakes, but it won't do ANYTHING to increase
your understanding of the lottery.
James, if you want to present your 8 mystical and magical ways
to improve your odds at the lottery (complete with many new
special effects), I'd strongly suggest that you just write it
all up in advance and post it ONCE. That way, it might actually
get posted, rather than another one of your endless series of
posts that all say "MORE TO FOLLOW" without ever actually saying
what your silly methods are.
Anyone else notice that James said WEEKS AGO that he was going to
post his methods and he still hasn't done it? He's posted several
times, yet he still hasn't said anything about his methods.
James, I challenge you to simply list your 8 methods with a brief
description of how they are used. Nothing major, just a paragraph
or two on each method. Put it all in one simple post. Prediction:
James can't/won't do this, as it would just be too simple and
sensible, and it would interfere with the one phrase that gives
meaning to James pitiful cyber existence: "MORE TO FOLLOW".
Kurtis Araki
If you happened to just skip through the last artice, go back...
It's the best flame I've ever seen in rec.gambling.*
-ksa
Edmund C. Hack
JAMES SARNELLI <jam...@ix.netcom.com> wrote:
[stuff deleted]
> Einstein spent his
>life seeking an underlying law that would tie all natural law together.
>there was more to discover.
He spent a lot of this time trying to disprove Quantum Mechanics, (cf
"God does not play dice with the universe.") He also never found the
Unified Field Theory, either.
> Perhaps, my critics would like to answer
>these questions, or at least consider them:
>
> 1. Is there any such thing as a perfect random number generator?
> 2. If such a machine could be discovered, would it defy the laws
>of physics and continue to generate perfectly random numbers, with no
>evidence of mechanical bias?
I'm pretty sure, remembering back to my original career as a Quantum
Mechanic, that a randome number generator can be constructed out of a
timer and a detector of radioactive decay products. The output of a
radioactive source can be considered a random process (Poisson process, as
I recall), with known statistics.
There have been hints around that such devices have been used to generate
random sequences for crypto work in the past. Such devices may require a
steady supply of cats to operate. :-)
On a more practical note, it is not necessary that a mechanical random
number generator be _perfect_, just _good enough_. Good enough so that
the biases cannot be detected within the lifetime of the machine. For
example, some crypto techniques are secure enough for routine data (takes
weeks to crack, by which time the data are useless) and some are not
(crackable in minutes or hours).
virgil adumitroaie
<jac...@xmission.com> wrote:
>In article <3t6272$k...@ixnews3.ix.netcom.com> jam...@ix.netcom.com (JAMES SARNELLI) writes:
Steve,
Although I agree on the final conclusion on the lottery predictability,
allow me to disagree on the means to that conclusion. Is it right to apply
the Heisenberg principle to macroscopic objects? As I understand, all lotteries
are performed in day-to-day environments and not in some particle physics
lab :-).
If you believe in deterministic theories, you would agree that the output of
the lottery could be predictable (as a deterministic system) by knowing the
exact initial conditions of the mechanical system (the ball configuration in
the pool, initial velocities, the properties of the objects and sufaces and
all that jazz). Theoretically this is OK, but even if you know all this, there
is a big practical problem. This is a non-computable set-up, principally due
to uncertainties in the order of collision of three balls. If one will insist
in doing the simulations still, the round-off errors will bring him down
anyhow, the final state of his computation being far away from the real
situation.
On the practical side now: what would you prefer to play in a lottery game
1. a combination that was a winner in a previous draw,
2. a pertubation of the above or,
3. a totally new combination (and this one changed every draw or keeping
the same one from start to the hit - if ever)? :-)
VA
fycnrdths,'mntnthrck,whrshldb.
{Fine "presentation of non-praising material" -- Since we all know that
flaming is immature.}
|I can guarantee you several things if you try to discuss this topic with
|James:
[snip]
| 2) If you disagree with him, he will say that you don't know what
| you are talking about. He will claim that he understands subjects
| such as probability theory, while his comments indicate clearly
| that he is completely clueless on the topic.
|
2a) He will use his age, experience and hoards of sastisfied
customers as evidence that he knows what he is talking about
and you don't.
| 3) If you disagree with him, he will claim that you do not
| understand simple English. Yet, instead of using simple
| English himself, he will try to obscure what he is saying
| in phrases like "randomness morphs into predictability".
|
| 4) If you ask him serious questions about his methods, he will
| label your post as "cyber-pollution". This is part of his
| ongoing ploy to avoid answering serious questions from those
| who are capable of showing that he doesn't know what he is
| talking about.
|
| 5) If you ask him to provide evidence for his claims, he will
| ignore you.
|
5a) When you've confounded him with logic, he will quote the bible.
| 6) If you prove that he is wrong, he will never admit it or
| retract a single word. He will however whine that you
| are being "unfair" to him, and say that you are "attacking"
| him.
|
6a) He will selectively forget things that he has said and call you a
liar or accuse you of making things up.
| 7) James will embarrass himself repeatedly, yet he won't ever
| realize what an embarrassment he is to himself.
8) He will either quote your entire post, or none of it. (And on
top of that, his line breaks are hosed and his first line will
invariably start tacked onto the last line of your post.)
ray (aka Robin, or am I Mathman?)
----------
Ray DeGennaro
dege...@bmsrs.usc.edu
----------
Did you know that 'gullible' is not in Webster's Dictionary?
fycnrdths,'mntnthrck,whrshldb.
|1. a combination that was a winner in a previous draw,
|2. a pertubation of the above or,
|3. a totally new combination (and this one changed every draw or keeping
|the same one from start to the hit - if ever)? :-)
Doesn't really matter -- they all have the same probability of winning.
ray
JAMES SARNELLI
writes:
>>>>>>>>>>>I deleted all of your message with just a quick review. We
all agree that you have a good deal of education and are not modest
about speading it arround with profuse verbiage. Did it ever dawn upon
you that millions of people play the lottery/lotto everyday and really
don't give a diddle about signing up for your course in advanced
irrelevant information. Even without a degree in physics people
understand that things wear out . The lottery/lotto machines are
mechanical and exist at the observable (by physical senses) and not at
the qnantum level. Who appointed you keeper of the rec.gambling.lottery
gate, and empowered you to exclude anyone who doesn't have multiple
degrees in math and physics. I don't have to prove anything to you, nor
do I have any desire to. I intend to discuss lottery/lotto techniques
with normal people who have paid their money and have a right to access
to these discussions. Why don't you math weenie/gurus talk to each
other in your secret code, and leave us normal people to communicate in
plain old English?
MORE TO FOLLOW
JAM...@AOL.COM
A.J. Norman
JAMES SARNELLI <jam...@ix.netcom.com> wrote:
> Perhaps, my critics would like to answer these questions, or at
> least consider them:
>
> 1. Is there any such thing as a perfect random number generator?
> 2. If such a machine could be discovered, would it defy the laws
> of physics and continue to generate perfectly random numbers,
> with no evidence of mechanical bias?
If (1) is true, then the machine would have to create perfectly random
numbers indefinitely. That's what being perfect means. Which
particular "laws of physics" would such a machine violate?
> 3. Is there any such thing as bias and is there then anyway to
>evaluate it?
> 4. Would such manifestation of bias be any different than the
> mechanical bias of a roulette wheel which was discussed in "BEAT
> THE WHEEL"?
> 5. Rather than remain locked in at the probabilities of matching
> one or more numbers on a given ticket, wouldn't it make more
> sense to study the theory and actuality of the random number
> generation process itself, so as to compare expectations with
> results?
A little question for James: Do you think that lottery companies (and
casinos etc) might possibly spend a good deal of time and energy
checking for bias in their gambling operations, and that they might be
better at it than someone who has no understanding of statistics?
--
Andrew Norman, Leicester, England // n...@le.ac.uk // 04/07/95
Jedermann sein Eigner Fussball // l
JAMES SARNELLI
>
>In article <3t6272$k...@ixnews3.ix.netcom.com>,
>JAMES SARNELLI <jam...@ix.netcom.com> wrote:
>> Einstein spent his
>>life seeking an underlying law that would tie all natural law
together.
>>In spite of his great achievements he never stopped believing that
>>there was more to discover.
>
>He spent a lot of this time trying to disprove Quantum Mechanics, (cf
>"God does not play dice with the universe.") He also never found the
>Unified Field Theory, either.
>
>>these questions, or at least consider them:
>>
>> 1. Is there any such thing as a perfect random number generator?
>> 2. If such a machine could be discovered, would it defy the laws
>>of physics and continue to generate perfectly random numbers, with no
>>evidence of mechanical bias?
>
>Mechanic, that a randome number generator can be constructed out of a
>timer and a detector of radioactive decay products. The output of a
>radioactive source can be considered a random process (Poisson
process, as
>I recall), with known statistics.
>
>There have been hints around that such devices have been used to
generate
>random sequences for crypto work in the past. Such devices may require
a
>steady supply of cats to operate. :-)
>
>On a more practical note, it is not necessary that a mechanical random
>number generator be _perfect_, just _good enough_. Good enough so that
>the biases cannot be detected within the lifetime of the machine. For
>example, some crypto techniques are secure enough for routine data
(takes
>weeks to crack, by which time the data are useless) and some are not
>(crackable in minutes or hours).
>
>
>>>>>>The only flaw in your brilliant presentation is that we are not
talking about a quantum physics RNG, BUT ABOUT A CHEESEY MACHINE THAT
USES SOME SORT OF VACUUM SYSTEM TO SPIT NUMBERED PIMG PONG BALLS OUT A
PLASTIC TUBE ONE AT A TIME. Are you guys so over educated that it is
impossible for a normal person to communicate with you without opening
up the floodgates of irrelevant verbiage. Are we now clear about the
kind of RNG WE ARE TALKING ABOUT? The name of this group should give
you a hint. Why not turn on the boob tube and watch the lotto/lottery
drawings and you will find what normal people mean when they talk about
a random number generator.
MORE TO FOLLOW
JAM...@AOL.COM
Olaf Vancura
>In article <3t9els$g...@azure.acsu.buffalo.edu>, adu...@eng.buffalo.edu wrote:
>
>|1. a combination that was a winner in a previous draw,
>|2. a pertubation of the above or,
>|3. a totally new combination (and this one changed every draw or keeping
>|the same one from start to the hit - if ever)? :-)
>
>Doesn't really matter -- they all have the same probability of winning.
>
Well, actually it matters if the numbers game is parimutuel or not. If
so, then winning tickets share the prize pool, so you'd want to pick
obscure numbers. It has been shown in Lotto games that previous winning
tickets (and also numbers that haven't come up in a long time -- for
the contrarians ;>) are exceedingly popular and likely to be picked.
So out of the 3 choices above, the best bet is to go with #3 in a
parimutuel...
Cheers,
Olaf
--------------------------------------------------------------------------
Olaf Vancura | "Yeah, we got their attention."
van...@cfa.harvard.edu |
Harvard-Smithsonian CfA | -- Bobby Baldwin, Pres. & CEO Mirage
60 Garden St., Cambridge, MA 02138 | regarding Caesars Palace.
--------------------------------------------------------------------------
Ron Meisenheimer
>In <3srgji$j...@mailer.york.ac.uk> Edward Croft <croft> writes:
>>
>>
>>In article <3rmk7u$h...@gateway.gtech.com>, psul...@gtech.com (Peter
>Sullivan) says:
>>>Ron (& anyone else who has been following this thread):
>>>
>>>I agree that there is no proven method for artificially increasing
>your
>>>probability of matching numbers drawn (as James Sarnelli promises).
>lower
>>>
>>
>>Regards,
>>
>>Ron Meisenheimer (ro...@ns.net)
>>
In <3srgji$j...@mailer.york.ac.uk> Edward Croft <croft> writes:
>
>
>In article <3rmk7u$h...@gateway.gtech.com>, psul...@gtech.com (Peter
>Sullivan) says:
>>Ron (& anyone else who has been following this thread):
>>
>>I agree that there is no proven method for artificially increasing
>your
>>probability of matching numbers drawn (as James Sarnelli promises).
>lower
>>
>>>>RON: We are in agreement. I do not believe that you can
>artificially increase your probability of winning a prize. However, I
>do believe that you can natually increase your probability of winning a
>prize relative to other players, by taking the time to study and
>understand the way that the lottery/lotto system of number generation
>works.
This illustrates the danger of letting go unanswered any
misunderstanding of your postition. James, I never used the word
'artificially.' Assuming that lotto numbers are drawn at random,
there is *absolutely nothing* you can do to increase your probability
of winning the jackpot, apart from buying more tickets. This result
is an elementary one in probability and statistics.
Note the claim is much stronger than "there is no artificial way to
increase the probability." The claim is that there is no artificial
way, nor natural way, nor any way whatever.
Wheels don't work (i.e., don't increase your probability of winning
the jackpot). Sums don't work. Systems, either known or unknown, do
not and will not ever work. It is easily proven, but the proof
depends on the assumption that lotto numbers are drawn at random.
Attacking that assumption is your only reasonable avenue of attack.
But the burden is on you to demonstrate a lack of randomness.
Furthermore, a failure to pass a test for randomness, does not in
itself imply that wheels or sums will work for you.
Recently I saw an interesting story on tv. It was about a casino
employee who noticed that whenever the lights went out, the keno
computer would initialize to some default state and begin generating
the same sequence of numbers. Based on that observation, he won a
fortune. The casino fired him and refused to pay. But he won in
court. The court ruled that he had done nothing wrong.
What this man did, in effect, was find and exploit a non-randomness of
the process. He successfully attacked the false assumption that the
process was random. And thus, in this particular case (not all
instances of non-random processes are as easily exploitable), he was
able to "devise" a system (namely, hang around and wait for the lights
to go out again) that worked beautifully and perfectly.
Wheels, sums, voodoo, etc., would *never* have worked (in the sense
above) for him. And they won't work for you either, if the lotto
process is random. And I defy you to show us how it is not random.
That systems don't work should please rather than annoy you. Your
chance of winning the jackpot is as good as that of the most brilliant
mathematician alive, but no better. (I take that back. The most
brilliant mathematician probably doesn't play, so your chance is
better.)
>It is unfortunate that the math weenie/guru contingent in this
>group have less confidence in the theory and laws of probability and
>statistics to have practical application than I do. Einstein spent his
>life seeking an underlying law that would tie all natural law together.
>In spite of his great achievements he never stopped believing that
>there was more to discover.
I think Einstein spent the last years of his life trying to unify
physics, something most experts think can be done. What you are
proposing to do is something everyone with even a modicum of
mathematical sophistication agrees is impossible.
Your claim that math weenies have no confidence in probability is a
hoot. It is *you* who have no confidence in it. And that is, of
course, because you have a very, very poor understanding of it.
>Perhaps, my critics would like to answer these questions, or at least
>consider them:
> 1. Is there any such thing as a perfect random number generator?
> 2. If such a machine could be discovered, would it defy the laws
>of physics and continue to generate perfectly random numbers, with no
>evidence of mechanical bias?
> 3. Is there any such thing as bias and is there then anyway to
>evaluate it?
This question is further evidence that you've never dipped even three
chapters into an introductory text in probability and statistics.
> 4. Would such manifestation of bias be any different than the
>mechanical bias of a roulette wheel which was discussed in "BEAT THE
>WHEEL"?
> 5. Rather than remain locked in at the probabilities of matching
>one or more numbers on a given ticket, wouldn't it make more sense to
>study the theory and actuality of the random number generation process
>itself, so as to compare expectations with results?
>>>>>>>Don't these subjects at least warrant discussion? The Bible
>states: "There is nothing new under the Sun" but it is also true that
>at any time in history man only knows a small fraction of what can be
>known ünder the Sun".
>
>MORE TO FOLLOW
>JAM...@AOL.COM
There are a lot of interesting questions you could ask about
randomness, random number generators, bias, etc. You've asked
none of them. You've merely piled one addled non sequitor upon
another.
Do yourself a favor. Either defer to the many knowledgeable people
who have attempted to steer you in the right direction, or at least
make an honest effort yourself to learn about these things.
And if you won't do that, perhaps you can tell us why you are so proud
to be so ignorant of this stuff? At this level it's not rocket
science, you know.
Rgrds,
Ron Meisenheimer (ro...@ns.net)
Regards,
Ron Meisenheimer (ro...@ns.net)
cmo...@sedona.intel.com
JAMES SARNELLI <jam...@ix.netcom.com> wrote:
Predicting the red powerball number 4 times in a row would cause your
detractors to eat crow. Why not publish your predictions?
cheers, cmoore (not speaking for my employer)
Dave Budd
>the qnantum level. Who appointed you keeper of the rec.gambling.lottery
>gate, and empowered you to exclude anyone who doesn't have multiple
>degrees in math and physics. I don't have to prove anything to you, nor
He's not excluding anybody. He's just pointing out false information in
postings, because he cares enough to want other people to have accurate
information. It just happens that you've posted several pieces of false
information.
>do I have any desire to. I intend to discuss lottery/lotto techniques
>with normal people who have paid their money and have a right to access
>to these discussions. Why don't you math weenie/gurus talk to each
>other in your secret code, and leave us normal people to communicate in
>plain old English?
Math is the least secret code in the world, and it scores over "plain old
English" in excluding ambiguity. You've occasionally shown that your grasp
of English isn't so hot, anyway.
--
How many user support staff does it take to change a lightbulb?
"We have a copy of your system here and it works perfectly"
Dave Budd: +44|0-161-275-6033 fax 6040 D.B...@mcc.ac.uk
Edmund C. Hack
[James' question on perfect random number generators and my response
about possible machines based on radioactive decay deleted, including
obscure reference to Schroedinger's Cat.]
>>
>>On a more practical note, it is not necessary that a mechanical random
>>number generator be _perfect_, just _good enough_. Good enough so that
>>the biases cannot be detected within the lifetime of the machine. For
>>example, some crypto techniques are secure enough for routine data
>>(takes weeks to crack, by which time the data are useless) and some are not
>>(crackable in minutes or hours).
>>
>>
>>>>>>>The only flaw in your brilliant presentation is that we are not
>talking about a quantum physics RNG, BUT ABOUT A CHEESEY MACHINE THAT
>USES SOME SORT OF VACUUM SYSTEM TO SPIT NUMBERED PIMG PONG BALLS OUT A
>PLASTIC TUBE ONE AT A TIME.
James, that is why I talked about the difference between a _perfect_
machine (a radioactive/quantum system) and _good enough_. It is probably
true that the ping-pong ball machines (used in Texas for Pick 3, but not
Pick 6) are _good enough_ systems. Again, what I mean is that the system
is random enough that you can't detect the bias before either the machine
is:
a - retired from use,
b - the ball set is changed, and/or
c - it breaks and a component is changed that affects the statistics.
I know that the Texas lottery discards the ball sets for Pick 3 after a
certain number of uses.
The Pick 6 is more complicated. The machine is more complicated, with 3
rotating arms grabbing balls that are being mechanically bounced around.
The balls are solid (rubber-like substance). They have 3 machines and at
least 6 sets of balls. The balls and machine are picked at random
(drawing lots) before each drawing, the balls are weighed and the machine
is tested. If the same number comes up "too many" times during the tests,
the machine and ball set are picked again. This complicates the task of
looking for any mechanical biases.
My opinion is that these machines are probably _good enough_ RNGs. It is
simply a trade off of cost, security, and ease of use. A Pick 6 number
generator based on radioactive decay might be "better" in the sense that
the numbers are more random, but the security problems would be enormous,
since it would be electronic and not electromechanical.
>Are you guys so over educated that it is
>impossible for a normal person to communicate with you without opening
>up the floodgates of irrelevant verbiage. Are we now clear about the
>kind of RNG WE ARE TALKING ABOUT? The name of this group should give
>you a hint. Why not turn on the boob tube and watch the lotto/lottery
>drawings and you will find what normal people mean when they talk about
>a random number generator.
James, I am sorry that you have chosen to take my post, an attempt to
sincerely answer the questions that you posed, as an attack. You asked a
couple of interesting questions ("Are there perfect RNGs?" "Would it
violate the laws of physics?") that I attempted to answer. It is
unfortunate that you have become so belligerent here. It makes your
claims appear to be much less credible.
To answer the "questions" above -
I've seen the Texas Pick 3 and Pick 6 drawings on TV several times. The
local stations carry the big Pick 6 drawings live (happens once a month
or so) and I saw the Pick 3 once while channel surfing, so I am familiar
with the equipment. I also buy Pick 6 tickets from time to time, esp if
the jackpot gets above break even (discounting multiple winners being
possible).
In short, I do live in the real world. Chill out.
--
Edmund Hack \ "The great prince issues commands,
ech...@crl.com \ Founds states, vests families with fiefs.
Houston, TX \ Inferior people should not be employed."-regnaD kciN
William Chin
>In article <3t6272$k...@ixnews3.ix.netcom.com>,
>JAMES SARNELLI <jam...@ix.netcom.com> wrote:
>[stuff deleted]
>> Perhaps, my critics would like to answer
>>these questions, or at least consider them:
>>
>> 1. Is there any such thing as a perfect random number generator?
>> 2. If such a machine could be discovered, would it defy the laws
>>of physics and continue to generate perfectly random numbers, with no
>>evidence of mechanical bias?
This is a joke, right?
>I'm pretty sure, remembering back to my original career as a Quantum
>Mechanic, that a randome number generator can be constructed out of a
>timer and a detector of radioactive decay products. The output of a
>radioactive source can be considered a random process (Poisson process, as
>I recall), with known statistics.
Even if you accept q.m. as gospel truth, you would then have to measure
the "random" process with a "timer" and then find out if the physical
devices introduce a bias. The question is more mathematical than physical.
>
>There have been hints around that such devices have been used to generate
>random sequences for crypto work in the past. Such devices may require a
>steady supply of cats to operate. :-)
In practical terms, the question of whether a generator is random is moot.
The question is how close to random. And it is a delicate question too.
Let me refer you (pl., at least the numerati) to Knuth's "Art of Computer
Programming", vol. II for a good easy explanation of random no. generators.
Of course you have to know a little math. If you read this material you
would realize that a generator can be constructed to almost any conceivable
degree of randomness, where the notion of degree of randomness is made
rigourous. This notion is a litlle technical, but says that (within a
specified degree of accuracy) no matter how you slice it the sequences
will always look random.
Given this (standard) material, you will agree that anyone
with a little training can make a rng that is perfectly random
for all practical purposes. It's not a mystical process, just
simple arithmetic.
If you can't do the math, you can just look it up in "numerical
recipes in C". This stuff is all old hat and standard to math weenies.
>
>On a more practical note, it is not necessary that a mechanical random
>number generator be _perfect_, just _good enough_.
Right. So I think that the lottery engineers are up to the task, ergo
for all practical pruposes, the lotteries are random. If you differ,
prove otherwise. The generators are run on computers, and as long
as they can do arithmetic correctly, the rng will be accurate. So let's
forget about the mention of "physical machines" and "laws of physics".
The problem with this group reflects the tragic enumeracy of a lot
of American population. The concepts of probability are beyond
the comprehension of people who don't understand even more basic
concepts such as "function" and variable. One can't really pound
a point home to someone who has a superficial and flawed understanding
of the basic concepts involved. Sarnelli and his cohort have rightly realized
that he does'nt even speak the same language as the math weenie's and
therefore doesnt' want to talk to them anymore. This is to his credit.
What he doesn't realize is the tragic part.
-Bill Chin,
slumming and procrastinating.
jac...@xmission.com
>In article <3ta0a6$j...@ixnews5.ix.netcom.com>,
>JAMES SARNELLI <jam...@ix.netcom.com> wrote:
>>Are you guys so over educated that it is
>>impossible for a normal person to communicate with you without opening
>>up the floodgates of irrelevant verbiage.
Translation: James asked some questions and didn't get the answer
that he wanted, so now he's upset.
Two questions:
1) Are you holding yourself up as an example of a "normal person"?
2) Where do you get off complaining about "irrelevant verbiage"? If
we're going to disallow irrelevant verbiage, we'll be forced to
ask you not to post, since 99% of everything you say here falls
into that category.
>>Are we now clear about the
>>kind of RNG WE ARE TALKING ABOUT? The name of this group should give
>>you a hint. Why not turn on the boob tube and watch the lotto/lottery
>>drawings and you will find what normal people mean when they talk about
>>a random number generator.
>
> James, I am sorry that you have chosen to take my post, an attempt to
> sincerely answer the questions that you posed, as an attack. You asked a
> couple of interesting questions ("Are there perfect RNGs?" "Would it
> violate the laws of physics?") that I attempted to answer. It is
> unfortunate that you have become so belligerent here. It makes your
> claims appear to be much less credible.
Edmund is being too kind. Many of us feel that it isn't possible for
James to appear any less credible than he already does.
This is simply another example of James pretending to want a serious
discussion, but being unwilling to "discuss" anything that doesn't
agree with his unorthodox (read "dimented") view of reality.
JAMES SARNELLI
>
>jam...@ix.netcom.com (JAMES SARNELLI) wrote:
>
>>In <3srgji$j...@mailer.york.ac.uk> Edward Croft <croft> writes:
>>>
>
>>>
I am sure is true of you also). Why do you and your contingent of math
weenie/gurus choose to arrogantly assume that everyone in the world
wants to be bored to death with your mathematical mumbo jumbo. You do
not need to have a degree in math to gamble (whether the lottery/lotto,
horses, dogs, casinos, etc) in any manner. Did it ever occur to you
that normal people might like to use this group to discuss techniques
that they have used in playing lotto/lottery, and that such discussion
might be enjoyable for them? If you are so certain that no system can
improve an individual's probability of winning a prize, why do you
waste your time trying to convert us hopelessly ignorant gamblers? I am
not proud to be ignorant of anything, I simply choose to allocate my
time to studying subjects that interest me. I am, however, proud not to
be a pompous, selfcentered, insecure individual(like some in this
group), who seizes every opportunity to stroke his ego, by pouring out
his pseudo knowledge all over cyberspace.
MORE TO FOLLOW
JAM...@AOL.COM
JAMES SARNELLI
>
>In article <3ta0a6$j...@ixnews5.ix.netcom.com>,
>JAMES SARNELLI <jam...@ix.netcom.com> wrote:
>>
>James, I am sorry that you have chosen to take my post, an attempt to
>sincerely answer the questions that you posed, as an attack.
>>>>Edmund: Please accept my apology for coming accross as beligerent.
My original questions were more rhetorical than serious. I have been
attacked by several individuals since I began posting to this group. I
agree that the machines and balls both vary from state to state and are
periodically changed; thus, making the detection of any bias more
difficult. The point I have been trying to make is that there appears
to be certain predictable characteristics in the lotto numbers drawn,
and it is not as necesary to know why these characteristics appear as
they do, as it is to just recognize that they exist and use this
information.. Mechanical and procedural bias may be one explanation.
Perhaps it is the math experts who are taking the loto/lottery
discussion too seriously. Why not discuss these predictable
characteristics with me and the others on a more common basis, as
opposed to a highly academic level, and see where it leads?
MORE TO FOLLOW
JAM...@AOL.COM
John Griffin
JAMES SARNELLI <jam...@ix.netcom.com> wrote:
[...]
>... The point I have been trying to make is that there appears
>to be certain predictable characteristics in the lotto numbers drawn,
>and it is not as necesary to know why these characteristics appear as
>they do, as it is to just recognize that they exist and use this
>information.. Mechanical and procedural bias may be one explanation.
>Perhaps it is the math experts who are taking the loto/lottery
>discussion too seriously.
> Why not discuss these predictable
>characteristics with me and the others on a more common basis, as
>opposed to a highly academic level, and see where it leads?
Excellent idea! Now you're talkin'! Finally, we can get down to
business here...
Please provide a list of these predictable characteristics.
cmo...@sedona.intel.com
>The point I have been trying to make is that there appears
>to be certain predictable characteristics in the lotto numbers drawn,
James, there is an article on this very subject in the June 1995,
"Communications of the ACM". It has almost nothing to do with
mathematics. It has everything to do with human cognitive biases.
Quotes are from the article which would be good reading for anyone
who tries to find purpose where there is only process.
"Cognitive scientists have discovered that people's intuitive
inferences and probability judgments do not strictly conform
to the laws of logic or mathematics, and that people are willing
to provide plausible explanations for random events."
As an example, which of the following coin tosses is more likely?
Heads, Heads, Heads, Heads, Heads, Heads
Heads, Tails, Tails, Heads, Heads, Tails
"If you picked the second sequence as more likely, you are like
most people who responded to similar questions in a classic study
of cognitive biases. The two sequences actually have an identical
probability of occurring, and, in fact, all sequences of six out-
comes are equally probable... The second sequence, with its mix
of (heads and tails), looks more *typical* or more representative
of the expected results than the first sequence, and hence is
judged to be more probable."
Normand Veilleux
>From: jam...@ix.netcom.com (JAMES SARNELLI)
>
>There are many subjects that I choose to remain ignorant of (as
>I am sure is true of you also). Why do you and your contingent of math
>weenie/gurus choose to arrogantly assume that everyone in the world
>wants to be bored to death with your mathematical mumbo jumbo.
ROTFL. Have you ever thought of becoming a cyberspace comedian James?
The only logical reason I can see why you attack mathematicians is
that they have shown your system has very big flaws in it and since
you can't get rid of the flaws, then you are attempting to get rid
of the mathematicians.
>You do
>not need to have a degree in math to gamble (whether the lottery/lotto,
>horses, dogs, casinos, etc) in any manner.
Correct. But, in order to PROVE that a WINNING SYSTEM really works,
you do need to understand the math. You have given ample evidence
that you do not understand neither probability nor statistics.
BTW, I noticed you didn't take me up on my previous offer. Post
your silly "morphing" mumbo jumbo to sci.math and ask what true
mathematicians think about it.
>Did it ever occur to you
>that normal people might like to use this group to discuss techniques
>that they have used in playing lotto/lottery, and that such discussion
>might be enjoyable for them?
James there is a world of a difference between saying "I have been
using these techniques when I play the Lotto" and "I HAVE A WINNING
SYSTEM. BELIEVE ME. IT REALLY WORKS." Just prove your system
improves the winning odds of a player and we will admit you have a
winning system: it's that simple.
>If you are so certain that no system can
>improve an individual's probability of winning a prize, why do you
>waste your time trying to convert us hopelessly ignorant gamblers?
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Sorry James, you might fit that category, but most people here
don't.
>I am
>not proud to be ignorant of anything, I simply choose to allocate my
>time to studying subjects that interest me. I am, however, proud not to
>be a pompous, selfcentered, insecure individual(like some in this
>group), who seizes every opportunity to stroke his ego, by pouring out
>his pseudo knowledge all over cyberspace.
Your last sentence is a perfect example of why no one believes you.
You just described your own behaviour and attributed it to others!
I warn you, if you don't stop your stupid attacks I will post your
"morphing" stuff to sci.math myself.
Jos van Kan
(Beating of James Sarnelli snipped)
>
>BTW, I noticed you didn't take me up on my previous offer. Post
>your silly "morphing" mumbo jumbo to sci.math and ask what true
>mathematicians think about it.
>
(More beating of James Sarnelli snipped)
>
>I warn you, if you don't stop your stupid attacks I will post your
>"morphing" stuff to sci.math myself.
>
Norman, on behalf of the regular readers of sci.math, I beg you, don't. As it
is we already are in possession of the regular contributions of various
crackpots and the amusement value of James's contributions is relatively low.
(As opposed to the rebuttals of Steve Jacobs, Ray de Gennaro and yourself, just
to name a few.) The whole thread has produced however, the mathematically
interesting problem of a minimal covering (a wheel) that, though not
influencing your expected return DOES increase your probability to win a prize
with a given number of tickets.
That this is not an interesting proposition can be shown by MY 8 ways to
increase this probability:
1. Buy 1/8 of all the tickets.
2. Buy 1/4 of all the tickets.
The other 6 methods I'll publish in a later posting. Well, on second thoughts
I'll leave them as an exercise to the reader. :)
Jos.
(Oh, note for roulette players: the equivalent of a wheel for roulette is to
cover all numbers with the same bet. Your probability of winning a prize would
be 1, but you'd be guaranteed to lose 1/37 of your money. (European roulette
that is.))
Thomas Sippel - Dau
In article <3tf6ib$i...@ixnews3.ix.netcom.com>, jam...@ix.netcom.com (JAMES SARNELLI) writes:
.....
- The point I have been trying to make is that there appears
- to be certain predictable characteristics in the lotto numbers drawn,
Quite.
1. There is a draw every week (for every weekly lottery...).
2. There are seven numbers drawn every draw (in the UK)
3. There are a few silly postings by JAMSAR (a drawback, I guess)
- and it is not as necesary to know why these characteristics appear as
- they do, as it is to just recognize that they exist and use this
- information.. Mechanical and procedural bias may be one explanation.
A fault in JAMSAR's keyboard ? In the software of his Internet Provider ?
- Perhaps it is the math experts who are taking the loto/lottery
- discussion too seriously. Why not discuss these predictable
- characteristics with me and the others on a more common basis, as
- opposed to a highly academic level, and see where it leads?
Hmm, Anglo Saxon Verbiage requested. In words of one syllable:
JAMES, SHUT UP.
I hope that was not too academic for you.
Thomas
--
*** This is the operative statement, all previous statements are inoperative.
* email: cma...@ic.ac.uk (Thomas Sippel - Dau) (uk.ac.ic on Janet)
* voice: +44 171 594 6904 (day) fax: +44 171 594 6958
* snail: Imperial College of Science, Technology and Medicine
jac...@xmission.com
>
> Hmm, Anglo Saxon Verbiage requested. In words of one syllable:
>
> JAMES, SHUT UP.
I feel compelled to comment that although I disagree completely with
almost every bit of nonsense that James spouts, I'm not particularly
out to silence him. His spoutings are on-topic for the group, if
nothing else.
Of course, I would prefer that James avoid posting about topics for
which he is completely "clue challenged". Hmm, if he did that then
I guess he wouldn't have much to say about the lottery...
"MORPHOOLISHNESS TO FOLLOW"
Dave Budd
>.....
>- The point I have been trying to make is that there appears
>- to be certain predictable characteristics in the lotto numbers drawn,
>- they do, as it is to just recognize that they exist and use this
>- information.. Mechanical and procedural bias may be one explanation.
>- discussion too seriously. Why not discuss these predictable
>- characteristics with me and the others on a more common basis, as
>- opposed to a highly academic level, and see where it leads?
OK, discussion of predictable characteristics:
What are they?
Do chi-square tests, t tests, or whichever standard statistical test is
appropriate (which we can't determine until you tell us which
characteristics you're thinking of) show them to be significant?
If you can't do the tests, post the data and we'll do them, as it may wise
up the few remaining folk who think it's possible to predict lotteries with
better than chance results.
Please note that chi-square and t tests are not "a highly academic level",
they're what you learn in the first year of a stats course (or earlier), are
basic to most evaluations of significance, and are well proven in the real
world.
Ron Blaylock
: >In article <3tf6ib$i...@ixnews3.ix.netcom.com>, jam...@ix.netcom.com (JAMES SARNELLI) writes:
: >.....
: >- The point I have been trying to make is that there appears
: >- to be certain predictable characteristics in the lotto numbers drawn,
: >- and it is not as necesary to know why these characteristics appear as
: >- they do, as it is to just recognize that they exist and use this
: >- information.. Mechanical and procedural bias may be one explanation.
: >- Perhaps it is the math experts who are taking the loto/lottery
: >- discussion too seriously. Why not discuss these predictable
: >- characteristics with me and the others on a more common basis, as
: >- opposed to a highly academic level, and see where it leads?
: OK, discussion of predictable characteristics:
: What are they?
: Do chi-square tests, t tests, or whichever standard statistical test is
: appropriate (which we can't determine until you tell us which
: characteristics you're thinking of) show them to be significant?
: If you can't do the tests, post the data and we'll do them, as it may wise
: up the few remaining folk who think it's possible to predict lotteries with
: better than chance results.
: Please note that chi-square and t tests are not "a highly academic level",
: they're what you learn in the first year of a stats course (or earlier), are
: basic to most evaluations of significance, and are well proven in the real
: world.
But, you may be talking to people who have never had a "stats" course. I
suppose this brings it into the realm of "highly academic level" to those
who don't know how to calculate stats but love to play the lottery anyway.
just MHO,
Ron Blaylock
< rg...@aimnet.com >
: --