Math Punter finds Lotto "Holy Grail"

[Moochie]

Believe it or not:

http://www.wnbc.com/money/17263616/detail.html


My take is not.


M.

[ynot]

Believe it or not:

http://www.wnbc.com/money/17263616/detail.html


My take is not.


M.
Surely one would only want to patent something if one was going to sell it. Why would one sell a system that could win Lotto? Perhaps thy want to patent dumb luck. :cool:

[Jackalgirl]

Are all lotteries determined by the random-ball method? I can't understand how you could find "an algorithm" unless the lottery in question were being determined by an algorithm. In other words, the only way to figure it out is if the process were somehow (intentionally or not) non-random, right?

[TobiasTheViking]

Are all lotteries determined by the random-ball method? I can't understand how you could find "an algorithm" unless the lottery in question were being determined by an algorithm. In other words, the only way to figure it out is if the process were somehow (intentionally or not) non-random, right?

right

[Jackalgirl]

Tobias, I love you. : )

[TobiasTheViking]

Tobias, I love you. : )
good

[geni]

If you are buying multiple tickets it is posible to develop formulars to prevent less than optimium number choices.

[Rasmus]

If you are buying multiple tickets it is posible to develop formulars to prevent less than optimium number choices.

I guess having multiple duplicate entries wouldn't qualify, though? ;)

Even though that is what I would do if I had a way of knowing the numbers in advance. Getting more shares of a large jackpot would be helpful.

But seriously: If I could calculate the numbers in advance here's what I would do:

a) win the next possible jackpot, no questions asked.
b) possibly try to sell the winning ticket at a loss, so my name won't be connected to the win.
c) wait for the next super-jackpot and play 4 or 5 tickets - possibly allow my wife etc. to do the same.

I should be incredibly rich by then. If I really wanted to make more money I could now sit down and work out a plan. At no time would that plan involve me announcing to the world that I was able to accurately predict the lottery numbers.

[Horatius]

Surely one would only want to patent something if one was going to sell it. Why would one sell a system that could win Lotto? Perhaps thy want to patent dumb luck. :cool:



The patent would give one the right to exclude others from using the formula. If you could show that someone else used the formula to win part of a jackpot you had also played, you could sue them for the damages caused by their infringement - which would be the portion of the jackpot they won.


Of course, here in Canada it isn't possible to get a patent merely for an equation - they would have to embody it in some sort of method or apparatus. Simply calculating a sting of numbers wouldn't cut it. I'm not sure about the US, though.

[Ratatoskr]

Strangely, they forgot to include this one (from the article on FoxNews.com no less):
He claims he developed a secret formula that has helped him make money in the lottery over the last 20 years, although he says his winnings have been small compared to this jackpot.

So, developing a formula, using it for 20 years, and then claim victory when you after 20 years win the jackpot?

Yeah, good luck with that one...

[Alkatran]

Well there are really two possibilities:
- This guy has found a way to [partially] predict lotto numbers
- This guy is in league with the company he won money from.
- This guy has been playing a lot of numbers in triples for a long time, knowing that even if he only eventually won a smaller prize [eg. 10K], that would get him into the news and he could make millions selling books or whatever, recouping his losses and then some.

Number 1 is ridiculously unlikely. Lottery companies exist because this problem is intractable. Number 2 is conceivable, but you would need evidence of the involvement. Number 3 is probably the most likely, although you probably need to be extremely non risk-averse to even consider trying it.

How many tickets would he have to buy to have a greater than 5% chance of winning a 10K prize at least once a decade?

[tsig]

If you are buying multiple tickets it is posible to develop formulars to prevent less than optimium number choices.

Sounds like you're saying that some numbers come up less often than others.

This would mean that the lottery was not random'

[Rasmus]

The patent would give one the right to exclude others from using the formula. If you could show that someone else used the formula to win part of a jackpot you had also played, you could sue them for the damages caused by their infringement - which would be the portion of the jackpot they won.

Also, whoever organized the lottery would see to it that you could no longer use your formula.

I would rather share a few of my jackpots.

[Horatius]

Also, whoever organized the lottery would see to it that you could no longer use your formula.




Well, yes, there's that too.

[neltana]

Sounds like you're saying that some numbers come up less often than others.

This would mean that the lottery was not random'
I don't know if this is what geni was thinking of. However, given a random lottery where you are placing multiple bets and near misses offer a reduced payout, you can increase your odds of winning something by avoiding overlapping numbers in subsequent bets.

For instance, imagine you had a lottery where 2 balls were chosen from a pool of 6 balls with no replacement. Matching both numbers wins the big prize while matching one number wins the small prize.

If you bet 1,3 on one ticket and 3,4 on the next, there is a 2/14 chance you will win nothing (a draw of 2,5 or 2,6 matches nothing). However, betting 1,3 and 2,4 has 100% chance of payout (12/14 times, the little prize, 2/14 times, the big one).

Now, scale that up to 6 numbers chosen from a pool of 42 balls. Clearly, avoiding overlapping bets will increase your odds of winning a prize from next to nothing to slightly better than next to nothing.

[Maus]

I think more likely the couple is patenting the formula so they can try and cash in on any future winnings by others. If someone else wins the couple can claim that person used the formula to pick the winning numbers even if they didn't (and despite the fact the formula is bogus.)

-Maus

[Mongrel]

I don't know if this is what geni was thinking of. However, given a random lottery where you are placing multiple bets and near misses offer a reduced payout, you can increase your odds of winning something by avoiding overlapping numbers in subsequent bets.

For instance, imagine you had a lottery where 2 balls were chosen from a pool of 6 balls with no replacement. Matching both numbers wins the big prize while matching one number wins the small prize.

If you bet 1,3 on one ticket and 3,4 on the next, there is a 2/14 chance you will win nothing (a draw of 2,5 or 2,6 matches nothing). However, betting 1,3 and 2,4 has 100% chance of payout (12/14 times, the little prize, 2/14 times, the big one).

Now, scale that up to 6 numbers chosen from a pool of 42 balls. Clearly, avoiding overlapping bets will increase your odds of winning a prize from next to nothing to slightly better than next to nothing.

But then add in multiple slips, using the same numbers, so that you get a larger share of the pot if it has to be split (more likely on rollovers when more people buy more tickets)

[Brian-M]

When I was a teenager (mid 90's), I bought a lotto winning system that claimed to be mathematically proven by some university, it's inventor possessing numerous lotto winning slips, and suggestions that lottery agencies over in England were trying to get it banned. I think it cost 50 or 60 dollars by mail-order.

It came with two systems.
In system 'A' you pick 9 numbers and play 12 games based on those numbers. If all 6 winning numbers were in the 9 numbers you picked, you were guaranteed at least a 5-number win.
In system 'B' you pick 12 numbers and play 42 games based on those numbers. If all 6 winning numbers were in the 12 numbers you picked, you were guaranteed at least a 5-number win.

I worked out how it worked, and checked it with my computer. Yes, it would work as claimed. Excitedly, I played system 'A' a couple of times, but with no result. It was only then that I took the effort to work out how many games, on average, I would have to play before all 6 winning numbers showed up in my 9 (or 12) chosen numbers. Needless to say, the results were very disappointing.

Over all, I don't feel I was ripped-off because of all the hours of amusement I got out of it, like a complex puzzle, seeing if I could improve on it or adapt it for other games like Keno for better results. It also improved my understanding of gambling, and years later when I found a library book called "How to win at Roulette", I used ideas I came up while playing with the system to prove to myself that none of the gambling systems in the book worked.

I did manage to improve on System 'A', coming up with my own system that gets the same results with only 8 games. I can post it here if anyone wants.

[CFLarsen]

Well there are really two possibilities:
- This guy has found a way to [partially] predict lotto numbers
...
Number 1 is ridiculously unlikely. Lottery companies exist because this problem is intractable.

Not quite.

The premise is that lottery numbers are completely random, but it takes a loooooong time for all numbers to be drawn equally often. All he has to do is look at those numbers who have not yet been drawn that often, and focus on those.

[Naughtyhippo]

Not quite.

The premise is that lottery numbers are completely random, but it takes a loooooong time for all numbers to be drawn equally often. All he has to do is look at those numbers who have not yet been drawn that often, and focus on those.

Why? Isn't each draw independent of the draw before, and each number still only has the same probability of appearing? Just because I've been flipping my fair coin and getting a chain of 5 heads, the probability of tails appearing in my next flip is still only 0.5.

[Ratatoskr]

Sure, we could look at statistics which tells us that 7 is drawn 50% more times than 6. That doesn't mean that the odds are that for the next draw...

Former patterns doesn't prove future patterns.

[CFLarsen]

Why? Isn't each draw independent of the draw before, and each number still only has the same probability of appearing? Just because I've been flipping my fair coin and getting a chain of 5 heads, the probability of tails appearing in my next flip is still only 0.5.

Yes, each draw is independent of the others, but it still takes a while before all numbers are evened out.

Check any statistics on the frequency of lottery numbers and you will find that there are slight variations.

[AntiTelharsic]

Those variations contain no information about future draws, however; you can't use them to give yourself better odds. Thinking otherwise is known as the gambler's fallacy (http://en.wikipedia.org/wiki/Gambler's_fallacy) (apologies if you know this).

[Naughtyhippo]

Those variations contain no information about future draws, however; you can't use them to give yourself better odds. Thinking otherwise is known as the gambler's fallacy (http://en.wikipedia.org/wiki/Gambler's_fallacy) (apologies if you know this).

Thanks, AntiTelharsic, I believe this was what I was trying to point out.

[AntiTelharsic]

Thanks, AntiTelharsic, I believe this was what I was trying to point out.

Yeah, your last post was entirely correct.

[Foolmewunz]

My Spidey Sense Is Tingling

> Why did they buy their four tickets at four separate locations? It's based on numbers you choose. Location has nothing to do with it.
> Seems a little early, already on the weekend to be ready with their lawyer and patent story.

If the story is even true* (I can't get to any Lottery results sites from work) I agree that if the guy's been playing for 20 years and it's a weekly lottery, he may have just been hoping for one big win (10, 20, 30, 50 K), but with multiple tickets, so he played multiples every week. It'd sure get him some publicity, or he could use the multiple winning tickets as the dust jacket to his book.

I don't know the price in Wisconsin, but assuming it's a one buck ticket, he could've spent anywhere from $4000 on up. Admittedly 1.4 million on a 4000 investment is good, but how many other tickets did they buy. (Not one article mentions whether or not they held, say forty or fifty tickets for the draw. For Lottery Junkies, that is not all that uncommon.)

*Just saying, ya know..... all the articles I can find (and there are surprisingly few hits for a pretty interesting bit of human interest) seem to be based on one mention.

[leon_heller]

When I was playing around with artificial neural net software a colleague of mine wanted a copy to predict lottery numbers. In spite of me trying to convince him that it was a waste of time he kept insisting that it could work.

There is an on-line source of true random numbers generated by a Quantis quantum random number generator. On the web page is a tongue in cheek suggestion that anyone who wins a lottery with one of their numbers donates half the winnings to the website owners:

http://www.randomnumbers.info/

Leon

[Ratatoskr]

http://streamofthoughts.com/files/dilbert2001182781025.gif

[NobbyNobbs]

The patent would give one the right to exclude others from using the formula. If you could show that someone else used the formula to win part of a jackpot you had also played, you could sue them for the damages caused by their infringement - which would be the portion of the jackpot they won.


Of course, here in Canada it isn't possible to get a patent merely for an equation - they would have to embody it in some sort of method or apparatus. Simply calculating a sting of numbers wouldn't cut it. I'm not sure about the US, though.


You cannot patent an equation in the US.

Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the confitions and requirements of this title.

And before someone asks, it has already been decided by a court of law that an equation is not a "process".

Now, if he were to build a machine that makes use of his equation, that would be a different matter.

[rjh01]

When I was playing around with artificial neural net software a colleague of mine wanted a copy to predict lottery numbers. In spite of me trying to convince him that it was a waste of time he kept insisting that it could work.

There is an on-line source of true random numbers generated by a Quantis quantum random number generator. On the web page is a tongue in cheek suggestion that anyone who wins a lottery with one of their numbers donates half the winnings to the website owners:

http://www.randomnumbers.info/

Leon

Scam way to earn heaps of money. Persuade many people you can predict things, like the winners of the horses. Give them random predictions, some of which will be right. They pay you nothing unless they win, in which case they pay you a % of the winnings. You only need a few people to send you money to make it worthwhile.

How to persuade your victims? Just a simple ad will be a start. I have seen a few of these myself. Usually they want money up front.

[Foolmewunz]

Well, well, well!!!! Does anyone live in Wisconsin who'd care to take this up with the local news or State Lottery?

http://www.wtop.com/?nid=456&sid=1462725

In the original article by the AP, they list the winning numbers...

6 - 10 - 11 - 19 - 21 - 49

And they cite the drawing as being from "this weekend's SuperCash state lottery". With the article dated 21 Aug, that'd make the drawing from the 15th, 16th, or 17th.


Yet, here are the official numbers from The State of Wisconsin's site for all the SuperCash drawings of the past however longs, and it seems that those six numbers have not come up at all, in the 13 months I backtracked.

http://www.wilottery.com/lottogames/ascash.asp

[Cuddles]

Well there are really two possibilities:
- This guy has found a way to [partially] predict lotto numbers
- This guy is in league with the company he won money from.
- This guy has been playing a lot of numbers in triples for a long time, knowing that even if he only eventually won a smaller prize [eg. 10K], that would get him into the news and he could make millions selling books or whatever, recouping his losses and then some.

Number 1 is ridiculously unlikely. Lottery companies exist because this problem is intractable. Number 2 is conceivable, but you would need evidence of the involvement. Number 3 is probably the most likely, although you probably need to be extremely non risk-averse to even consider trying it.

You missed out what I think is the most likely explanation:
- This guy has a method of predicting lottery numbers that isn't actually any better than chance, but he has fooled himself into thinking it is.

As pointed out already, the gambler's fallacy is a very common way for people to fool themselves into thinking they can predict random events. People often attribute success in lotteries and similar to the way they choose numbers, and misunderstanding randomness leads to many people thinking that certain numbers and combinations of numbers are more likely than others.

I think it's entirely possible that this guy really believes he has something. So do millions of other people. The only difference between them and him is that he happened to get lucky eventually.

[AndyD]

People get frustrated when you point out to them that the odds of their six randomly chosen numbers coming up are exactly the same a 1, 2, 3, 4, 5 & 6 coming up - regardless of where they buy the ticket, whose birthdays their numbers represent, what their astrologer almost, maybe hinted at telling them or what numbers have already been drawn in the past.

WRT to CF Larsen's point (since I'm not a stats or maths geek), is there a subtle, or not so subtle, difference between odds and likelihood?

[sanguine]

People get frustrated when you point out to them that the odds of their six randomly chosen numbers coming up are exactly the same a 1, 2, 3, 4, 5 & 6 coming up - regardless of where they buy the ticket, whose birthdays their numbers represent, what their astrologer almost, maybe hinted at telling them or what numbers have already been drawn in the past.


I bought a lottery ticket (the only one I've bought to date) two years ago, and selected the numbers 2, 3, 5, 7, 11, 13, and 17 (the last being a "bonus value" slot). The checker at the store remarked on how low they all were, and I said, "Well, it doesn't matter..."

But intuitively, people expect "random" to mean "evenly distributed over my sample set, regardless of size," and that makes a series like the one I picked feel like a bad play.

[Horatius]

Former patterns doesn't prove future patterns.


...if the draws are truly random. There was a case in Canada back when the Casino du Montreal first started playing keno (http://catless.ncl.ac.uk/Risks/15.80.html):



Montreal -- Daniel Corriveau said he hopes that his 'victory over the
system will give hope to others.'
The computer analyst and his family received more than $620,000 [1C$ =
U$0.75], including interest, from the Montreal casino yesterday, weeks
after they overcame odds of one in six billion and beat an electronic keno
game three times in a row."

The author explains the following key points:

o Corriveau used an "antique 286" computer to analyse 7,000 combinations from
the keno game, [which uses an electronic pseudo-random number generator].

o Corriveau noticed that the electronic game was repeating numbers in a
predictable pattern.

o Corriveau and several family members bet on what they predicted would be due
to come up; they won three times in succession.

o The Casino managers shut the game down and called the police.

o The Surete du Quebec [provincial police] fraud squad investigated; Corriveau
and his family even took polygraph tests.

o The president of Loto-Quebec, the Crown Corporation which owns the Montreal
Casino, admitted that the problem was theirs: they had failed to test the
electronic game before using it.

o The keno game is missing its clock, [used to reset the pseudo-random
number generator]; therefore it started all over again every time it was
powered off.

o Police are continuing their investigation to find out if the clock was
missing when the game was delivered or whether it has been stolen.


But note, this guy had detected a major flaw in their system - if the system had been working properly, he'd have been out of luck, as it were.

[JohnG]

Lotto "Holy Grail"? Pff, it's only a model.



BTW, a lot of what you've been discussing is covered in a very entertaining and informative book called CONNED AGAIN, WATSON by Colin Bruce. I didn't understand just how counter-intuitive statistics, probability, etc. could be until reading the book.

[Foolmewunz]

Well, well, well!!!! Does anyone live in Wisconsin who'd care to take this up with the local news or State Lottery?

http://www.wtop.com/?nid=456&sid=1462725

In the original article by the AP, they list the winning numbers...

6 - 10 - 11 - 19 - 21 - 49

And they cite the drawing as being from "this weekend's SuperCash state lottery". With the article dated 21 Aug, that'd make the drawing from the 15th, 16th, or 17th.


Yet, here are the official numbers from The State of Wisconsin's site for all the SuperCash drawings of the past however longs, and it seems that those six numbers have not come up at all, in the 13 months I backtracked.

http://www.wilottery.com/lottogames/ascash.asp

BUMP TO CORRECT MY ERROR.

I have no idea why the article above cited the wrong winning numbers. I found an earlier article (21/08) though, which cited the correct numbers.
It only briefly mentioned that the husband says he has a system. But it quotes the wife as saying WE'VE BEEN PLAYING THE SAME NUMBERS FOR YEARS.

Playing the same numbers for years is not a system. This is a daily lottery. Four separate tickets is $4 per day. That's $1460 a year. Not a bad return, even if they'd been playing the numbers for ten years straight. But playing a set of numbers you like over and over and over and over might be defined by some as a system, but I can't see it being patent office material! I'd love to see the look on the gullible when they open the sealed book of "The Secret" and find that it has 75 nearly blank pages, with "Play the same numbers every day until you win" on each page.

[NobbyNobbs]

Scam way to earn heaps of money. Persuade many people you can predict things, like the winners of the horses. Give them random predictions, some of which will be right. They pay you nothing unless they win, in which case they pay you a % of the winnings. You only need a few people to send you money to make it worthwhile.

How to persuade your victims? Just a simple ad will be a start. I have seen a few of these myself. Usually they want money up front.

Even better. Send out a letter to 1000 people saying that Stock A will rise. Send a letter to 1000 people saying that it will fall. Whatever it does, those people to whom you predicted correctly are your new list. Send a letter to 500 saying that Stock B will rise, 500 are told it will fall. Rinse, repeat.

In the end, send a letter to the dozen or so who are left, "I have now correctly predicted stocks 7 times in row. Don't you think you should invest with me?"

[Cuddles]

I bought a lottery ticket (the only one I've bought to date) two years ago, and selected the numbers 2, 3, 5, 7, 11, 13, and 17 (the last being a "bonus value" slot). The checker at the store remarked on how low they all were, and I said, "Well, it doesn't matter..."

Well, it doesn't matter for the probability of the numbers coming up, but it does make a difference to how much you win. For example, many people choose numbers based on dates, especially birthdays, which means if you choose numbers below 31 you are more likely to share the jackpot than if you choose higher numbers. Of course, this only applies to lotteries where the prizes are split between winners, if the prizes are fixed, there really is no difference between numbers.

[Darat]

I'm going to use an argument from incredulity! If they had such a system why didn't they just keep winning the jackpot week after week, after all what they are claiming they can do is not illegal or even unethical?

[CFLarsen]

I'm going to use an argument from incredulity! If they had such a system why didn't they just keep winning the jackpot week after week, after all what they are claiming they can do is not illegal or even unethical?

Think, McFly. Think.

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

Why else would psychics claim that they can't use their own powers to win the lottery?

Think, McFly. Think.

[3point14]

Yes, each draw is independent of the others, but it still takes a while before all numbers are evened out.

Check any statistics on the frequency of lottery numbers and you will find that there are slight variations.


Like there are when you're tossing a coin? (http://www.youtube.com/watch?v=X1uJD1O3L08)

[Cuddles]

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

Why else would psychics claim that they can't use their own powers to win the lottery?

No there isn't. Even the Sylvia Brownes of the world aren't making several million dollars every week. The only reason it's more profitable to make money fooling people is because there isn't actually any way to win lotteries other than chance. If someone actually had a method of winning, they could easily make far more money playing the lottery than they could doing pretty much anything else.

[HarryKeogh]

No there isn't. Even the Sylvia Brownes of the world aren't making several million dollars every week. The only reason it's more profitable to make money fooling people is because there isn't actually any way to win lotteries other than chance. If someone actually had a method of winning, they could easily make far more money playing the lottery than they could doing pretty much anything else.

I think he was being sarcastic.

[sanguine]

Well, it doesn't matter for the probability of the numbers coming up, but it does make a difference to how much you win. For example, many people choose numbers based on dates, especially birthdays, which means if you choose numbers below 31 you are more likely to share the jackpot than if you choose higher numbers. Of course, this only applies to lotteries where the prizes are split between winners, if the prizes are fixed, there really is no difference between numbers.

True, true. How many people each drawing bet on the early primes, do you suppose? :)

[Foolmewunz]

I'm going to use an argument from incredulity! If they had such a system why didn't they just keep winning the jackpot week after week, after all what they are claiming they can do is not illegal or even unethical?


As with all the "You Can Control Your Destiny and Be a Millionaire" self-help revivalist preachers, the answer (as you no doubt know) is that they can't do it and the way to real wealth is in the potential suckers they can fleece to buy their book, atend their seminars, buy their DVD, and attend the Super Gold Success Seminars at the end of the year.

In this case, the question, albeit rhetorical, to the "winners" is why they bought four tickets. The lottery in question will pay out 350,000 for a six number win. It will pay to a maximum of 20 winners, and then split 7 million over any number of winners over that. So they bought four tickets? But they had a system that was a locked-in certainty to win? Why didn't they buy ten or twenty?

The wife says they've been playing those numbers for several years. Was that the only set? Or did their system, like most supposed random number mastery systems, have five or six "possibles". Were they merely spending four bucks a week? Or forty? Or four hundred?

They won once.... on four identical tickets. They are playing on the public's basic misunderstanding of math. "Hey, we won the big prize four times in one week! You can do it, too!" If they want me to buy their book, they need to win four times in a row.

[Brian-M]

Like there are when you're tossing a coin? (http://www.youtube.com/watch?v=X1uJD1O3L08)

Anyone can get 10 heads in a row. Just keep flipping the coin for a few hours, restarting the count every time you get a tail until it happens. Once you've done that, you can show people the part of the film where heads came up ten times in a row, and impress them with your greatness.

What's that basis for the system Derren Brown is talking about in the video?

[AntiTelharsic]

Anyone can get 10 heads in a row. Just keep flipping the coin for a few hours, restarting the count every time you get a tail until it happens. Once you've done that, you can show people the part of the film where heads came up ten times in a row, and impress them with your greatness.

What's that basis for the system Derren Brown is talking about in the video?

Apparently it took more than nine hours:
http://en.wikipedia.org/wiki/Derren_Brown#The_System

[CFLarsen]

No there isn't. Even the Sylvia Brownes of the world aren't making several million dollars every week. The only reason it's more profitable to make money fooling people is because there isn't actually any way to win lotteries other than chance. If someone actually had a method of winning, they could easily make far more money playing the lottery than they could doing pretty much anything else.

I think he was being sarcastic.

Actually, no, I wasn't.

The Math Punter will make more money from selling his system than he could make just winning the lottery. Think of the number of people who buy amulets, go to psychics, or have their own "system". Think of the number of people who are just desperate enough.

Think there isn't enough people who will buy into this scam of his? You bet there is.

Of course, he would have to sell so many of them that the buyers would not win as much as he would, if he played alone. But then again: Since when were lottery freaks rational?

What we see here is exactly the same method as psychics use, when they are met with the question why they just don't use their prediction abilities to win the lottery: Oh, no, they would rather help other people. And, it doesn't work every time...

[HarryKeogh]

No there isn't. Even the Sylvia Brownes of the world aren't making several million dollars every week. The only reason it's more profitable to make money fooling people is because there isn't actually any way to win lotteries other than chance. If someone actually had a method of winning, they could easily make far more money playing the lottery than they could doing pretty much anything else.

Apologies, Cuddles. He wasn't being sarcastic. He thinks someone can make more money selling nonsense than walking into a convenience store, spending two dollars and winning tens (and sometimes hundreds) of millions in return. There are billions of dollars awarded annually in lotteries all over the world which dwarfs the income of the most successful psychics and scam artists (and pretty much everyone else on the planet).

Again, sorry. I totally thought he was kidding.

[CFLarsen]

Apologies, Cuddles. He wasn't being sarcastic. He thinks someone can make more money selling nonsense than walking into a convenience store, spending two dollars and winning tens (and sometimes hundreds) of millions in return.

Well, yeah. I do.

Because selling nonsense is a far better way to make money than playing the lottery.

There are billions of dollars awarded annually in lotteries all over the world which dwarfs the income of the most successful psychics and scam artists (and pretty much everyone else on the planet).

Again, sorry. I totally thought he was kidding.

...thereby missing the point entirely.

[HarryKeogh]

Well, yeah. I do.

Because selling nonsense is a far better way to make money than playing the lottery.

Watch those goalposts!

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

If I could win the lottery week after week I could make more than Kevin Trudeau, John Edward and Sylvia Browne in their best years combined!

[3point14]

Anyone can get 10 heads in a row. Just keep flipping the coin for a few hours, restarting the count every time you get a tail until it happens. Once you've done that, you can show people the part of the film where heads came up ten times in a row, and impress them with your greatness.

What's that basis for the system Derren Brown is talking about in the video?

That was kind of my point - coins (or dice or lottery balls) have no memory, therefore the odds remain... etc. etc. etc.

The system he's talking about was a system designed to pick the winners in eight (I think, might have been six) races. And he did. He gave a lady the winner in each and every race. It's a perfect system.

More here (http://en.wikipedia.org/wiki/Derren_Brown#The_System)

[CFLarsen]

Watch those goalposts!

I'm not moving any goalposts. To win the lottery, you have to play it.

If I could win the lottery week after week I could make more than Kevin Trudeau, John Edward and Sylvia Browne in their best years combined!

Then do it. Buy this guy's system.

In the meantime, I'll keep an eye on what Sylvia Browne makes.

Who do you think will earn the most?

[HarryKeogh]

Then do it. Buy this guy's system.

In the meantime, I'll keep an eye on what Sylvia Browne makes.

Who do you think will earn the most?

Stick with your claim:

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

The Math Punter will make more money from selling his system than he could make just winning the lottery.

I don't claim his system works. I'm sure it does not. Your claim is that there's more money in bilking people than winning the lottery week after week. That claim of yours is nonsense.

Think, McFly, think!

[AntiTelharsic]

Oh for crying out loud.

If you have a system that works, you're better off playing it. If you have a fake system, you're better off selling it.

Are you guys really going to argue like this about two completely different scenarios?

[HarryKeogh]

If you have a system that works, you're better off playing it. If you have a fake system, you're better off selling it.

Of course, you are 100% right.

If Claus stated "playing" the lottery instead of "winning" the lottery in his two quotes above I'd agree with him.

[CFLarsen]

I don't claim his system works. I'm sure it does not. Your claim is that there's more money in bilking people than winning the lottery week after week. That claim of yours is nonsense.

Think, McFly, think!

O..........K.

Can you please explain how you can win the lottery without playing the lottery?

I may have assumed that you would understand that to win, you would have to play. Pardon me if I assumed too much.

[Cuddles]

If I see a man playing the lottery without winning it, I'll shoot him.

[3point14]

If I see a man playing the lottery without winning it, I'll shoot him.

How many bullets do you have???

[HarryKeogh]

Can you please explain how you can win the lottery without playing the lottery?

I didn't make that claim. Everyone knows "You have to be in it to win it".

If you want to address your silly claims (quoted below) we can do that instead of addressing claims you imagine I make.

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

The Math Punter will make more money from selling his system than he could make just winning the lottery.

[CFLarsen]

I didn't make that claim. Everyone knows "You have to be in it to win it".

If you want to address your silly claims (quoted below) we can do that instead of addressing claims you imagine I make.

What "silly claims"?

Did I make the claim that, to win the lottery, you won't have to play it?

[HarryKeogh]

What "silly claims"?

..

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

The Math Punter will make more money from selling his system than he could make just winning the lottery.

[CFLarsen]

..

Did I make the claim that, to win the lottery, you won't have to play it?

[HarryKeogh]

Did I make the claim that, to win the lottery, you won't have to play it?

Did you? Who said you did? These are the claims I'm aware you've made in this thread:

There's a lot more money in bilking the easily fooled, than in winning the lottery, week after week.

The Math Punter will make more money from selling his system than he could make just winning the lottery.

[GreyICE]

What the dick just happened?


I've seen some dumb arguments in my day, but this is taking the cake. Claus can't write, we know what he meant to say, even though he can't type it correctly. He's the proverbial monkey with a typewriter, and it ain't Shakespeare.

He MEANT to say:

There's more money in selling this system then in playing the lottery week after week with it.
The Math punter will make more money from selling his system then he could make by playing the lottery with it.

We have 41,000 posts by him conclusively proving he can't write worth a damn and will argue about stupid things all day long. Stop poking the bear with a stick.

[jadebox]

I bought a lottery ticket (the only one I've bought to date) two years ago, and selected the numbers 2, 3, 5, 7, 11, 13, and 17 (the last being a "bonus value" slot). The checker at the store remarked on how low they all were, and I said, "Well, it doesn't matter..."

Yes, it does.

While your numbers are just as likely to win as any others, they are likely to win less money if you do win. :-)

Edit: Sorry ... I missed the earlier reply from Cuddles before I posted this ....

-- Roger

[Moochie]

I'm not moving any goalposts. To win the lottery, you have to play it.



Then do it. Buy this guy's system.

In the meantime, I'll keep an eye on what Sylvia Browne makes.

Who do you think will earn the most?

With 30 years (and counting) of playing the lottery, I reckon Sylvia and co. have made inestimably more money than I have. The only consolation I have is that a good slice of what I spent went to worthwhile causes.


M.

[Brian-M]

Wouldn't it depend on how profitable the system is?

If it gave you an even chance of a first division win, you'd keep it to yourself. If only it gave you a one-in-a-hundred chance of a second division win, you'd might want to sell it to tens of thousands of people for a several hundred dollars each.

[balrog666]

Yes, each draw is independent of the others, but it still takes a while before all numbers are evened out.

Check any statistics on the frequency of lottery numbers and you will find that there are slight variations.


The Law of Large Numbers says it will be arbitrarily long before the numbers "even out".

Which reminds of the casino story about when they put up displays on roulette showing the last 15 rolls (numbers and black/red) so that people could spot the "trend" - plays on roulette games increased markedly and profits on roulette went up even more. ;)

[Delvo]

All I don't get is how any of this has to do with "punting".

[AntiTelharsic]

All I don't get is how any of this has to do with "punting".

"Punter" is a British slang term for (among other things) a customer or a gambler.

[Starrman]

What the dick just happened?


I've seen some dumb arguments in my day, but this is taking the cake. Claus can't write, we know what he meant to say, even though he can't type it correctly. He's the proverbial monkey with a typewriter, and it ain't Shakespeare.

I would normally agree - but he made the statement twice. And despite clearly being wrong, he won't admit it.

From an outsider's perspective he looks pretty silly right now. I don't even know why I am commenting, but my four year old is quicker to admit she is wrong than this dude. And it annoys me greatly when someone has so much trouble admitting they either made a mistake or are wrong. Why is that so hard for some people?

[Brian-M]

The Law of Large Numbers says it will be arbitrarily long before the numbers "even out".

Theoretically, they don't have to "even out" at all. It's random. :)

[balrog666]

Theoretically, they don't have to "even out" at all. It's random. :)


What part of "arbitrarily long" did you not understand? ;)


And anyway, if you want to make money on the lottery, you *SELL* tickets.

[Mongrel]

Theoretically, they don't have to "even out" at all. It's random. :)


Reminds me of the Dilbert sketch when he's taking to the realm of the accounting trolls (paraphrased, can't find it at the moment);

Troll guide: And this is our random number generator <points to another troll>
RNG Troll : 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9...
Dilbert : But he's saying the same number over and over again
TG : Yeah, that's the thing with random numbers, you never can tell

[Skeptical Greg]

Reminds me of the Dilbert sketch when he's taking to the realm of the accounting trolls (paraphrased, can't find it at the moment);

Troll guide: And this is our random number generator <points to another troll>
RNG Troll : 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9...
Dilbert : But he's saying the same number over and over again
TG : Yeah, that's the thing with random numbers, you never can tell

Uh, huh ..

http://forums.randi.org/showpost.php?p=3977562&postcount=28

[Skeptical Greg]

.... And despite clearly being wrong, he won't admit it. ...

...When has CFL ever admitted being wrong ?

[Richard Masters]

Even better. Send out a letter to 1000 people saying that Stock A will rise. Send a letter to 1000 people saying that it will fall. Whatever it does, those people to whom you predicted correctly are your new list. Send a letter to 500 saying that Stock B will rise, 500 are told it will fall. Rinse, repeat.

In the end, send a letter to the dozen or so who are left, "I have now correctly predicted stocks 7 times in row. Don't you think you should invest with me?"

One of my college professors claimed to have invented this. The idea was borrowed by one of his students and presented on some TV show.

It's a far better way to make money than playing the lottery.

Now back to the original topic (sorry!) Since computer generated numbers aren't completely random, it may be possible to find very subtle patterns, but you'd also need a very large sample of winning combinations. It would also help to know what algorithm and computer is being used, and more what the seed values are, which could range from the current time, to the cpu's temperature reading or the latest memory dump, or a combination of those.

Along the lines of what NobbyNobbs just said, though, you could analyze all state, local, lotteries with a long history and play only those lotteries which show a particular statistical tendency (internally);

As a way to mine a larger sample, you could analyze ALL lotteries which use the same selection method and treat all the results as if they came from the same lottery.

But in practice, predicting the lottery will generally be an exercise in futility

[TobiasTheViking]

When has CFL ever admitted being wrong ?

I've read him say that say that once.. it is going to be a bitch to find.. but if you really want to put me through that punishment i guess i'll do it to get the facts.. But he has done it

[Skeptical Greg]

I'll take your word for it ...

[Tunny]

Theoretically, they don't have to "even out" at all. It's random.
What part of "arbitrarily long" did you not understand? ;)


Although I don't think balrog666 is claiming this, a lot of people misinterpret the Law of Large Numbers to mean that, for example, if you bet on the flip of a fair coin long enough, you will eventually break even. This is not necessarily true, even with an arbitrarily long series of flips. The only thing guaranteed is that the distribution of head and tails will approach a .5/.5 ratio. The absolute difference in number of heads and tails will often get larger. There is no assurance that once you get behind, things will inevitably "swing back your way", or "even out" in the sense of getting you back to even.

If you've bet $1 on tails 1000 times, and are down $50 (i.e., tails came up 475 times, while heads came up 525 times), it's little consolation to go to a million flips, only to find that while the percentage of tails has now converged to .499, you've lost an additional $1950...

Tunny

[jadebox]

Although I don't think balrog666 is claiming this, a lot of people misinterpret the Law of Large Numbers to mean that, for example, if you bet on the flip of a fair coin long enough, you will eventually break even. This is not necessarily true, even with an arbitrarily long series of flips. The only thing guaranteed is that the distribution of head and tails will approach a .5/.5 ratio. The absolute difference in number of heads and tails will often get larger. There is no assurance that once you get behind, things will inevitably "swing back your way", or "even out" in the sense of getting you back to even.

I had a co-worker who analyzed all the lottery drawings to determine what numbers to pick. He had different ways to determine numbers. Since the Florida lottery uses balls to pick numbers, it is possible that the numbers picked aren't completely random. So, most of his schemes at least seemed somewhat plausible. But, when he described a scheme based on numbers "that were due" - since they had appeared less often than chance would dictate - I couldn't help but to laugh out loud.

-- Roger

[3point14]

Although I don't think balrog666 is claiming this, a lot of people misinterpret the Law of Large Numbers to mean that, for example, if you bet on the flip of a fair coin long enough, you will eventually break even. This is not necessarily true, even with an arbitrarily long series of flips. The only thing guaranteed is that the distribution of head and tails will approach a .5/.5 ratio. The absolute difference in number of heads and tails will often get larger. There is no assurance that once you get behind, things will inevitably "swing back your way", or "even out" in the sense of getting you back to even.
Tunny

I tried to explain this recently by describing one coin, but having multiple people each starting the betting at a different point along the string. Unless it's bang on 50:50 when someone joins, then there's no way that more than one person is ever going to work out with a balance of zero.

I think I've explained that really badly. I hope it makes at least a little sense.