What are the odds of winning the lotto twice?

Faith+1 · 05-03-2011 12:41AM

?

whysomoody · 05-03-2011 03:29PM

Blisterman wrote: »

The odds of a given ball coming out are the same regardless of how long it's been since that number has come out.

No, you are ignoring the theory of mean reversion, so they are in fact increased.

Iwannahurl wrote: »

P=SFA?

Somehow the drip-drip of not very useful information in this long thread brings to mind this joke.

A mathematician organises a raffle in which the prize is an infinite amount of money paid over an infinite amount of time. People are so impressed by this amazing prize that the mathematician has no trouble selling a large number of tickets. He holds the draw in due course, the winning ticket is drawn, and the excited winner shows up to claim the prize. Then the mathematician explains the payout method: "One Euro now, half a Euro next week, one third the week after that..."

go on....

Pherekydes · 05-03-2011 10:56PM

whysomoody wrote: »

No, you are ignoring the theory of mean reversion, so they are in fact increased.

You are making the same blunder as keane2097. The draws are independent and random. The odds on a given ball coming out in these circumstances cannot change!

go on....

That's the joke. The explanation was given earlier:

Pherekydes wrote: »

Yes. You have to know that the series mentioned sums to infinity, otherwise it's meaningless.

Bruthal · 05-03-2011 11:02PM

whysomoody wrote: »

No, you are ignoring the theory of mean reversion, so they are in fact increased.

No that is not right, 1 draw previous or 100 draws previous without a given number appearing will not change its chances of coming out in any single draw subsequent.

All that is happening is that the more draws there are, the more likely it will be to appear, and people will be thinking it must appear soon as we have not seen it in a while. But on each individual draw, it is no more likely to appear than any other number no matter what happened in previous draws.

whysomoody · 07-03-2011 12:14AM

robbie7730 wrote: »

No that is not right, 1 draw previous or 100 draws previous without a given number appearing will not change its chances of coming out in any single draw subsequent.

All that is happening is that the more draws there are, the more likely it will be to appear, and people will be thinking it must appear soon as we have not seen it in a while. But on each individual draw, it is no more likely to appear than any other number no matter what happened in previous draws.

The theory of large number disagrees with you, that mean reversion is a valid concept.

excuse me Law of ...http://en.wikipedia.org/wiki/Law_of_large_numbers

lugha · 07-03-2011 01:01AM

whysomoody wrote: »

The theory of large number disagrees with you, that mean reversion is a valid concept.

excuse me Law of ...http://en.wikipedia.org/wiki/Law_of_large_numbers

Mean reversion, as you call it, doesn't have much relevance to what is being discussed here. It is true that if in one particular draw, a lot of high numbered balls are drawn then the next draw will tend to have lower numbered balls. I.e. They revert or regress to the mean.

But that does imply the balls have any capacity for memory. The chance of a ball being drawn (conspiracy theories aside) is always the same, and is entirely independent of that ball's draw history.

07-03-2011 02:04AM

whysomoody wrote: »

The theory of large number disagrees with you, that mean reversion is a valid concept.

excuse me Law of ...http://en.wikipedia.org/wiki/Law_of_large_numbers

Mean reversion is a theory used to describe systems which exhibit certain characteristics. It is valid for some systems, not valid for others. It is most clearly visible in systems which combine luck and skill, like poker or the stock market. It has no place in a discussion about truly random systems.

It has nothing to do with the law of large numbers, which is applicable to systems based entirely on chance (die rolls, coin flips). Use coin flips as an example. The expected result will be approximately equal numbers of heads and tails. The law of large numbers tells us that the more coin flips we do, the closer the difference between the number of heads and tails as a percentage of the number of flips.

The bolded part is the important part. Say I have a coin which I've flipped 50 times, and it's come up heads every time. I then flip it 4,950 more times, and I end up with an exactly equal number of heads and tails. The absolute difference between the two is still 50. The difference between the two as a percentage of the total number of flips is only 1%. That's what the law of large numbers state. At no point does the probability of heads (or chance, or non-mathematical likelihood) change from 1 / 2.

Jesus Wept · 07-03-2011 02:32AM

28
06
42
12

No sign in Saturday's main draw though lotto plus 1 was a different story.
Stay tuned.

Bruthal · 07-03-2011 03:17AM

whysomoody wrote: »

The theory of large number disagrees with you

So does the lotto numbers drawn every week, not just for me, but for almost everyone else in the country.

Mellor · 07-03-2011 04:25AM

Iwannahurl wrote: »

P=SFA?

Somehow the drip-drip of not very useful information in this long thread brings to mind this joke.

A mathematician organises a raffle in which the prize is an infinite amount of money paid over an infinite amount of time. People are so impressed by this amazing prize that the mathematician has no trouble selling a large number of tickets. He holds the draw in due course, the winning ticket is drawn, and the excited winner shows up to claim the prize. Then the mathematician explains the payout method: "One Euro now, half a Euro next week, one third the week after that..."

That wouldn't be an infinite amount of money.
It would end up being only 2 euro at infinity

Jesus Wept · 07-03-2011 04:34AM

Approximately how much is an infinite amount of money?
100 trillion or something?

Mellor · 07-03-2011 05:06AM

At least a googolplex

Bruthal · 07-03-2011 11:45AM

Mellor wrote: »

That wouldn't be an infinite amount of money.
It would end up being only 2 euro at infinity

Your a true genius alright:rolleyes:

07-03-2011 12:02PM

Mellor wrote: »

That wouldn't be an infinite amount of money.
It would end up being only 2 euro at infinity

Nope, you're thinking of the series 1/2 + 1/4 + 1/8 + ... which is equal to one, so 1+1=2.

The series 1 + 1/2 + 1/3 + 1/4 + ... (the harmonic series) is equal to infinity

Rabble Rabble · 07-03-2011 12:18PM

Did keane2097 get super-rich yet with his novel approach to the lottery which has been scientifically disproven?

As an aside I tried this in Vegas once. I am not a gambler, and was in Vegas under duress. I understood the logic of losing on this system, but I reckoned may have paid out in a short time and there is a bit of ability in blackjack. So I played a blackjack machine - you played against the computer as "house" - and doubled my bet every time I lost. The machine took bets of $2 to $1000.

I was up $100, or so, in about 2 days. That last day I had a run of 7 losses, and then I packed it in. A few more losses would have killed me. I was also only three losses away from having to make the maximum bet.

I suspect this will happen to Keane. Still it keeps libraries open.

Rabble Rabble · 07-03-2011 12:22PM

That story is what happens in that system. You win most of the time but without infinite resources you eventually lose big.

keane2097 · 07-03-2011 12:23PM

Rabble Rabble wrote: »

Did keane2097 get super-rich yet with his novel approach to the lottery which has been scientifically disproven?

Won another couple of hundred over the last week between the UK lotto and a few others.

Lol at "scientifically disproven".

Rabble Rabble wrote: »

As an aside I tried this in Vegas once. I am not a gambler, and was in Vegas under duress. I understood the logic of losing on this system, but I reckoned may have paid out in a short time and there is a bit of ability in blackjack. So I played a blackjack machine - you played against the computer as "house" - and doubled my bet every time I lost. The machine took bets of $2 to $1000.

I was up $100, or so, in about 2 days. That last day I had a run of 7 losses, and then I packed it in. A few more losses would have killed me. I was also only three losses away from having to make the maximum bet.

I suspect this will happen to Keane. Still it keeps libraries open.

You obv play blackjack badly.

Rabble Rabble · 07-03-2011 12:25PM

keane2097 wrote: »

Won another couple of hundred over the last week between the UK lotto and a few others.

Lol at "scientifically disproven".

You obv play blackjack badly.

Seriously mate. This is simple statistics. You - a gambler - are arguing with people with degrees in the subject. That makes you look very very stupid. And you are going to lose all your money some day.

But it's good for the libraries.

keane2097 · 07-03-2011 12:29PM

Rabble Rabble wrote: »

Seriously mate. This is simple statistics. You - a gambler - are arguing with people with degrees in the subject. That makes you look very very stupid. And you are going to lose all your money some day.

But it's good for the libraries.

I doubt anybody with a degree worth its salt spends their time posting links to wikipedia on After Hours on boards.ie, making irrational claims about their point being proven by completely unrelated data.

Like I said, these theories are all well and good, but they're pretty well exposed as woo-woo by the reality of the history of lottery draws.

Rabble Rabble · 07-03-2011 12:36PM

keane2097 wrote: »

I doubt anybody with a degree worth its salt spends their time posting links to wikipedia on After Hours on boards.ie, making irrational claims about their point being proven by completely unrelated data.

Like I said, these theories are all well and good, but they're pretty well exposed as woo-woo by the reality of the history of lottery draws.

Lol.

You seem to think the "history of lottery draws" is your evidence.

Look, I'll try to be as remedial as I can to someone who thinks that if 4 hasnt come up for the last 5 draws it is likely to come up this draw.

Doubling your bet if you lose will win you money if you have infinite resources.
If you dont sometime - and it might be a year or so - you will lose 6,7, or 8 - times in a row. You have to double each time. Eventually you wont have the money to continue and that will wipe out your previous wins.

By the way, this idea is utterly simple. It's occurred to everybody. If it worked then gamblers would be rich, and bookies poor. The lottery would make a loss. Everybody would do this.

So some day , as theory predicts, you will lose your gains. Be sure to post here.

07-03-2011 12:41PM

keane2097 wrote: »

I doubt anybody with a degree worth its salt spends their time posting links to wikipedia on After Hours on boards.ie, making irrational claims about their point being proven by completely unrelated data.

Given you don't understand what an axiom is, linking you to complex proofs requiring a knowledge of mathematical syntaxes would seem pretty pointless. Wikipedia is as good a source as anything else, unless you're going to actually point out where it's wrong. And I never claimed to have a degree related to maths, this is all pretty basic stuff.

But since you came back, any chance of answering your fundamental questions?

28064212 wrote: »

The mathematical odds that 4 comes out in the next 1,000 draws are exactly equal to any other number, regardless of how long its been since 4 last came out. The odds do not change at any point along those 1,000 draws regardless of whether you take it as 1 series of events or 1,000 single events. True or false?

Can you give your definition of "non-mathematical possibility", "likelihood" or "chance"?

lugha · 07-03-2011 12:41PM

keane2097 wrote: »

Won another couple of hundred over the last week between the UK lotto and a few others

I fear you are making another oft made mistake here. Assuming that a bet where you usually win cannot be a bad bet. It is easy to devise a bet where the punter will almost certainly win but is still a bad bet.
Ie Were you to bet that the first number drawn in the lotto was NOT a 5 (say, or any specified number) at odds of 1/50 you would almost always win.
But it is still a bad bet.

Bruthal · 07-03-2011 12:47PM

keane2097 wrote: »

Like I said, these theories are all well and good, but they're pretty well exposed as woo-woo by the reality of the history of lottery draws.

Well with 2 main draws a week (was only one draw originally), back 24 years, thats about 2500 draws, out of 8 million combinations, its hardly a history that can be analysed in any way at all, which i mentioned something about already here.

keane2097 · 07-03-2011 01:59PM

Rabble Rabble wrote: »

Lol.

You seem to think the "history of lottery draws" is your evidence.

Look, I'll try to be as remedial as I can to someone who thinks that if 4 hasnt come up for the last 5 draws it is likely to come up this draw.

Doubling your bet if you lose will win you money if you have infinite resources.
If you dont sometime - and it might be a year or so - you will lose 6,7, or 8 - times in a row. You have to double each time. Eventually you wont have the money to continue and that will wipe out your previous wins.

By the way, this idea is utterly simple. It's occurred to everybody. If it worked then gamblers would be rich, and bookies poor. The lottery would make a loss. Everybody would do this.

So some day , as theory predicts, you will lose your gains. Be sure to post here.

My bank account is my evidence.

keane2097 · 07-03-2011 02:00PM

28064212 wrote: »

Given you don't understand what an axiom is, linking you to complex proofs requiring a knowledge of mathematical syntaxes would seem pretty pointless. Wikipedia is as good a source as anything else, unless you're going to actually point out where it's wrong. And I never claimed to have a degree related to maths, this is all pretty basic stuff.

But since you came back, any chance of answering your fundamental questions?

No.

keane2097 · 07-03-2011 02:02PM

robbie7730 wrote: »

Well with 2 main draws a week (was only one draw originally), back 24 years, thats about 2500 draws, out of 8 million combinations, its hardly a history that can be analysed in any way at all, which i mentioned something about already here.

Yeah I agree it's a valid point (I did originally as well).

My point however was that we're both making claims based on incomplete data.

You say trends don't exist and expect it to be proven over a suitably large data sample.

I say trends do exist and expect it to be proven over a suitably large data sample.

It's obv still unclear, but you have to admit that reality seems to be backing my assertion up thus far.

keane2097 · 07-03-2011 02:03PM

lugha wrote: »

I fear you are making another oft made mistake here. Assuming that a bet where you usually win cannot be a bad bet. It is easy to devise a bet where the punter will almost certainly win but is still a bad bet.
Ie Were you to bet that the first number drawn in the lotto was NOT a 5 (say, or any specified number) at odds of 1/50 you would almost always win.
But it is still a bad bet.

Probably, but it obv depends.

Fremen · 07-03-2011 02:27PM

28064212 wrote: »

Nope, you're thinking of the series 1/2 + 1/4 + 1/8 + ... which is equal to one, so 1+1=2.

The series 1 + 1/2 + 1/3 + 1/4 + ... (the harmonic series) is equal to infinity

The best thing is, if you square each of those terms, it all adds up to (pi squared)/6:

1 + 1/4 + 1/9 + 1/16 + ... = (Pi x Pi)/6

Fremen · 07-03-2011 02:34PM

keane2097 wrote: »

You say trends don't exist and expect it to be proven over a suitably large data sample.

I say trends do exist and expect it to be proven over a suitably large data sample.

We're saying trends don't exist based on a mathematical model that mimics the real world sufficiently well that it has been used all over physics, engineering, computer science and medicine.

You're using a data sample.

Fart · 07-03-2011 02:34PM

I'll win it this Friday. I promise this time, I've been lying for the past two years.

What are the odds of winning the lotto twice?

Comments