# "Somebody has to win" a lottery. True? Not true?
**URL:** https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021
**Category:** Factual Questions
**Created:** [April 6, 2012, 6:06pm UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021 "2012-04-06T18:06:30Z")
**Posts on this page:** 12
**Page:** 2
### Author:  [@Jenaroph](https://boards.straightdope.com/u/Jenaroph)
#### Post date: [April 6, 2012, 11:37pm UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/21 "2012-04-06T23:37:02Z")
> [@Jonathan\_Chance](#):
>
> For myself, I prefer the odds of my underwear spontaneously transmuting into gold.
>
> Because that would be awesome.
Not if you’re wearing them at the time.
---
### Author:  [@Sanity\_Challenged](https://boards.straightdope.com/u/Sanity_Challenged)
#### Post date: [April 7, 2012, 12:27am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/22 "2012-04-07T00:27:10Z")
Yes, someone always wins.
The people selling the tickets.
---
### Author:  [@Francis\_Vaughan](https://boards.straightdope.com/u/Francis_Vaughan)
#### Post date: [April 7, 2012, 1:27am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/23 "2012-04-07T01:27:13Z")
Not all lotteries are pick number lotteries. The vast majority of lotteries are simple numbered ticket lotteries. These always have a winner. But they are not now the preferred way of running major, state run, lotteries. But for charities, and other minor lotteries they are usually mandated. Indeed the common way of running such a lottery is with two component tickets, where the purchaser gets one part, and the other part goes in the barrel to be drawn. Thus ensuring that only sold tickets may be part of the selection. Thus there is a simple mechanical process to ensure that there will always be a winner. Clearly lotto style lotteries, as observed above, do not guarantee a winner, nor guarantee a single winner either. One suspects the maxim of always having a winner has been misapplied from the common lottery case, where it is trivially true, to the case that is in everyone’s mind, where it is not.
---
### Author:  [@leahcim](https://boards.straightdope.com/u/leahcim)
#### Post date: [April 7, 2012, 1:31am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/24 "2012-04-07T01:31:16Z")
> [@ultrafilter](#):
>
> … there will be a winner in finite time with probability one.
…Which is a very careful phrasing designed specifically for the purpose of not saying “there is no way for there not to be a winner in finite time”.
---
### Author:  [@Corcaigh](https://boards.straightdope.com/u/Corcaigh)
#### Post date: [April 7, 2012, 10:34am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/25 "2012-04-07T10:34:43Z")
> [@KneadToKnow](#):
>
> Before the big jackpot deal last weekend, a co-worker was saying how she really was thinking of buying a ticket because “somebody has to win.”
You are taking her too literally. It’s highly doubtful that she meant someone _will_ win, but that someone _might_ win, and that someone might well be her.
---
### Author:  [@Alley\_Dweller](https://boards.straightdope.com/u/Alley_Dweller)
#### Post date: [April 7, 2012, 10:42am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/26 "2012-04-07T10:42:27Z")
> [@leahcim](#):
>
> …Which is a very careful phrasing designed specifically for the purpose of not saying “there is no way for there not to be a winner in finite time”.
And just so everyone understands, “two years after the sun explodes and turns the earth into a bunch of hydrogen atoms” is a finite amount of time.
---
### Author:  [@KneadToKnow](https://boards.straightdope.com/u/KneadToKnow)
#### Post date: [April 7, 2012, 11:42am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/27 "2012-04-07T11:42:44Z")
> [@Corcaigh](#):
>
> You are taking her too literally. It’s highly doubtful that she meant someone _will_ win, but that someone _might_ win, and that someone might well be her.
With respect, you don’t know her, and you are incorrect. I told her why I thought that statement wasn’t true and her response was, “Oh, I hadn’t thought of that.”
She meant exactly what she said.
---
### Author:  [@md2000](https://boards.straightdope.com/u/md2000)
#### Post date: [April 7, 2012, 12:25pm UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/28 "2012-04-07T12:25:38Z")
> [@Corcaigh](#):
>
> You are taking her too literally. It’s highly doubtful that she meant someone _will_ win, but that someone _might_ win, and that someone might well be her.
Yes, apparently your odds of winning go up tremendously when you buy your first ticket. No so much for the 100th ticket you buy…
> [@leahcim](#):
>
> …Which is a very careful phrasing designed specifically for the purpose of not saying “there is no way for there not to be a winner in finite time”.
It is theoretically possible for there to be no winner ever from now on. It’s also theoretically possible for half a million quick-pick winners to share the prize. (I knew of a group that won in a 10-million lotto prize; but, there were 3 tickets, so it split 3 ways. Then, they had 11 people in the group, so ultimately, they each took home $30,000 or so… no chump change, but not $10M either. )
IIRC, the lotto pays out about half it’s earnings; the guaranteed jackpot, from their site if I read it right, is $40M. So I assume when all is added up, they sell $100M or about 50M tickets? Your odds of winning are 1 in 179M or so; suggests the odds of a winner any draw are 1 in 3.5; slightly worse that throwing a dice and getting 3 or better.
What are the odds you could roll a dice forever and not hit 1 or 2?
---
### Author:  [@ultrafilter](https://boards.straightdope.com/u/ultrafilter)
#### Post date: [April 7, 2012, 3:17pm UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/29 "2012-04-07T15:17:00Z")
> [@md2000](#):
>
> It is theoretically possible for there to be no winner ever from now on.
>
> What are the odds you could roll a dice forever and not hit 1 or 2?
Both of these events have probability zero by the theorem I referred to above.
---
### Author:  [@md2000](https://boards.straightdope.com/u/md2000)
#### Post date: [April 7, 2012, 6:56pm UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/30 "2012-04-07T18:56:26Z")
> [@ultrafilter](#):
>
> Both of these events have probability zero by the theorem I referred to above.
It’s the same logic that says 3.99999… = 4.0 - yes it does for infinte repetitions of 9; but generally we are talking about the practical lifetime of the exercise - until the glaciers eventually cover the USA, or until the last person who gives a damn about the theorem stops flipping coins coninuously, etc. So let’s be explicit; it is theoretically possible that nobody will win the lottery for 10 years; or for so long that the jackpot basically encompasses all the money in country and nobody can buy a ticket. In practical terms, the odds are sufficiently astronomical, why even argue it…
Its like the halting problem for computer programs; you can prove a program will halt eventually by waiting until it halts, but after X millennia if it still has not halted, then that does not necessarily prove it will never halt…
---
### Author:  [@leahcim](https://boards.straightdope.com/u/leahcim)
#### Post date: [April 8, 2012, 1:50am UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/31 "2012-04-08T01:50:34Z")
> [@md2000](#):
>
> It’s the same logic that says 3.99999… = 4.0 - yes it does for infinte repetitions of 9; but generally we are talking about the practical lifetime of the exercise - until the glaciers eventually cover the USA, or until the last person who gives a damn about the theorem stops flipping coins coninuously, etc.
But even accepting the premises of the problem (an infinite number of random drawings, random numbers played by at least one player) there is a distinction to be made. There is nothing inconsistent about an infinite sequence of no-winner drawings, in that there is nothing about the set-up of the problem that forbids it, so it is not “impossible” per se. The set of universes in which there is no winner ever form a set of measure zero in the space of all universes, so the probability of there being no winner is exactly zero. I was just pointing out that **ultrafilter** was being particularly careful in his language to say that instead of the less-well-defined statement, “it is impossible”.
Contrast with a drawing, where one entry at random is selected to be the winner. Then the set of all universes where there is no winner is not just of measure zero, but is actually the empty set.
---
### Author:  [@Corcaigh](https://boards.straightdope.com/u/Corcaigh)
#### Post date: [April 8, 2012, 5:02pm UTC](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021/32 "2012-04-08T17:02:50Z")
> [@KneadToKnow](#):
>
> With respect, you don’t know her, and you are incorrect. I told her why I thought that statement wasn’t true and her response was, “Oh, I hadn’t thought of that.”
>
> She meant exactly what she said.
So I take it she was one of the winners then?
😛
[Previous page](https://boards.straightdope.com/t/somebody-has-to-win-a-lottery-true-not-true/618021.md?page=1)