Mathematics desk
< January 25	<< Dec \| January \| Feb >>	Current desk >

Welcome to the Wikipedia Mathematics Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

January 26

Powerball Paradox of Probability

In powerball lottery in USA.

The probability of ZERO pairs of adjacent numbers is 73.5% and

the probability of at least one pair of adjacent numbers is 26.5%

I have two candidates for powerball ticket

Candidate A is 2,7,10,14,15 and powerball is 17

Candidate B is 2,7,10,14,29 and powerball is 17

Unfortunately I only have money for a single powerball ticket.

therefore 73.5% of the time Candidate A has no chance of winning because it has one or more pairs of adjacent numbers (aka {14,15} )

therefore 26.5% of the time Candidate B has no chance of winning because it has zero pairs of adjacent numbers

Therefore I should spend my money on Candidate B when I buy my powerball ticket.

But the lottery officials tell me that the drawing of powerball is random therefore Candidate A has as much chance of winning as Candidate B

Now this is a paradox indeed! Ohanian (talk) 14:52, 26 January 2019 (UTC)[reply]

Ohanian, there is no paradox, 26.5% is simply the odds of picking any two consecutive numbers. That is independent of it being the winning pick. Roger (Dodger67) (talk) 18:01, 26 January 2019 (UTC)[reply]

By analogy, consider a random drawing which picks a single number from 1 through 3. Two thirds of the time the pick is odd (1 or 3) and one third of the time the pick is even (2). If you played odd/even, then odd is a more likely result, but that doesn't mean that 3 is more likely than 2. Likewise, if you could place a side bet that the winning Powerball draw will have zero adjacent numbers, then you would win that side bet 73.5% percent of the time. But that doesn't make one particular zero-adjacent-numbers ticket more likely to win than one particular at-least-one-pair-of-adjacent-numbers ticket; it just means that there are nearly three times as many possible tickets which fit that first description than there are those which fit the second description. -- ToE 02:03, 27 January 2019 (UTC) BTW, what you did here -- setting up a well defined problem which appears to be a paradox and asking for an explanation -- is an excellent learning technique. Good job![reply]