Naked Science Forum

General Science => General Science => Topic started by: oatman on 20/01/2011 20:01:19

Title: Can the probability of a random event change with frequency?
Post by: oatman on 20/01/2011 20:01:19

Sorry if this is bit of a stupid question and not terribly well worded!

But for example,

if the number 9 came out in the lottery one week, would it not be slightly more unlikely to come out again, if it did, then the chance of it coming out again even less likely and so on, and the same with flipping a coin etc.

Something tells me this isnt the case, but cant really see why!

Title: Can the probability of a random event change with frequency?
Post by: Joe L. Ogan on 20/01/2011 20:11:18

My limited knowledge of probabilities, it does not change because something has happened.  i.e. If you flip a coin and it comes up heads, the probability will remain 50% on the next flip.  This would carry through with any other probability.  Thanks for comments.  Joe L. Ogan

Title: Can the probability of a random event change with frequency?
Post by: RD on 20/01/2011 21:13:25

Quote from: Joe L. Ogan on 20/01/2011 20:11:18

If you flip a coin and it comes up heads, the probability will remain 50% on the next flip.

i.e. the coin has no memory of previous outcomes (http://en.wikipedia.org/wiki/Gambler%27s_fallacy), (same for roulette wheels and lottery machines).

Title: Can the probability of a random event change with frequency?
Post by: CliffordK on 20/01/2011 23:40:37

Many lottery machines involve some kind of a mechanical system.  For example some are based on balls blown in air.

If #9 is 1mg lighter than #10, then it could potentially be favored by a very slight amount.  Obviously there are tight controls on the system, but nothing's perfect.

Title: Can the probability of a random event change with frequency?
Post by: Geezer on 21/01/2011 04:14:47

There was a case recently (Eastern Europe I think?) where the same numbers came up in a lottery only one week apart. This raised some eyebrows, but as the two events are not related by any means known to science, it's just as likely (or unlikely)as any other set of numbers coming up a week later. It just seems strange, but that's probably because we try to find patterns, even when there really are none.

BTW, if anyone can prove me wrong, I think we should take a little trip down to Vegas.

Thankuveramuch.

Title: Can the probability of a random event change with frequency?
Post by: grizelda on 25/01/2011 02:19:38

I think there was a case at a casino in Montreal where the numbers in a keno game were chosen by computer which automatically reset after a power failure and generated the same numbers when it did (due to the fact that the backup battery was dead). The pseudo-random number sequence was randomized to the time which always reset to zero. Some observant patrons made some money off this.

Title: Can the probability of a random event change with frequency?
Post by: Geezer on 25/01/2011 03:12:53

Quote from: grizelda on 25/01/2011 02:19:38

I think there was a case at a casino in Montreal where the numbers in a keno game were chosen by computer which automatically reset after a power failure and generated the same numbers when it did (due to the fact that the backup battery was dead). The pseudo-random number sequence was randomized to the time which always reset to zero. Some observant patrons made some money off this.

Why didn't you tell us about this sooner?

I'd have been up there quicker than you can say "Don't step on my blue pseudo-shoes."