Telepathy test odds

[DrMatt]

Part of this query has already been discussed, but I need to know the odds of positive trials back to back as in the prelim and formal tests. Maybe you odds-makers and probability buffs can help with this.

The published odds are 2.6 million to one against a chance event of correctly guessing five non-repeating numbers from one to fifty two (preliminary).

The published odds are 135 million to one against a chance event for correctly guessing five non-repeating numbers from one to fifty-two and a wild card from one to fifty-two which may be a repeatable number. (formal test)

What are the odds of telepaths sucessfully completing these two situations back to back? I wish I knew how to calculate these odds, but I do not . Can anyone help.

This scenario seems mind-boggling but would be parlor games for true telepaths. Do you think Randi would accept this protocol?

I think that in consultation with a statistician, Randi wouldn't approve a protocol with such a small number of guesses. The objective of the test is to convince people that you really have a supernatural ability, not that you got lucky one day.

 

[geni]

I think any odds over 2500 to 1 against chance is a fair demonstration. With the proper protocals of course to prevent any cheating intentional or otherwise.

The odds for the priliminary test are 1000 to 1

The odds for the final test are 1000000 to 1

 

[epepke]

1/52*1/51*1/10*1/49*1/48 should be how it is done but that gives me odds of 311,875,200 to 1

If we go for 4 picks the odds are 6,497,400 to 1

If we go for 3 picks the odds are 132,600 to 1

I'm not sure where the orignal number came from.

No, that's for picking five in that order (permutation). You have to divide by the number of orders you can pick the cards, (permutation of the picked cards), which is 5!, or 120, in order to get the combinations. This gives 2598960, or about 2.6 million.

See Ask Dr. Math at http://mathforum.org/dr.math/faq/faq.comb.perm.html for more info.

Also, beware of these probability games. Bigger doesn't always mean better. The possibility that everyone involved in the test has been paid off and is just plain lying about everything may be small, but it isn't in the trillions and probably not in the billions either.

 

[ceptimus]

Here's a little applet I wrote that calculates things like 52C5 within the range useful to lotteries - alter the 'pick' and 'from' drop down boxes, and read the answer in the 'out of' box.

<APPLET codebase="http://www.mround.pwp.blueyonder.co.uk" code="lotto.class" width="400" height="200" align="left" hspace="20" vspace="14"></APPLET></P>