Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: 527 heads out of 1,000 coin flips redux...
Replies: 9   Last Post: Jul 2, 2012 4:59 PM

 Messages: [ Previous | Next ]
 Douglas 73 Posts: 52 From: Pacific northwest, USA Registered: 1/15/11
527 heads out of 1,000 coin flips redux...
Posted: Jun 26, 2012 6:42 PM

Referencing:

http://www.nytimes.com/2011/01/11/science/11esp.html

My apologies for taking so long to answer to this subject, but I plead to
being old, slow thinking, and generally thorough.

Lately I have succeeded in creating a number of basic standard probability
scales as usually defined: more than zero, through to, and including, 1.
When I took Stat 101, someone forgot to tell me that these are merely
midpoint interval scales for all points except for the 1.

I have always taken an answer like 527 heads resulting from 1,000 coin
tosses as an actual 0.043852515 one-tail probability occurrence in the
infinite scheme of the universe. It turns out this probability amount is
merely a midpoint scale amount. The actual comparable one-tail probability
of 527 heads occurring is 0.005874295. This is a difference of almost 7 1/2
times. May I be allowed to call this a significant difference?

Now this is mostly statistics talk so far. I think us ordinary humans would
better understand these results in the form of odds against. In this
example, the generally accepted scale amount translates to almost 22:1 odds
against being by mere chance. Since I am an ordinary human, I take that as
being pretty chancy, but maybe possible. I think I would ask for the coin
flips to be repeated to see what happens for the next 1,000.

But the actual odds are 169:1. You have little chance convincing me these
527 heads are an actual fair result.

Douglas

Date Subject Author
6/26/12 Douglas 73
6/27/12 Richard Ulrich
6/27/12 Douglas 73
6/27/12 Richard Ulrich
6/28/12 Douglas 73
6/28/12 Ray Koopman
6/28/12 Douglas 73
6/28/12 Richard Ulrich
6/28/12 Douglas 73
7/2/12 Douglas 73