Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!akgua!sdcsvax!sdcrdcf!hplabs!sri-unix!crummer@AEROSPACE
From: crummer%AEROSPACE@sri-unix.UUCP
Newsgroups: net.physics
Subject: Bad Coin or Bad Luck?
Message-ID: <1183@sri-arpa.UUCP>
Date: Thu, 24-May-84 13:44:20 EDT
Article-I.D.: sri-arpa.1183
Posted: Thu May 24 13:44:20 1984
Date-Received: Fri, 1-Jun-84 03:59:50 EDT
Lines: 11

From:            Charlie Crummer <crummer@AEROSPACE>

You're right about the coin test not being a proof.  In any sufficiently long
string of flips of a good coin there will be all imaginable sub-strings.  So
with a good coin there would be, e.g. a sub-string of 50 heads, one of 50 
tails, ... .  Since the probability of observing such a string is very low for
a good coin, however, if I observed such a string with a particular coin I 
would be willing to BET that the coin is biased.  No one expects a sure thing 
in gambling.

  --Charlie