From: flaps@dgp.toronto.edu   
      
   johnf@panix.com (John Francis) writes:   
   >2)  Given the specific list of AFU birthdays, what are the odds that just   
   >    one of those birthday dates matches any particular day?   
      
   That one's easy.  There are 198 entries in the list Nick just posted,   
   of which 13 list only years; of the remaining 185, there are 147 unique   
   birthdates (discarding the year), so the answer is 147/365 (0.4 or 40%)   
   in non-leap-years, or 147/365.24 on average.   
      
   >1)  Given an (otherwise undefined) list of 198 people, what are the odds   
   >    that exactly one of them has a birthday that falls on a particular day?   
      
   Ignoring leap years, I think that this would be 1-(364/365)**198, which is   
   about 0.42 (i.e. 42%).   
      
   In the standard birthday problem, the reason that the probability of a   
   match is so high for small numbers is that _anyone_ can match _anyone_ else.   
   Once you're insisting that someone match just one particular date, the odds   
   are much lower, and are going to be closer to the intuitive answer -- 198 is   
   about half the number of dates in a year, and the probability is about half.   
      
   Btw, for those of you who forget, I'd like to brag that I compiled the very   
   first AFU ages list, in about 1997.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)