>music in which the probability (or frequency) of a measure
>not in common time is greater than 0.0000000001.

Even technically speaking, there is a problem in this definition.  When
you have two sets of music "items," you can measure the actual ratio
of common time vs.  mixed in both sets and say that with such and such
probability set A is popular, whereas set B is much more likely to be
classical.  In your problem, though, we have only single items.  To make
it simpler, how do you know each set is homogenous, i.e.  contains only
popular or only classical music? You can only define that composition
X is classical with probability 0.9999, 'cause it's 3/4 or 13/8, while
composition Y is popular with probability 0.9999, 'cause it's 4/4.
Personally, I think the probabilities are lower, and the distinction is
extremely blurred.  Frank Zappa with the baroque instruments from another
thread, is he 4/4? Is he popular?

According to Pushkin, Salieri tried to "measure harmony with algebra"
(sorry for lame translation).  According to Forman he ended up "King of
Mediocrities." I do not share this view, though.  If memory does not betray
me, in the mediaeval times, music and math were one subject, weren't they?

Alexey Fuchs