🚀 go-pugleaf

It would seem to me that whether probability of exit or deaths per passenger
mile is a more valid measure would depend on why you want the info.

For example, the airplane vs car comparison probably makes more sense in terms
of miles, because I interpret the question to be "If I want to travel from
Boston to Chicago, is it safer to use a car or a plane."  Chances are
that you will exit the car more times between Boston and Chicago than you will
by plane.

On the other hand, if you are comparing it with the shuttle, the "per mile" is
specious.  The fact that you went 2 million miles to get from LC39 to the 
runway across the island, or the fact that the runway is 2 miles from the pad
is irrelavent.  So in this, I think I would compare entry/exit.

The other interesting part of this is that most airplane fatalities
(and so far all shuttle fatalities) occur close to entry/exit.  In other words
the length of the trip is unlikely to be significant to the probability of
fatality.  I doubt this is strictly true in cars.  While I have heard that most
accidents happen within 10 miles of home, I suspect that is because most of
person-miles spent in cars are within 10 miles of home.

Anyway...(why did I get into this discussion!!!!???)

Burns