Assuming Yahoo! is working
--
One school
of thinking goes that the 15 of a
5-10-15 and the 30
of a 30-20-10 can both be not just
hard but
unreasonably hard -- thus the 5-10-15
where the answers are,
say, Isaac Newton, Isaac
Newton's chair, and the name
of Isaac Newton's dog
(Diamond). I think it's
reasonable that a bonus testing
depth of knowledge should
include a part whose
correct answering should be
considered an impressive
event; this means that the average
team should
average 10 on 5-10-15s and 15
on
30-20-10s.
Meanwhile, a breadth-of-knowledge 10-10-10 should,
on
average, probably be 20d.
Assuming a
clearly
unrealistic distribution of bonus types and purposes, about
a
third each of each of the above types mentioned,
that
averages out to 15ppb or 50%
conversion.
Note that
the number of fudges in the above
argument exceeds
the number of actual
numbers.
Edmund