>This brings up a stat I'd like to see along with PP/PK percentages:
>amount of PP/PK time per goal scored/allowed.  THN has reported this
>in the past for the NHL.
 
I did a quick little statistical test on the ranking produced by
"power play goals per minute" vs "power play goals per opportunity"
with THN's data from a few years ago. There was no significant
difference. That is, teams were ranked about the same using both
statistics. If you were #1 in one stat, you were also near #1 in the
other.
 
So, the implication is, if you only want to evaluate who is better or
worse than someone else on the power play, it doesn't matter which
stat you use. Since PPG/opportunity is MUCH easier to do, I see no reason
to bother doing the other. PPG/opportunity is not perfect, but it appears
to be good enough. The flaws of PPG/o apply to everyone.
 
 
Keith