Also why not just track the amount of time the puck is in each zone, I think that would give a decent idea about puck possesion as well..and which team is the agressor...
I think if they're going to be timing possession, for it to be meaningful it needs to be only when the attacking team has the puck, otherwise time spent when one team is setting up a breakout would count towards the other team's possession time. I don't really know why they can't do that though, I guess maybe the official stat keepers could since they already keep track of time on ice, but since Corsi and Fenwick seem to kept track of by blogger type people, it is too much work for them to watch each game with a stopwatch and everyone's numbers would differ slightly by the end of the game. Corsi and Fenwick are more cut and dry and thus easier for different people counting to end up with the same numbers. That's my guess.
EDIT: I just looked into it and found this article that sheds some light on this.
The easiest way to measure possession would be with a stopwatch of course, generating a time-of-possession number similar to what we're familiar with in, say, football. And the NHL actually did measure time of possession (and time in the attacking zone) for a brief period of time between 1997 and 2004. In this 2011 article from SI, the league's chief statistician said they stopped because, well, no one used it.
When I first learned about Corsi and Fenwick I remember thinking that they were already keeping track of possession time, well I guess this explains why I was confused. Turns out we don't really need Corsi or Fenwick to become more popular, we just need the NHL guys to start recording time of possession again, obviously it is doable since they did it in the past.
Edited by devilsfan26, 22 April 2014 - 08:26 AM.