I don't find that there's anything advanced or fancy about the newfangled stats. They're all just basic arithmetic.
But the way some of the fancy stats are justified does actually turn out to be pretty complex, and I want to explore that more someday. Folks say a player "drives possession" or is better at "possession", when they're simply looking at a player's positive shot plus/minus (corsi), a move that is justified because Corsi supposedly highly correlates with time of possession. I'd like to test out just how good that correlation is, because there are a lot of other factors (like proclivity to shoot) that would seem to influence shot attempts.
I agree.
Hockey analysts have sort of taken the lead from Sabremetrics in baseball, but ignore a few simple things:
1) Baseball is both more static and more linear, hockey more fluid
2) Baseball stats are tracked to a much greater extent than in hockey
3) Baseball went through a span of about 25 years of new stuff coming out into the open, being tested and analyzed itself for importance by others, and resulted in
a lot of new stuff falling by the wayside. Some of it failed the eyeball test, some of it fell apart because it was arbitrary, and some of it was more for entertainment and settling arguments than for meaningful analysis in the moment.
4) Baseball has cycled back and forth between bringing new numbers and analyses out into the open and attempting to find a "grand unified number" that explains everything. I find myself in the former camp: invent, test, study, publish, and see if it holds up. Too much of hockey right now is in the latter camp, where a single number or two is held up as the real important part.
5) Baseball did go through a span where there were statistical idiots having too much power. These are the guys who are convinced that the game is no more complex than the numbers themselves, and that other externalities like a locker room cancer or a neurotic rookie or a high-strung coach are either minimized or eliminated.
I think that a lot of us who have a background in other sports and have done actual analysis in any sport tend to be pretty skeptical of what's going on in hockey. Hardcore baseball analysis has done incredible things for the game on the field, for the kids coming up, and for the Hall of Fame. Hockey...not so much. Yet. It reminds of me of what Bill James said in 2003, which was something about how we're now drawing information with a bucket instead of a thimble, but it's still being drawn from an ocean.
