Hockey's statistical revolution is in full swing, and it's time to take the next step
The numbers, man. This summer the hockey world has been all about the numbers.
Four NHL teams have made interesting analytics hires this off-season, and two of them - Tyler Dellow (Edmonton) and Eric Tulsky (still unknown) - and all their excellent work were readily accessible online to anyone with reasonable interest. Mix in stat-heavy acquisitions Kyle Dubas (Assistant GM, Toronto) and Sunny Mehta (New Jersey), and it’s obvious that hockey is embracing more than just goals and assists these days.
That this is happening is beyond debate. If you’re on the side opposing numbers, call your friend on your Nokia flip-phone and see what he thinks, ‘cause the rest of us don’t have time to listen.
With that dispute long settled, it’s time for those that embrace numbers to leave that fight (stop, he’s already dead) and embrace new ones - like making sure they’re not just throwing a bowl of number soup at the fridge and pointing to parts that stick.
What we’ve learned over the past few seasons of following stats beyond the “boxcars,” (goals, assists, points, etc.), is that our ability to assess players and teams is evolving, as it always should. We haven’t arrived at a destination. Nor will we - we’re always on the journey, constantly seeking ways to better inform our opinions. We want to better inform our opinions, you see, so we have more opportunities to call our friends idiots when they talk hockey like they’re Scotty freakin’ Bowman.
One of the many reasons Dellow separated himself from the rest of his stat-friendly community was his early understanding that sorting Corsi percentages didn’t plainly organize players in the league by their value. While that’s common knowledge these days, I found the stat conversation in years past, particularly on Twitter, glossed over circumstances and overvalued lists. (Players with a ton of offensive zone starts were assumed to be poor defensively, for example. That ain’t necessarily the case.)
I’m a little proud that something I quipped a year ago launched one of Dellow’s best pieces “Corsi and Context.” In it Dellow writes (emphasis mine):
There’s a point that has to be made here. When we’re talking about a player’s Corsi, what we’re really saying “This player had a shot attempt share of X% when he was on the ice in his particular mix of circumstances, with his particular mix of players.” It bears repeating a lot, because it gets abused at both ends of the spectrum. In the case of the Oilers mess at the bottom of their roster, I tend to think that there are some guys there who can play who are being buried in a sea of terrible hockey players. The defencemen who are on the ice will have an impact. The flip side’s true as well – take a look at Jordan Eberle’s Corsi% with and without Taylor Hall for an example of this. The job for hockey teams, and for commentary sources, is to suss out which guys are really pushing the bus and which guys are passengers.
The numbers we currently have are helpful in this regard, but I still develop a minor tic when I see the implication that adding Quality of Competition, Quality of Teammates and Zone Starts definitively tells us “what a player is.” That’s some nice context, absolutely, but it’s not nearly beyond refute.
Hockey analysis is moving in a new direction now. The Maple Leafs are one of the teams adding NBA-style movement tracking cameras to their rink to gather spatial information. In basketball, those cameras have spawned concepts such as the “gravity score,” which measures how tightly defenders monitor a player even when he’s away from the ball. They also have a “distraction score” which measures how often a defender strays from his particular check to, as Zach Lowe puts it “patrol the on-ball action.”
This is where I see hockey stats going. Certain players do earn themselves more space when shooting than others. Certain guys do “go to the dirty areas” more. We can’t track these things yet, but it doesn’t make them any less true while we await the technology.
While we thrum our fingers on the table wishing on the future, we have what we have - mostly raw counting numbers - and those have proven valuable. Ask the Leafs, Devils, and Oilers. Just because some people fear numbers, fear change, or just fear that sports will cease to become the mindless distraction that accompanies four Coors Light bottles at the end of a long work day (it won't), doesn’t mean the rest of us shouldn’t feel free to dig in and learn things about the game we otherwise would’ve missed.
We’re always on the journey to better analysis in sports, and while significant strides have been made in hockey, we’ve got a long ways to go yet. The next few years should see our evaluation skills grow at an exponential rate, so here we go - on to the next step.