The way we see the game has fundamentally changed. Gut feelings may still be a factor but younger general managers at all levels are allowing data to come to the forefront. Numbers are influencing how hockey is played in just about every aspect. Heck, I used numbers to help create a strategy for the JV high school hockey team I coach.
That said, some might argue that numbers are being too heavily relied upon. People argue that tools like Wins Above Replacement (WAR) and Player Isolates don’t take everything into account, and they’re right, to a degree. A lot of the advanced analytics tools rely upon data gathered by the NHL which, for anyone who doesn’t know, isn’t great.
There are gaps in the data and while some will argue they’re minuscule, or even micro, I believe they play a decent part. This is where my story begins. I began tracking data a few years ago for a blog I wrote in college. That blog was objectively terrible but the data I had was good enough to be given an opportunity on a larger stage.
Eventually, I decided to track the Nashville Predators. As I write this on April 6th, I’ve tracked 40 games, or about 2,400 minutes of Predators hockey. It’s my first project in five years where I had creative control, and I loved every second of it. I tracked everything from zone exits and entries, to puck battles and slot passes.
I will show you my data over the next few articles and when we get to the end, I will make all my data public. That means everything, whether as a final result or on a game by game basis.
So before we dive in, two things. All data below is at five on five and I want to give a big shoutout to my editor, Justin, and Michael Wade for being supportive. I couldn’t have done this without both of you.
I figured we’d work our way up the ice for this series, so zone exits come first.
I’m sure we can all agree that exiting the zone is important, but are all exits created equal? I would argue that they’re not and that maintaining possession is the highest priority. This goes against conventional hockey theory, as anyone who played defense years ago has the phrase, “hard and off the glass” burned into their memory.
This shift in play has allowed defenders to shine in roles that they wouldn’t be suited for 30 years ago. Players like Roman Josi would also dominate as did their predecessors (Leetch, Niedermayer), but the third and fourth defenseman who can move the puck is who my data puts a spotlight on. Players like ex-Predator Sam Girard and current Predator Dante Fabbro (although to a lesser degree), make up for their defensive lapses by playing the defensive zone as little as possible.
So, with this emphasis put on moving the puck with possession, let’s take a look below.
Before we proceed, let’s talk about what we’re looking at. A possession-based exit is when the Predators leave the zone with ownership of the puck. That means carrying the puck over the blue line or making a pass directly (read: to a player in mind, not to space) out of the zone. A non-possession exit is basically anything else, that means dump outs, failed passes, or just generally when a player hits the puck wildly and it leaves the zone. Note that I have not included failed exits here, I tracked that stat but I’m saving it for another day.
Also, don’t trust Miikka Salomaki’s, Daniel Carr’s, or Steven Santini’s numbers. Their sample sizes are all around the five-game mark. But enough of this.
As we can see, Roman Josi sits at the head of the pack. He has over two times as many possession-based exits than non-possession exits. That’s not a typo or some problem in my math, Roman Josi is just that good, and while he’s been great his entire career, he took it to a whole new level this season.
The defense is who really shines in zone exits. They were heavily relied upon under Peter Laviolette’s system and while John Hynes’ system has the forwards playing deeper and carrying the puck up the ice more, the defense still does most of the heavy lifting. I think this is pretty clear in that most of the defense are near the top.
The part I like the most is how the forwards are mostly possession exiters. A lot of them don’t create a zone exit often, but at least they’re not just dumping the puck out at the first sign of trouble. Although it’s a little strange to see Rocco Grimaldi being the best exiter among the forwards as he just inches out Filip Forsberg. Grimaldi wasn’t nearly this good last year, and I don’t think anyone saw this coming.
In a perfect world, you’d like to see all of your skaters playing like Josi, but I think having the majority of players creating more possession exits than non-possession exits is a fine compromise. That said, let’s look at how these players are creating possession exits.
Carry vs Pass
I’ve thought long and hard about what is more valuable, a carryout or a pass out, and I can safely say that I have no idea. That’ll probably be my next project, but for now, let’s stick to what we know.
You’re going to notice something fun in the next graph. There are three types of players, defenders who mostly pass out, forwards who mostly skate out, and, as always, Roman Josi.
The same sample size issues from the last graph apply to this one for reference.
I’m sure the first thing you notice, besides Josi, is that little cluster of players on the right side. That’s Fabbro, Ellis, and weirdly enough, Hamhuis. Ellis has been dynamite all season long at passing the puck out of the zone, but as you can tell, he rarely skates. Any takers on why that is? If you guessed because Josi dominates zone exits, then hey, you’re right!
Fabbro and Hamhuis haven’t been nearly as consistent as Ellis, but it averages out to pretty good. Fabbro, especially, was absolutely terrible at moving the puck in the first part of the season. He would move the puck to Ekholm almost automatically. As the year has gone on Fabbro has stopped dumping the puck out and started taking control of exits. Hamhuis, on the other hand, has no discernable trend. He’ll have games with five or six pass-outs, and then have games with a dozen dump outs. Actually, he leads the Predators with five games with double-digit dump outs, no other Predator has more than one.
Otherwise, it’s kind of expected to see Grimaldi and Arvidsson lead the forwards in pure carry-outs. They’re the fastest and probably the most elusive, although I have a theory that Arvidsson’s numbers suffer because of playing with an injury.
I don’t think it’s a surprise to see most of the bottom six in the bottom right quadrant. They have a tough time moving the puck with possession and mostly rely on the defense to do it for them. That said, there is a very interesting phenomenon that I’ve dubbed the “Craig Smith Special”. Once a game, without fail, Craig Smith will tear up the ice and create a zone exit and entry. It has happened in every single game and it’s usually in the second half of the game. I cannot describe how it happens nor do I know why he doesn’t do it more. All I know is it happens like clockwork and it’s honestly just funny at this point.
Shameless Self Promotion
So this was part one in my multipart series on the advanced analytics I’ve personally tracked. If you want to see more, check here. If you want to know more about the data, how I track, or if you’re just angry, feel free to reach out on Twitter @Datarangas