Gardiner, Phaneuf, and quality of competition

A reader writes:

Bolding these four words will draw the eye to those four words since they’re starkly different from the rest of the text on the page. The words themselves don’t matter, but the presentation is slightly different. It’s the departure from familiarity which attracts our attention.

In my last post I wrote out the Corsi rates (table at the bottom) of the six defensive pairing combinations that were together for more than 200 minutes in 2013-14. What stood out was Jake Gardiner and Morgan Rielly, posting an elite-level 54.6% Corsi rate on a team where so many players fail to crack 45%.

That stands out. That’s the highlighted section. The next two highest Corsi rates for a defensive pairing also happen to have Jake Gardiner on them. That also stands out. It’s no secret that the Maple Leafs are a team that struggles with puck possession, but they also happen to have a player who struggles a visibly less.

We’ve established that this stands out. What we haven’t established is whether or not the numbers mean anything. After all, this was also written in the comment section:

Hmm. This is a testable hypothesis from leafdreamer. What really stuck out to me was the next line in the commenter’s case against our defendant:

I’m going to snip the rest there. It’s interesting that “talent” doesn’t get equated with being the best defenceman on the team. It goes beyond talent—Gardiner gets results like no other defenceman. Since the end of the 2011-12 season, all you can say about Dion Phaneuf, the other man in the discussion, is that he faces the best competition in the league. Phaneuf doesn’t have the results to show for it (he used to!), however. Toronto were out-shot significantly with the Captain on the ice for the past two seasons.

Over at mc79hockey, Tyler has done some great work comparing “skilled” defencemen to “defensive” defencemen against the same competition. His latest post, where he looked at Matt Niskanen versus Brooks Orpik, shouldn’t surprise too many people: Niskanen is a quality blueliner, and Orpik’s major selling point is that he’s big.

Using the pages at Hockey Analysis, I spent some time looking at the star players Phaneuf spent the most time against, and compared how Gardiner played against the same players. The minimum qualifier was that the players needed 15 minutes of ice-time against Jake to qualify:

(Note that since Hockey Analysis calculates shot attempts in rates per 20 minutes rather than raw, the numbers may not be exact)

If our earlier hypothesis was correct, then Phaneuf should have much better numbers against these players than Gardiner. But he doesn’t. In fact, Gardiner’s Corsi rate is much higher. Perhaps not to the extent that we can definitively say Gardiner is better than Phaneuf as much as Niskanen is better than Orpik, but considering these numbers are since Gardiner’s rookie season and include that one last great year from Phaneuf and Carl Gunnarsson, it’s impressive that Gardiner’s numbers would be a percentage point higher.

And, yes, Phaneuf plays more against stars. He plays much more against stars. But he still doesn’t have results that clearly separate him from Gardiner when you try to control for the fact that Gardiner plays against mostly scrubs.

Gardiner doesn’t visibly play much defence. He represents the future of defencemen, however. I think in a few years, the majority of defencemen coming up through the junior ranks are going to be more of the puck-moving defencemen, as teams begin to recognize that in hockey, the best defence is a good offence. Since the game is played with a single puck, every second spent with the puck on Jake Gardiner’s stick is a second that the opposition does not have the puck. It’s why we’re more willing to forgive giveaway-prone players at this level: if you never have the puck, you’re never going to give it away.

Those of you who will write in the comment boxes about the defensive gaffes that Gardiner makes that hurt the team are sort of missing the bigger picture. The purpose of hockey is neither to score goals nor to prevent goals. It’s a combination of both, and the best players will do things that help you score more goals than the other team. Corsi, an approximation of possession, is an indicator of whether a team is doing the right things with a player on the ice. If the Leafs, a terrible, terrible possession team, suddenly become respectable when Gardiner and Rielly are on the ice, is that not an indication that the pairing is up to something that you’re missing as a viewer with your own inherent biases about the way defence should be played? Being a good defenceman doesn’t mean being big and mean. Some players, like Zdeno Chara and Chris Pronger, can work being big and mean to their advantage, but a lot of players who can’t move the puck as well as those two get snuffed out of the NHL before they get close to making it.

To conclude: if I’m Jake Gardiner’s agent, I bring these numbers up when I’m negotiating my client’s contract this summer. Once the Leafs get rid of Randy Carlyle and start to play like a real hockey team instead of one that relishes being caught in its own end, Gardiner’s possession and scoring rates will probably some of the best in the league. If I’m Brendan Shanahan, I’d recognize this and try to sign Gardiner on a long, cheaper deal now that will pay Gardiner slightly more in the first year or so, but eventually provide far more value on the back half. He’s going to be really, really good some day. Don’t be surprised when it happens. If he’s standing out this much now, imagine how much he’ll stand out when he’s on a good team.