Young, puck-moving defencemen gettin' paid!

Cam Charron
August 04 2014 11:45PM

The NFL season is due to start soon, which is a good thing considering my manager at work* changed me from the Sunday weekend shift to the Saturday, meaning I get Sundays off.

That doesn't seem like an appropriate header for a hockey post, but all these posts wind up having a point a little further down. Please continue reading this blog post, and if you don't enjoy it, I'll refund you all the money you spent purchasing access to this website.

* - Please change my Internet nickname from "Known Girlfriend-Haver" to "Known Job-Haver"

Patrick Peterson of the Arizona Cardinals recently signed an extension to make him the highest paid defensive back in the league, which has provoked a bit of a debate around who is the best coverage man in the league.

The argument is interesting since it relates to hockey somewhat. Statistically, the honour falls on Richard Sherman of the Seattle Seahawks. Quarterbacks rarely target Sherman, they have lousy completion percentages when they do, and passes are far more likely to wind up in interceptions than touchdowns. This isn't so with Peterson or Cleveland's Joe Haden, who is also in the discussion.

The caveat is that Sherman doesn't always line up against opponent's number one receivers like Haden or Peterson do. This is a bit of a familiar argument: Internet people, particularly Toronto Maple Leafs commenters, like to point out that Jake Gardiner's high Corsi number shouldn't mean as much as we make it out to be because Gardiner doesn't play against tough forwards.

If this football stuff interests you, the best post I've found on the Sherman-Peterson debate is from Pro Football Focus, published all the way back in May:

One of the key talking points at the heart of the issue seems to be what a “No. 1 corner” should be doing in terms of his assignment. There is a sentiment in some quarters that a truly elite corner should be tracking an opponent’s No. 1 receiver if he wants to be considered among the league’s elite. Others feel that locking down one side of the field on the quarterback’s open side is just as merit worthy.

The article's conclusion?

In terms of production and performance Sherman clearly separates himself from the other two with the only counter argument available that Haden and Peterson operate in more varied and difficult assignments. However, is the performance level close enough for that difference in assignment level to bridge the gap?

In my opinion, no.

The article was written by Ben Stockwell, and his conclusion should be applied to hockey somewhat. I've already taken a brief look at how both Gardiner and Phaneuf play against the elite forwards of the Eastern Conference and while comparable, the Leafs generate that extra shot per 60 minutes of play with Gardiner on the ice, slightly tilting the needle in favour of Toronto.

The conclusion can also be drawn when looking at the Maple Leafs division rivals. A few stodgy hockey types no doubt had to polish their monocles after seeing that one Pernell Karl Subban signed for an eight-year, $72-million contract out of his arbitration. Subban is a bit like Gardiner, in that he's a defenceman whose defensive miscues are overstated because the vast majority of the time, he's doing things that help his team hold onto the puck. Like Gardiner, Subban played on a lousy possession team, and like Gardiner, Subban didn't play against the toughest minutes.

Extra Skater has a quality of competition ranking that weights the ice-time a player played against the ice-time of opposing forwards. The definition is fairly technical so I'll spare you, but essentially the higher the number, the tougher the competition. Here are Toronto and Montreal's top defencemen in terms of forward quality of competition, our two flashy offensive defencemen, and their respective Corsi percentages:

Avg F TOI Corsi %
Josh Gorges 26.4% 45.2%
P.K. Subban 25.8% 49.9%
Avg F TOI Corsi %
Dion Phaneuf 27.5% 40.8%
Jake Gardiner 24.5% 46.4%

In either case, I don't think that the gap in Corsi can be explained solely by quality of competition, especially not for Montreal. Even if a team is going for a hard matchup, keep in mind the opposition is doing its best to avoid that matchup, so you do wind up with a lot of players playing similar opponents for 41/82 games a season on the road. At home, a coach like Randy Carlyle who really likes to match his lines will see more separation in competition between his top and bottom pair. In the end, the Gardiner pairing got results and the Phaneuf pairing didn't, which tells you something about the overuse of Phaneuf, the on-ice tactics, or as a last resort, that Gardiner is a plain better defenceman than Phaneuf.

If our option is the latter, than Toronto got a hell of a deal on a very good defenceman for the next five seasons. I wouldn't put Gardiner in Subban's class, but I'd say he has a good shot at joining that second tier. Quality of competition, in my estimate, should be available if we want to settle a tiebreaker between two very close Corsi (or other preferred statistic) score but it shouldn't have to lead the conversation. That's the conclusion the smart football people seemed to come to and one I think we'd be smart to follow in hockey.

63811cbf517d2d685ea09e103488ea3a
Cam Charron is a BC hockey fan that writes about hockey on many different websites including this one.
Avatar
#1 SPYDER CANADA
August 05 2014, 09:05AM
Trash it!
17
trashes
Props
14
props

Overjoyed to hear that this isn't your day job.

Avatar
#2 Poluza
August 05 2014, 06:04AM
Trash it!
0
trashes
Props
6
props

Great article. I love reading about the Gardiner comparisons, and how he's great and all, but I'm also very interested in Morgan Rielly. I thought that Morgan seemed to be a bit more poised offensively, and I'm a lot more excited to see him when he's finished developing. Will he ever be considered an "elite" guy, in the Doughty/Weber tier?

Avatar
#3 deaner
August 05 2014, 06:12AM
Trash it!
1
trashes
Props
5
props

I like the football analogy to a degree, but it loses symmetry with respect to zone starts. In football I don't think there are "red-zone specialist" cornerbacks, where as the NHL has defenseman who are (rightly or wrongly) more zone-deployed based on skillset - and this would certainly influence shot differentials.

I have nightmares of Roman Polak getting like 80% DZ starts and getting hemmed in his zone for 5 mins straight so hopefully Leafs (Carlyle) smarten up and put skilled D out in all situations.

Avatar
#4 Jeremy Ian
August 05 2014, 06:30AM
Trash it!
0
trashes
Props
4
props

Interesting to watch the game change and adapt. But I wonder how much one can separate the Subban-Gorges/Gardiner-Phaneuf contrasts? It may well be that they are co-dependent, something perhaps even more glaring with teams with poor possession records.

The way you have told this story, Subban and Gardiner operate as discrete players, independent variables, at best affected by their partners. But what about their ties to the other pairing? If you took Phaneuf out of the lineup, or Gorges, how would this affect Gardiner or Subban? I have a feeling their performance is more dependent on the work of the other pairing. Certainly, we'll see in the case of Subban.

Here's another possibility. That Gardiner is more dependent on Phaneuf than Phaneuf on Gardiner. Given how Carlyle uses tactics, is Gardiner's record a "response variable;" Phaneuf's more of an "explanatory variable?"

I know, I know, a lot of people are going to say this is just a disguised defense of Phaneuf. But the point is, one might have to think about Gardiner's contract in relationship with someone like Phaneuf's. No Phaneuf contract, then Gardiner's starts to look less economical.

On a totally separate note, since you started with the football parallel, what do you think of the concussion issue? Malcolm Gladwell's predicted an imminent crisis in football (http://www.newyorker.com/magazine/2009/10/19/offensive-play

Just wondering.... Thanks for the great pieces and congrats on the job.

Comments are closed for this article.