Jeff Petry signed a one year contract today that was a pretty good deal for the Oilers, I think. Petry will make $3.075M, which is so low that I kind of wonder if he wouldn’t have been better off going to arbitration to try and bump the number up by a few hundred thousand dollars. That said, the NHL’s arbitration process, particularly for defencemen, is a mess because there’s not widely accepted data that is used. It’s an open question whether or not you can use things like Corsi and Fenwick in an arbitration and even then you’d have to educate the arbitrator in it. I can understand why you wouldn’t want to fight an arbitration on an issue without precedent over a few hundred thousand dollars.
A Maple Leafs fan on Twitter pointed out someone who might be the real beneficiary of Petry’s contract: Cody Franson.
Petry, with similar ice time to Franson and half as many points last year, just signed a 1-year deal for $3.1M. Good news for Franson.
— Draglikepull (@draglikepull) July 7, 2014
A couple of observations about this. First of all, the difference between Franson and Petry in points this year is pretty much entirely due to the power play. Petry played about an extra 140 minutes of ES TOI; they each had 15 points at 5v5. As has been pointed out to me by Twitter’s @tsetse_fly, this was an unusual season for Franson – he got points on 31.1% of the 5v5 goals that the Leafs scored when he was on the ice, down from his 44.5% from 2009-13. That’s an extraordinarily high number for a defenceman. Petry, in comparison, had recorded points on 31.7% of the 5v5 goals that the Oilers scored when he was on the ice entering this season and pretty much matched that this year with a 28.6%.
As I’ve mentioned before, I’m kind of uncomfortable with points as a measure of a defenceman’s 5v5 offensive contribution. Over the past five seasons, Franson has posted 1.08 5v5 PTS/60. Petry’s posted 0.67 5v5 PTS/60. That’s a 61% edge for Franson. If you look at the shot attempts, Franson’s teams have generated just 7.4% more shot attempts when he’s on the ice. The shot attempts against are virtually identical.
Franson’s teams have scored 20% more goals when he’s on the ice than Petry has but obviously there’s a shooting percentage edge there, given the narrow difference in terms of shot attempts for. If you don’t believe that defencemen drive shooting percentage – and the evidence in support of that is pretty strong – then the edge in goals is probably just chance. So that 61% edge in points, what does it actually amount to in terms of offence? I’m not sure it’s very much but putting up points gets you paid.
Leave that aside though – I wanted to talk about the power play points. That’s where the scoring difference lies between Petry and Franson this year. Petry had 2 points on the power play (in 67:23) and Franson had 19 (in 230:03). At 5v4, which I’m going to focus on, it was 15-2 for Franson. If you look at it as a rate stat, which we should, Franson produced 4.0 5v4 PTS/60 while Petry produced 1.9 5v4 PTS/60.
I’ve been kind of beating the “points aren’t offence” drum for a while so I thought it’d be cool to put together a little two minute video of Franson’s 5v4 points this year. For those of you who aren’t familiar with him, he wears number four shoots right – he’s generally on the right side of the ice in these clips. None of these happened to be goals – if you can’t tell if Franson’s shot was tipped, it was. (The goal at 1:02 is pretty awesome.)
The most obvious takeaway here is that Phil Kessel is awesome. That’s sort of the problem with points though – it’s an arbitrary way of allocating credit for offence. I suspect that it works a lot better for forwards than it does for defencemen. How much of the offence that we’re seeing here is offence that Franson is creating, as opposed to offence that he’s being credited with by virtue of having a great net front guy in JVR and hockey’s chubby genius Phil Kessel on the ice with him?
That’s a harder question to answer. Then, you can lay the structure of the power play on top of that. Does the Leafs PP generate more touches for defencemen then usual? I have no idea. It will influence how many points a defenceman puts up though.
The correct way of evaluating Franson’s offensive contribution would involve understanding how everything he does changes the probability of a goal being scored. When the NHL starts cranking out the SportVU data, that’s going to be much easier to do. If we learned, for example, that Franson’s pass selection and shot selection was identical to that of Jeff Petry in the same positions on the ice, we might conclude that they’ve got equal offensive skill and that it’s external factors making them look different. Or we might see that there’s a real difference with Franson. For now, we’re left to draw inferences based on point totals and I’m skeptical that this is a realistic way to do it.
None of this is intended as a criticism of Franson. I like puck moving defencemen with positive possession numbers and he’s a puck moving defencemen with positive possession numbers who has continued to post positive possession numbers as his role has expanded. His boxcars give him a better arbitration case than Petry though and I’m not sure that that makes a lot of sense. It will be very interesting to see what his number comes in at.Email Tyler Dellow at firstname.lastname@example.org