Figuring out why Braydon Coburn was so bad in 2013 (with advanced stats!)
Braydon Coburn has long been considered a solid, consistent defender and has been a fixture on the Flyers’ top-two defensive pairings since he was acquired at the 2007 trade deadline. So consistent, in fact, that during the 2011-12 season, the Flyers jumped at the chance to extend his contract for 4 years at a healthy $4.5 million-per-year price tag. Hardly a small commitment for a team that is perennially bumping its head against the salary cap.
But Coburn’s 2013 season was – and I’m being kind here – a break from the mold.
Oh who am I kidding? It was awful. In order to set up the second half of my analysis, I’ll spend the first half explaining just how awful it was.
Here is a quick recap of Coburn’s career in terms of the standard statistics:
WARNING: Minors, the elderly, and those living with a heart condition may wish to avert their eyes from the row marked ‘2013’
|Scoring Stats||Goals||Assists||Ice Time|
Those numbers look bad. Really bad. But the season was shortened to start and Coburn was on the DL for 1/3 of those games, so it may help to break down his 2013 stats in terms of ice time.
Stats per 60 minutes of ice time (with career rank):
Goals/60: .08 (2nd worst)
Assists/60: .322 (worst)
Points/60: .402 (worst)
Plus-Minus/60: -.804 (worst)
PIM/60: 3.3 (2nd worst)
Shots/60: 3.06 (worst)
Even Strength Points/60: .402 (worst)
Yuck. How about a nail in the coffin? In the 33 games in which Coburn played, the Flyers went 13-17-3, averaged 2.64 goals for and 3.12 goals against. In the 15 games in which Coburn did not appear, the Flyers were 10-5-0, averaging 3.07 goals for and 2.53 goals against.
Now, I already trashed Coburn’s lockout season a month ago, but I swear I’m really not trying to pile it on. I am simply attempting to provide some perspective. I went into this study because I was genuinely baffled that such a dependable player could just roll over and die at the prime age of 28. How was this possible? Outside of my private suspicions that Dan Bylsma spent his off-hours poking needles into a Braydon voodoo doll somewhere in the bowels of the CONSOL Energy Center, there was seemingly no explanation for such a precipitous fall…
Unless, of course, you like advanced stats. I’ve recently shown my ardent support for the proliferation of advanced statistics, and this seemed like a ripe opportunity to break them out. As it turns out, the fancy numbers indicate that it wasn’t simply poor luck or bad mojo that led to Coburn’s bad year.
There was another reason:
More specifically, it was Peter Laviolette’s deployment of Coburn. Braydon was not put in a position to succeed.
Before I go any further, I should note that the basis for the remainder of my analysis is a statistic called Corsi and a few variations thereof. It’s a +/- metric that records ‘shot attempt differential’ rather than ‘goal differential.’ It’s used as a proxy for possession (as in, the more your team has the puck, the more shots they will attempt, and vice versa). It is an especially helpful metric for evaluating defensive-minded players that generally don’t score a lot of points. Click here for quick definitions of Corsi, Corsi Rel (Relative Corsi), and Corsi Rel QoC (Quality of Competition). To read about them in depth, click here. For an eloquent defense of advanced hockey stats, read this. For evidence that shot differential is a leading indicator of Stanley Cup success, check this out.
Coburn’s Corsi numbers, like his standard numbers, were terrible. He registered a Corsi rating of -7.22, by far the worst regular season score he’s posted since BehindTheNet started tracking Corsi in ’07-08. (That is, the opposing team averaged 7.22 more shot attempts than the Flyers for every 60 minutes that Coburn was on the ice. Got it? Good. I’m done explaining things now.) Even worse Coburn registered a Corsi Rel of -7.9, 7th on the team and ahead of only 3 other players – oft-scratched Kurtis Foster, waiver-wire pickup Kent Huskins, and the perpetually injured Andrej Meszároš.
But there is an important mitigating factor that could have contributed to his terrible Corsi. Laviolette used Coburn as the primary defender in the Flyers’ defensive zone. Coburn only started shifts in the offensive zone 40.5% of the time, the lowest mark on the team among defensemen, and the lowest mark of Coburn’s career. Of course this usage would skew his Corsi numbers against him, as it’s far easier for the opponent to get shots if they take the faceoff in the Flyers end. Still, Coburn’s D-zone deployments aren’t solely responsible for his dip in production.
Despite the strong slant towards defensive responsibility, the effect is slightly mitigated by the fact that Coburn faced more-or-less mediocre opposition throughout the season (.416 Corsi Rel QoC, lower than Timonen, Schenn, Grossman, and Lauridsen). For reference, Coburn performed extremely well against much more difficult competition in 2012 (1.079 Corsi Rel QoC), so it’s not as if he’s never been capable.
So what else did Lavy do to mess with Braydon’s game? He paired Coburn with Nicklas Grossman for the whole year. Of Coburn’s 586 minutes at 5v5, Grossman was on the ice for 250 of them.
Mind you, I’m not saying that Grossman is an especially bad defender. Yes, he’s slow and he gives you almost nothing as far as offense, passing, or puck handling is concerned, but he’s willing to give up the body, block shots, hit, and generally does a decent job with all of that. The problem is that Coburn is built from the same mold, and when the two are together, it’s an unmitigated disaster.
Corsi For % (CF%) is an adapted form of Corsi where you express the numbers in terms of the proportion of shots directed at each net. Using the With Or Without You (WOWY) breakdowns from HockeyAnalysis, it’s apparent that the Coburn-Grossman duo was doomed. In 2013, when together, their CF% was a hideous 45.8%. When Coburn was apart, his CF% was 48.6%, and Grossman while apart scored 48.8%. It wasn’t just that Grossman held Coburn back, but rather it seems playing together made them both significantly worse. (This also happened in 2011-12, when their 202 minutes together yielded a CF% of just 44.7%, compared with Coburn’s 51.4% while apart and Grossman’s 45.7%. Seriously, these guys should never play together.)
On it’s face this makes some logical sense. When paired with Grossman – who is pretty good without the puck but basically a pylon with a stick while in possession – Coburn was forced to take on the responsibility of making breakout passes and trying to start the rush. Coburn is an average-to-below-average puck handler for a defenseman, and with no help from his partner he was often cornered or rushed into making bad/dumb plays. This poor pairing certainly contributed to Coburn’s 31 turnovers for the season, most on the team among defensemen, and especially damning since he only played 33 games.
It seems apparent that if he is to excel, Coburn must be paired with a defenseman who is at least a decent puck mover. In 2013, Coburn’s 2nd most frequent pairing was with Bruno Gervais (185 even-strength minutes on ice together). Gervais is not the most excellent defensive defenseman to ever live, nor is he really great at offense, but at the very least he has some sense of what to do with the puck when in possession. Together their CF% was 49.4%, Coburn apart was 46.6%, and Gervais apart was 47.4%. So they were better together than they were otherwise, which is certainly a more positive result than I expected to find.
Because of his contract, and after this awful season, Coburn likely isn’t going anywhere unless the Flyers are willing to sell him for peanuts. That would be a mistake. So while he’s on the team, history suggests that would be in everybody’s best interest if he is always paired with a puck-mover.
Laviolette’s best bet for success may be to pair Coburn with Timonen again. The two played over 2500 minutes together at 5v5 from 2008 to 2013, and both had improved Corsi numbers when they were paired. Their CF% together was 52%, compared with 49.4% for Coburn apart and 49.7% for Timonen. But this pairing would not be viable if Kimmo is asked to play on the top line with Luke Schenn again next season.
If he were healthy, Andre Meszároš would be also a good option (although his injury concerns have only gotten worse over the last two years). They were only paired at 5v5 for 147 minutes from 2010-13, a very small sample size, but again Coburn and Mez were better together (53.1%) than they were apart (49.3%, 49.5% respectively).
But perhaps Lavy’s best move would be to put Coburn with Erik Gustaffson. The 24-year-old Swede has only played 52 even strength minutes with Coburn in his career, but he profiles as the same kind of defenseman as Timonen and Mez and has improved every time he was called up from the Phantoms. Gus is a restricted free agent and very likely to play the whole year for the Flyers in 2013-14, he’s got better upside than Gervais and has none of the injury concerns of Meszároš. Pairing the youngster with Coburn would allow each to focus on his respective strengths – Gus’s puck handling and passing, Coburn’s defensive responsibility – and could be the best-case scenario for a team looking to improve its defense in a year when free agency is nearly devoid of top talent.