Public Service Announcement: Damon Severson is the Devils’ Best Defender

I don’t think this is something that should really have to be said, but Damon Severson ought to be in the lineup for the Devils every night that he is healthy. Why have I decided to address the matter now? Well, the situation has gotten more dire in recent days, and someone needs to say something before it gets out of hand.

The Situation

The reason I’ve determined this needs to be said is, after Hynes’s decision to sit Severson last night, I stumbled on a Facebook poll the other day that caught my eye.

Oof. That hurt me. The reason I put this at the top of the article is not because I take any issue with the individual that posted it, the group it was posted on, or those who voted in it. I am a member of this particular group because the fan part of me finds the content enjoyable. The analyst part of me, though, has concerns. The reason I put it at the top is because I am not of the illusion that I have the power to significantly impact coaching decisions. Hopefully, though, this blog and community — all of us, not just the writers —- do have at least marginal influence on how some of the players are perceived by the public, especially given that we are among the public. And any sample of 100 fans that agrees Andy Greene is the worst defender on the team and Damon Severson is the next worst, likely doesn’t utilize the same data we here at AAtJ typically do. So here’s a healthy dose of that information.

The Statistics

This data is from Natural Stat Trick which has been mentioned by Ken Daneyko on the live Devils broadcast so it is edging dangerously close to mainstream. The stats below are 5v5 only and are adjusted for score and venue.

If you’re a regular reader, and/or know what all the column names in that chart are, feel free to skip past the next paragraph and go to the asterisk (*). For those reading this article who are unfamiliar with the statistics used here, this is a breakdown:

For all the % columns above, 50.0 is league average — above is good, below is bad. CF% is “Corsi For%”. It tells what percent of shot attempts (shots on goal, blocked shots, and missed shots) come from the Devils while a given player is on the ice. It is often used as a proxy for time of possession. FF% is “Fenwick For%” and it’s the same except only unblocked shot attempts. SF% is the same but only using shot attempts that reach net — in other words, your friendly neighborhood shots on goal stat. GF%, “Goals For%” is the percent of goals that were scored by the Devils while the player was on ice — it’s pretty similar to plus-minus without some of the silly quirks, which I’ve addressed here at AAtJ -- oddly, also in an article about Severson. SCF% and HDCF% and “Scoring Chance For%” and “High-Danger Chance For%” and they each take the shot attempts that occur in dangerous and very dangerous areas (seen here) and calculate the percent of them that were recorded by the Devils when a player was on ice.

*

Okay, phew. So now that we’re past that, let’s get to how Severson ranks as of this writing. Among Devils D, he’s 1st in CF%, 1st in FF%, 1st in SF%, 4th in GF%, 2nd in SCF%, and 2nd in HDCF%. A reminder that this is the player that we just sat for Ben Lovejoy. This means that the Devils are producing a larger share of shots with Severson than with any other defender, and a larger share of scoring chances than with anyone except Will Butcher.

That could just tell us how other players are doing when he’s on the ice though. How do we know he is personally impacting these numbers? Well Corey Sznajder is tracking some additional statistics that can give us a window into other aspects of his game (Also Prasanth Iyer & Co. over at Hockey-Graphs have created an awesome DIY tracking applet if you want to verify findings at home). For instance, in tracked games as of this writing, he has been the hardest Devil to gain the zone against — according to my compilations, he’s broken up a higher percentage of entries and allowed carry-ins at a lower rate than any other defender on the team. He also leads all Devils defenders in how frequently he is able to carry the puck into the offensive zone. He is also the most prolific shot generator on the team — he leads in shot assists any way you slice it. The only area he isn’t among the best on the team in exiting the zone -- where he’s average.

Edit: I’ve since made a player-comparison version where you can find all these stats on one screen here.

The Narrative

So why is it that, if Severson leads the Devils in so many important statistical categories, he is so maligned by the staff and, apparently, a subset of the fanbase? Well, I think it has to do with some biases involving high-profile events. By “high-profile events” I mean stats that are low in frequency, but easy to see and identify as a good or bad thing. The “Ugh, that was bad.” moments.

Turnovers — particularly in your own end — are a good example. Severson leads Devils in turnover rate which is definitely a bad thing and probably part of why his exit numbers are low, but it’s probably perceived as a much worse thing than it is. It’s a really easy thing to see, ESPECIALLY when it leads to a chance or, worse, a goal; or, worse still, a game-winning-goal. But here’s the reason that’s unimportant. His team-high turnover total is only 23. That’s 23 events in 37 games. Turnovers are really ugly when they happen, but in the aggregate they just don’t matter enough. They are rare and don’t even always result in a shot attempt for the other team. Case in point, I had to ask twitter for help to get those linked clips of moments that mattered — and one of them was from over a year ago. Furthermore, he leads Devils defenders in takeaways by more than he leads in giveaways. His differential is only -8 which is middle of the team. But since he’s a defender, all a takeaway does is prevent a chance rather than produce one so it’s harder to “see.” That’s 8 more giveaways than takeaways in 37 games, remember those figures.

Another thing working against him is special teams. And this isn’t captured by my fancy chart above or colorful viz because those were only 5v5 situations. He doesn’t kill penalties, and regardless of total performance, as long as you’re not making really ugly mistakes, you tend to get street cred for just being a penalty killer. He was on the powerplay before being removed for Vatanen — though I think he was doing just fine there. And, possibly most importantly, he leads the Devils D-men in penalties taken. It’s easy to blame someone for a penalty taken — he’s the guy that’s in the box. What isn’t as easy to square with memory though is penalties drawn, because nothing happens to that guy after the moment of the penalty. When you account for those equally important events though, he’s team-average in penalty differential — Butcher, Lovejoy, and Santini are better, Greene, Moore, and Vatanen are worse. And the difference between the Severson and the best guy in that category is worth about 3 penalties so far — in other words 1 every 10 games. A powerplay on average is worth just under 2.8 shots. So that’d be around 10 shots to this point in the year. Remember that number.

To put those last two stats in perspective, the Devils have outshot their opponents by 25 attempts with Severson on ice. That differential is 4 higher than Will Butcher, 30 higher than Sami Vatanen and Ben Lovejoy who haven’t even played a full slate of games yet, 34 higher than John Moore, 61 higher than Mirco Mueller, 186 higher than Andy Greene, and 224 higher than Steve Santini. You see why those couple bad giveaways and 3 extra penalties don’t matter much now?

This brings us to the final misconception, which is defenders are there to prevent goals. That’s not what defenders are there to do. EVERY hockey player’s job is to make it more likely that his team is going to score than the opposing team — defenders aren’t special. Severson is marginally worse than his peers at preventing goals, but significantly better at gaining the zone, creating shots, and putting the Devils in a position to score goals. And that makes him still the largest net positive among our blueliners. When all is said and done, the Devils as a team are at their best with Severson on the ice. Severson, good and bad, with all of his setup passes, giveaways, carry-ins, penalties, entry breakups, etc., isn’t perfect, but he’s the best defender on this roster.

Thank you for reading. If you still disagree, feel free to voice your displeasure with myself and the numbers in the comments section. Until next time, AAtJ!