clock menu more-arrow no yes mobile

Filed under:

Player Types and Free Agency

There is a lot of work on the internet now classifying types of players and aspects of their games. I look at those attributes in the scope of impending Free Agency.

Arizona Coyotes v Washington Capitals Photo by Patrick Smith/Getty Images

As free agency rapidly approaches us, I thought it would be a good opportunity to investigate what quantifiable deficiencies the Devils have and what players available could help us. I’ll be utilizing data from Ryan Stimsons Player Type work as well as XtraHockey’s positional skater ratings to identify solutions for our squad accessible through free agency.

XtraHockey and Team Ratings

The Devils need help in a lot of areas. The only place the Devils are performing well is in the defensive play from forwards where we are tied for 5th in the NHL. This is according to XtraHockeyStats Team Ratings.

FGS = Forward Goal Scoring: The ability to score goals of the team's forwards.
FPA = Forward Passing: A ability to generate offence with passing of the team's forwards.
FOFF_OV = Forward Offensive Overall: Overall offensive ability of the team's forwards.
FDEF_OV = Forward Defensive Overall: Overall defensive ability of the team's forwards.
FOV = Forward Overall: Overall rating including Offense and Defense of the team's forwards.
DGS = Defence Goal Scoring: The ability to score goals of the team's defencemen.
DPA = Defence Passing: A ability to generate offence with passing of the team's defencemen.
DOFF_OV = Defence Offensive Overall: Overall offensive ability of the team's defencemen.
DDEF_OV = Defence Defensive Overall: Overall defensive ability of the team's defencemen.
DOV = Defence Overall: Overall rating including Offense and Defense of the team's defencemen.
GOAL = Goalie Rating: Overall rating of the goalies used by the team.
TOV = Team Overall Rating: Overall rating of the team including goalie, defence, and forwards.

Our forward play in general is well below-average, but it is HHoF-material in comparison to the play from our defenders. The numbers may be 1 spot off in the chart because, as I realized afterwards, XtraHockey included Montreal twice. So the Devils were actually 30th (aka last) in 5 of the 7 defender categories and were second to last overall. We are middle of the pack on defensive play from defenders but way ... WAY behind the pack in offensive contributions. This needs to be viewed as the chief concern going into free agency.

Ryan Stimsons Player Type Work

The pride and joy of AAtJ, Ryan Stimson’s most recent venture has been a foray into “player types.” He used a k-means clustering analysis to figure out what the statistical profile of different types of players is with regards to shot ratios, shot-generating passing proficiency, and transition game. I downloaded and parsed some of the data here and displayed below the important players from the Devils.

Build Up Index (secondary and tertiary shot assists)
Danger Index (Shot Contributions from below the end line or across the slot)
Influence Index (Total shot contributions)
Pass Index (Total shot assists)
Shot Assist Index (primary shot assists )
Shot Index (individual shots)
Transition Index (controlled entry assists)

Okay so what can we learn from this? Well, Taylor Hall is far and away the best player on the Devils team. Does it count as learning if you already knew it? Probably not. So what have we actually learned? Well, with regards to forwards, we may actually have a serviceable core in the works. Henrique and Zajac are predictably balanced players. Hall, Zacha, and hopefully at least one of McLeod of Hischier will be playmakers. Palmieri’s profile is actually quite close to that of “shooter” but his passing stats bumped him down to “dependent” instead. Bennett ranked unsurprisingly high as advanced stats continue to favor him — making the decision not to qualify him troubling. All said, if the established veterans maintain their production and the young guys improve along current projections, we have the making of a good forward group in place. It could even be great if some unexpected/ascending prospects or recent draftees like Boqvist, Big Poppy, Quenneville, Anderson, Wood, Blandisi end up becoming a indepedently and consistently valuable. That said, we are certainly currently behind in general and so another playmaker would help and, since we don’t seem to want our only shooter, maybe one of those would be nice too.

But that brings us to the much bigger problem which is defense. All of our defenders except for John Moore are labeled as “defense-oriented” which is the worst classifier to have. Worth pointing out though that Severson is decent in these metrics in general and could easily ascend to better types with continued improvement and John Moore has his own problems.

But, we desperately need an elite defender. Any other type of defender would be good, but based on Ryan’s work it seems an all-around or a volume shooter would be best — we’d settle for a puck-mover.

What UFAs Can Help Us?

So are there any solutions in free agency? I cross referenced the free agency list from CapFriendly with the player type data, and was able to produce a filterable chart. The settings I’ve chosen for this article are as follows:

  1. Less than 35 years old — I want a minimum 3-year deal and production seems to pass the point of no return around 37.
  2. Time on Ice (17 minutes/GP for D, 12 minutes/GP for F) — no bottom pairing defenders or 4th line forwards.
  3. Defenders must by Volume-Shooters, Puck-Movers, or All-around. Forwards must be Playmakers or Shooters — These are the player types that address our offensive needs as seen in our roster types and our XtraHockey ratings.
  4. At least 10 GP in 2016-17 — should be based on recent data and Stimson’s data needs to have been able to classify them (that might need even more GP).

Those four restrictions dwindle the desirable list from 321 UFA players down to 11.

Okay so let’s get into it. There were no all-around defenders available. You just can’t get a stud #1 d-man in free agency. There are only 4 defenders that meet our restrictions at all. I’ll show you their player type charts in comparison to our current top 2 defenders — Severson and Greene. One is obvious...

Kevin Shattenkirk

If you’re looking for a more cost-effective, yet slightly older, option -- Trevor Daley is similarly proficient in the measured categories to Shatty.

Trevor Daley

And honestly, that might be just about it for defenders. Michael Stone is classified as a “Volume Shooter” and he’s 24, but he’s bad at most other things and guys that already have seen him don’t seem to want him back. Dennis Wideman fits the bill, but he’s just at the end of the acceptable age range at 34.


So let’s look at the forwards. I’m going to rule out Hemsky for due to health and achievement issues. I’m going to put Drew Stafford, Brandon Pirri, and Beau Bennett all into a bucket together as possible 3rd line type guys that could be interesting. Bennett is the best of them though, and we didn’t qualify him as an RFA so I’m not going to spend too much time analyzing these other guys.

Versteeg, Radulov, Hanzal

What’s left is Kris Versteeg, Alex Radulov, and Martin Hanzal. All are playmakers, all can help us immediately, and none are over 31.

According to Darren Dreger, a dozen teams have contacted Versteeg so don’t get too excited about hat price point because it’ll go up. Hanzal has probably retained value after performing well with the deadline-sweepstakes-winning Wild. Radulov is the big-money guy here. He’s looking for a long-term deal and it’ll be a pricey one. Because of the amount of buzz surrounding our draft and prospects, I wouldn’t be surprised to see Shero pass on every one of these guys.


John was right. We should push to sign Shattenkirk. I’ve actually been a bit of a flip-flopper on this personally because I think someone is going to have to overpay him and I don’t consider him a clear #1 defender — though he is a top pairing defender and if you think he isn’t then you’re either not paying attention or you’re just being a contrarian. He has finished 8th in all-star voting and 14th in Norris voting 3 consecutive years, his WOWYs paint him as a generator of possession, and he has logged an average of 21:17 minutes per game over his career on a Blues team that has been amazing over that span. To imply he isn’t one of the 60 best defenders in the league is to live under a rock. DTMAboutHeart’s metrics rank him at 27th among defenders in WAR and 35th in XPM (expected plus-minus) which sounds about right to me — a tail-end #1 or top-end #2 defender.

Everyone else (including the aforementioned forwards, IMO) is a consolation prize.

Your Thoughts

Is there anyone you think I glazed over too fast on the list? Is there anyone left of the list entirely that you feel was mischaracterized or a victim of too-strict filtering rules on my part? Do these findings confirm your perception of the team, or are you surprised that we have/lack certain types of players?

Thanks for reading and leave your thoughts below!