The advanced stats of ‘goons’

I hardly know where to begin with this.

I know is that Cam Janssen said some pretty awful things, but misogyny and homophobia in hockey aren’t anything new. The last thing that the You Can Play team wants is to turn their campaign into some sort of witch hunt, so complaining about Janssen for the horrible things he’s said is counter-productive.

If you follow my Twitter account, last night I posted a few facts about Cam Janssen and the role he brings to the table. Nothing. He’s your prototypical fourth line winger, who is probably below replacement level, and his coach knows it. I bet until yesterday, you didn’t know that Janssen was on New Jersey’s roster for their Stanley Cup run. He didn’t play a single playoff game.

I’m not necessarily anti-toughness, but I’m firmly in the “anti-goon” camp. I just don’t see the role Janssen plays in hockey as having too much of an impact on a hockey game or a hockey team. You could say that teams with players with 10 or more fights this last season averaged 91.4 points, to teams without the 10+ fights guy having 92.6. It’s not a huge, huge difference, and the other thing to take into consideration is that teams with two or more “goons” averaged 94.2 points, largely on the success of the New York Rangers and Boston Bruins.

(It does, however, make for the second straight year that teams without a fighter eclipsed teams with a fighter in points)

I don’t particularly think that fighting causes losing. I just don’t think there is too much use for it. Fourth liners realistically play such little ice time that they’re all going to be slightly below or above replacement level.

Besides, not ever player with 10+ fights was a fourth liner. Wayne Simmonds scored 17 goals at 5-on-5 last season and played over 1000 minutes. Derek Dorsett and Chris Neil each played over 11 minutes a game, and Dorsett has the capability of being a pretty good defensive forward. He tread water last season against tough competition (his average opponent on-ice had a relative Corsi rate of .956) and started 348 shifts in the defensive zone to 198 in the offensive zone. He wasn’t all-world, but he was decent.

There’s a few other reasons for these players to exist, and they happen to be good hockey players who happen to fight. Gregory Campbell is a alright defensive forward and Matt Hendricks is a shootout specialist. It’s when you get into guys like George Parros or Ryan Reaves or Tim Jackman that you think “uh, what’s this guy doing here, again?”

Here, I made a handy chart of these players’ advanced stats. I looked at their goal-scoring ability (looking at even strength goals per 60 minutes) possession ability (using Corsi) and their usage in quality of competition and offensive zone starting rate. It’s not exact, but I combined all the numbers to give us a general picture of your average knuckle-chucker. No real scientific purpose, but it’s mid-July and I’m curious:

Name GP TOI TEAM G/60 Corsi On Rel QoC Ozone%
Kyle Clifford 81 727.52 L.A 0.41 3.55 -0.960 50.0%
Ryan Reaves 60 389.08 STL 0.46 7.86 -1.382 62.0%
Chris Neil 72 797.45 OTT 0.83 3.84 -0.353 54.5%
Jamal Mayers 81 660.57 CHI 0.54 -2.18 -0.930 45.8%
Derek Dorsett 77 920.63 CBJ 0.59 -9.12 0.956 35.0%
Mike Brown 50 430.57 TOR 0.28 -5.99 -0.175 41.4%
Tim Jackman 75 655.07 CGY 0.09 -2.11 -1.648 49.3%
Matt Hendricks 78 839.43 WSH 0.29 -5.72 0.725 44.2%
Brandon Prust 82 836.03 NYR 0.22 -11.34 0.153 33.7%
Wayne Simmonds 82 1017.23 PHI 1.00 0.18 0.065 57.7%
Cody McLeod 75 533.87 COL 0.67 -2.47 -1.172 54.9%
Matt Martin 80 837.35 NYI 0.50 -8.81 0.166 43.5%
Zac Rinaldo 66 489.27 PHI 0.25 -6.87 0.303 47.6%
Gregory Campbell 78 842.37 BOS 0.57 -9.76 -0.492 42.2%
Stu Bickel 51 514.38 NYR 0.00 -5.13 -0.890 51.9%
Zenon Konopka 55 375.5 OTT 0.32 -7.19 -0.755 48.9%
Shawn Thornton 81 737.32 BOS 0.33 -6.92 -0.674 51.9%
Jared Boll 54 434.7 CBJ 0.28 -9.52 -1.200 49.3%
Krys Barch 51 372.75 FLA 0.32 -9.18 -1.570 50.9%
Chris Thorburn 72 691.55 WPG 0.35 -17.7 0.631 38.8%
Michael Rupp 60 386.98 NYR 0.62 -16.28 -1.279 43.2%
Brad Staubitz 62 403.1 MTL 0.15 -20.99 -0.998 44.9%
George Parros 46 293.27 ANA 0.20 -15.34 -0.925 61.3%
Total 0.44 -6.27 -0.383 46.7%

[Data obtained via Behind the Net]

Your average player with 10+ fights played just over 9 minutes a game, was -6.27 Corsi with a 46.7% offensive zone start rate against pretty soft competition. Since they played enough games (Janssen didn’t) to qualify for this list, they’re all pretty much regular players, so most found some sort of usage role in the lineup.

It wasn’t particularly the toughest job. Few were counted on to score. On average, the “goons” had the regular season scoring ability of Dale Weise or Dustin Penner. Possession-wise, they had the abilities of Eric Nystrom.

I’m sure if you took a detailed look at most other fourth liners in the league, you’d come up with something similar. The point isn’t that “fighters are bad players, fighters will lose you games,” it’s that “specialized fighters are meaningless in today’s NHL”. Teams that dressed a regular forward with the jersey number of 91, for instance, averaged 79 points and all other teams averaged 93.4.

That stat shouldn’t make Tampa Bay want to trade Steven Stamkos or Long Island John Tavares. It means that if a guy is going to fight, you want to assess his individual worth. Cam Janssen has no value because his team doesn’t play him enough to be considered a “regular fighter” in the NHL. Certain players like Brandon Prust and Zenon Konopka do. They do something that made playoff teams want to keep them in the lineup.

Because this doesn’t look good on an organization teetering towards bankruptcy: