The Caps tried that last year – they went without the enforcer. Their star player missed six games with a shoulder injury after an incident that likely would not have happened if an enforcer had been on the bench.

I see what you’re getting at, and it’s true as far as it goes. But, there’s an intangible there that a goon can provide, and that is heart and soul and energy, which is something the Capitals are short on. When a guy goes to war for you, you can’t help but step up your game. The stakes become higher. There was an incident last year in which one of their players got one-shotted in a dirty play. The Caps responded by taking that 9 minute PP they were handed and proceeding to torch the other team for three goals on that PP and four more later. My thought is that it’s that energy, that solidarity, that the Caps were looking for in this trade.

Only when you find a way to factor in this team-wide collective delusion into the stats can you provide an adequate value for a goon.

This really is the only argument in support of the “goon” (and to a slightly lesser extent fighting in general), and, since I am totally anti-fighting in hockey, I don’t buy this. Sure, once in a while a team may get a lift when a goon goes out and pounds another goon – but even then I have yet see a someone make a causation link between having a goon on the team and improved team play.

Sow if we take a goon off the 4th line and add a decent Nhl 4th line player, he will be able to do a bit more for the club with his on-ice performance. So having a goon not only costs the team a win, it also costs them the opportunity to have a player who might be able to get them another win.

Do these two year contracts come with the bright red nose that these guys should wear?

The funny thing with Sutter is: he signs a goon almost every season, the guy ends up being either a detriment or a non-factor….and then he does it all over again the next summer. It really is quite baffling. As is his predilection for really terrible goons (in terms of actual hockey skills).

Fidelity to convention before thought.

Although I completely agree with your overall sentiment (i.e. goons suck), I do think you’re overstating their impact on team performance. Firstly, I think that 0.2 GD per penalty is an aggressive figure that doesn’t leave much room for actual power play skill, but I’ll leave that aside for now. But:

The average goon plays 6.9 minutes per game and 51 games per year… The average NHL goon has a penalty differential of -6.43 per 82 games. As Desjardins showed in the linked posts, a minor penalty is worth about .2 goals based on the value of the power play. Using that information, we can see that the average NHL goon gives up 1.28 goals per season based on penalty differential alone.

If the average goon only plays 51 games in a season, then the average impact is only 0.80 goals per season. I think it’s important to remember that goons don’t generally play every night, thereby minimizing their opportunity to be a drag on the team. Further, a greater percentage of their ice time (and goals against) will be “low leverage” minutes when a goal for or against matters less. That’s awfully difficult to factor in to the analysis, but it’s another thing that will push their impact on standings points and wins down, even if it doesn’t show up in goal differential.

The average goon plays 6.9 minutes per game and 51 games per year. The average goal differential per 82 games is -5.55… the average NHL goon gives up 1.28 goals per season based on penalty differential… Using these numbers, the average NHL goon is responsible for -6.83 goals per season

Again, the overall number of -6.83 goals per season is predicated on the goon playing in all 82 games which is not the case for the average NHL goon. Further, attributing all of the responsibility for the -5.55 goal differential to any one player and none of it to other players on the ice would seem to overstate the impact of that individual. Even if we say that, on average, half of the responsibility for each surplus goal against should be attributed to the goon, the impact lessens quite a bit. Over 51 games, the impact would be -1.86 goals.

Adding the penalty number and goal differential number (and both still strike me as aggressive valuations), the average goon only costs his club 2.66 goals versus average performance over the course of a season. The actual figure is probably a little bit less than that. Again, I agree with the overall premise that playing goons is a very bad idea, but as a group, they don’t have the kind of drag on team performance that you seem to suggest, mostly because they just don’t play very much.