It's almost impossible to watch a hockey game without hearing someone complain about a penalty. While most fans direct their ire towards the refs (for good reason, mostly) you'll also hear fans complaining about certain players taking dumb penalties, especially at inopportune moments. Taking bad penalties can be a momentum killer and can cost wins. So, forget the refs for a moment... the players are the ones committing the penalties, shouldn't they be held at fault?
To that end, I wanted to try an experiment with penalty minutes to see if there were certain players that hurt their team more by committing too many penalties. The essential question here is this: which players hurt their team the most by spending too much time in the penalty box?
First, Some Givens
Before tackling this question, there's two axioms that must be accepted. The first is that when a player is in the penalty box, they hurt their team by not being able to contribute offensively. Yes, certain players hurt their teams defensively by being in the box, but that's not really what I'm looking at here. Maybe defensive absence is a future addendum to this experiment. Admittedly it would be difficult, though not impossible, to measure the defensive contributions of all players, although Stephen Burtch has developed dCorsi to try do just that . In addition, we would all agree as Predators fans that having Shea Weber or Paul Gaustad in the box during a penalty-kill hurts the team and the coaching staff, but again, that's not what this experiment is about.
The second axiom is that when it comes to measuring a player's absence on the ice by being in the box, all penalties are essentially the same in terms of severity of penalty and time served. We would all agree that a 5 minute major hurts the team far more than a 2 minute minor, but for this experiment I really only looked at overall PIMs. The total amount of time that a player spent in the box is what matters here.
Given this, the goal was to see which players are hurting their team from an offensive standpoint. By not being in the box, they can't be out on the ice dishing assists and scoring goals.
If you've made it this far and you've accepted the premise of the experiment, the method should start to crystalize. Take a player's average time spent in the penalty box, compare it with his average point production and his average time on the ice, and you've got a rough idea what he should be contributing if not for his time in the box.
(PIM/60 * P/60)/ATOI
This formula gives you an idea of the point production a player might have contributed if not for being in the box. Let's call it Points Missed/60.
But wait. We've got to figure out what do we do about the extremes. There are a niche of bottom tier forwards that by their sheer volume of penalty minutes might throw the numbers off a bit. I'm not even really talking about 3rd and 4th line forwards that fore-check and play aggressively, though they do throw the numbers off a little. I'm talking about the enforcers, the grinders, the checking unit guys that no one is really expecting to contribute anything offensively. Chris Neil, Cody McLeod, Zac Rinaldo, Jared Boll, etc. These guys are not hurting their team offensively; they serve their purpose, albeit an outdated one. So in order to really see which skaters are hurting their team offensively the most, we have to filter out players that are simply not out there to contribute offensively or that don't play enough to warrant an "offensive contributor". The easiest way to do this is to put a floor on the ATOI. So I chose a very generous 14:00 minute ATOI floor.
Finally, before we look at the numbers, the other thing to consider is the sample size. It simply wouldn't make sense to look at the current year's PIMs or even just a single year's PIMs. The average player only spent 21 minutes in the penalty box last year. We need to look at these numbers in aggregate to see more of a pattern. So I took numbers from the last 5 NHL seasons, from 2010/11 to 2014/15. The column on the far right is calculated using the formula above.
For reference, the league averages for 2011-2015 are 2.95 PIM/60, 1.29 P/60, and 0.30 PtsMissed/60.
So let's talk about the top of the list first. Steve Downie isn't exactly an offensive superstar. He's not even a top-6 forward. The only reason he should be at the top of the list is because he takes a crazy amount of penalty minutes per 60, right? Well, yes, but don't discredit what he brings to the ice from an offensive standpoint. He has a career 12.4 SH% and his 1.98 P/60 is certainly respectable. There is also a reason he is currently a 4th liner on the Arizona Coyotes: no one is chasing after Downie as an answer to their scoring deficiencies. And he is right on the cusp of that 14 minute ATOI floor, so we can probably call Downie an exception to the rule here.
But let's briefly talk about two of the other players at the top of the list: Scott Hartnell and Milan Lucic. If not for being in the box, these players could have increased their P/60 at or above 3.00, instead of hovering around 2.40.
Scott Hartnell has been a top-6 forward for several years now, and his reputation follows him. He's always been a great shooter (career 12.6 SH%) and is always a quality forward in the offensive zone. But his style of play leads him to the penalty box quite often. So his name being at the top may not be a surprise: coaches know when they have Hartnell in the lineup he is as likely to serve 2 minutes for roughing as he is to score a goal. And they apparently seem willing to take that gamble.
But what about Milan Lucic? A 30 goal/60 point forward and yet he finds his way to the box at 1.93 PIM/60 more than the average skater. His tendency for penalties probably resulted in a lower scoring output during his time with the Bruins, even though they won a Cup while he was there. Nothing has really changed since he moved to L.A. In 13 games so far, he has 18 PIMs in 215 TOI, for a 5.02 PIM/60.
Some other interesting names in the top 20 include Corey Perry, Evgeni Malkin, and James Neal. Those guys are on the ice for one reason: to score goals as top line forwards. And yet they have clearly missed out on a significant number of scoring opportunities simply by taking so many penalties over the last 5 years. Or have they?
This brings me to the next phase of the experiment.
Adjusting The Numbers
Penalties are an inherent part of the game. It would be unreasonable to expect a player to go un-penalized over the course of a season. So how relative is one players PIM/60 verses another? Is it reasonable for Hartnell to be taking so many penalties? Or Lucic? Or what about those 40 goal scorers Perry and Malkin? This led me to another essential question: if these players had merely received an average number of PIMs during this span, what might their production have looked like? This might help us realize what is really the heart of this inquiry: are these players really hurting their team by their play or not?
So let's look at this another way. Let's suppose that all these players received PIM/60 right at the league average, or 2.95 PIM/60. How much time above average are they spending in the box relative to another player? And is the extra amount of time they are spending in the box hurting their team significantly? A simple adjustment to the formula should tell us. Let's adjust the PIM/60 by subtracting 2.95 (the league average for 2011-2015) from each player's PIM/60. We'll call it RelPIM/60, leading us to an adjusted PtsMissed/60, or:
(RelPIM/60 * P/60)/ATOI
So the RelPIM/60 number is telling you many more minutes per 60 that player is serving time in the box against the average NHL player, and the AdjPtsMissed/60 is how many more points per 60 they could have contributed if not for that extra time.
You'll notice a couple new names in the list (guys that pile up the PIMs), as well as you'll notice some names disappear. Guys like Malkin, Neal, and Perry disappear. They receive penalties somewhere around the league average, so they are probably not hurting their team by taking the penalties they take. But there are Hartnell and Lucic, still in the top 4. As well as another top-6 forward I've yet to mention, Wayne Simmonds. These three seem to be taking an above average amount of penalties and, given their offensive skill sets, are probably hurting their team.
One third of a point per 60 minutes may not seem like that much. But if you were to tack on 0.36 to Hartnell's 2.40 P/60, he'd be in good company: Joe Thornton (2.76), Patrick Sharp (2.75), Phil Kessel (2.73), Jonathan Toews (2.72). The same goes for Lucic.
It's obvious that top Hart-candidate players like Sidney Crosby and John Tavares don't do their team any favors by getting sent to the penalty box. Same thing with elite goal scorers like Vladimir Tarasenko and Steven Stamkos. But let's return to the first essential question: which players hurt their team the most by spending too much time in the penalty box? Players like Crosby, Stamkos, Tavares, etc. don't find themselves getting penalized as much, thus they really don't hurt their team in the long run. They might get an occasional penalty, but in aggregate their penalty numbers are more than reasonable. So the key here is trying to find out which players hurt their team the most by being in the penalty box too often.
By that regard it would appear your best candidates are Scott Hartnell, Milan Lucic, and possibly Wayne Simmonds.