clock menu more-arrow no yes mobile

Filed under:

Evaluating Defensemen: Paul Martin

A look at how good Paul Martin really is and whether his partners were holding him back.

Bruce Bennett

As Rich Miller from The Pensblog likes to say, Paul Martin is an American hero. He summed up this sentiment quite nicely in this tweet:

Martin has been with the Penguins for four years now, and has been either their best or second-best defenseman during this time. Nevertheless, much of the mainstream media coverage surrounding his tenure with the Penguins hews to the narrative that Martin struggled mightily in the beginning. A quote from a recent article:

Following the 2011-12 season, Martin was disgusted with his performance through two seasons of a five-year, $25 million contract. That summer, he told the Tribune-Review that, "If you had told me this is where my game would be, I wouldn't have believed it."

Years ago, Josh Yohe again wrote that Martin's play could be placed on a spectrum of indifferent to disastrous. That article prompted me to write this piece. Nevertheless, the Penguins have had far bigger culprits on their blueline during Martin's tenure. We've seen the immobile Doug Murray, the disastrous Rob Scuderi, a version of Zbynek Michalek who seemed lost all of the time, and the steady decline of Brooks Orpik. None of these (bigger) problems received the consistent negative attention that Martin did, though. And I always wondered why.

I try to answer that question in this article.

Prior Work

Some very smart people have already discussed the biases inherent in how we evaluate defensemen. The short of it is that we seem to be very patient with the unskilled guys who simply block shots and lay hits, but tend to get fed up--very quickly--with the offensive guys who carry the puck and help drive play. Here's a great quote from the astute Adam Gretz:

Why is it that players like Karlsson, Green, P.K. Subban and Kris Letang are kept on such a short leash, while stay-at-home defensive-defensemen (who are often times just as one-dimensional as their more skilled counterparts are) are given a mile to work with before they face the same intense criticism? If one of the aforementioned players is on the ice for a goal against because he tried to jump into the play at the other end of the ice and couldn't get back in time (and it does happen), it's replayed 100 times, fans and media are screaming about how that player needs to focus more on defense, and it's probably time to sit them on the bench so they can learn their lesson and understand their role.

But if a guy that stays in front of his crease and blocks a lot a shots just flat out gets beat because he's not good enough to make a play or too slow to get the puck out of his zone, it seems to be more forgivable because, hey, at least he's playing defense and that's what a defenseman is supposed to do.

One of the focal points in how a lot of people evaluate defensemen seems to be turnovers. Guys like Brooks Orpik and Rob Scuderi didn't have a lot of turnovers this year for the Penguins, so some conclude that they were "more responsible" than guys like Letang, Niskanen, and Maatta. And from this, people will intuit that Orpik and Scuderi are the good players (few turnovers = solid players) and that Letang/Niskanen/Maatta are the reckless ones who took unnecessary risks.

Of course, that's not what happened. Orpik and Scuderi had so few giveaways because they almost never handle the puck. Why don't they do that? Because they always pass it off to their partner or throw it up the boards as soon as they get it. And why is that? Because they don't skate, pass, or carry the puck well. In short, they don't drive play. Because they can't get the puck to the forwards in an efficient manner, it falls to Letang and Martin to do so. Naturally, they'll have more giveaways. But that doesn't mean they stink; it just means their partners are akin to boat anchors.

This dovetails with an excellent article from Tyler Dellow on big mistakes. The piece is outstanding, and focuses on how bad the Leafs were at puck possession and how that affects our evaluation of that team. The key part, to my mind, is quoted below:

There’s a reason that "Hockey is a game of mistakes" is a hoary cliche. It’s because it’s true. The more opportunities you have to make mistakes, the more mistakes you will make. No team in the NHL has more opportunities to make mistakes in the defensive zone than the Maple Leafs. That’s the underlying cause of this rot.

The Leafs can attack the underlying issue and try to become a 50% Corsi% team on their way to being one of the league’s elite. Or they can try and find guys who make mistakes 14% less often than the average guy so that they can survive as a 43% Corsi% team. Either way focusing on the big mistake misses the easiest way to stop suffering from them: be in a position to make them less often.

Too often we seem to be focused on the mistake that led to the goal against. But because there are so few goals against compared to shots and opportunities, this doesn't paint a reliable picture of one's true talent. Instead, we should be looking at how often a player puts himself in a position to make those mistakes, and the guys who skate, pass, and carry the puck well don't often put themselves in vulnerable positions.

Paul Martin's Stats

Below is a table with Martin's even strength stats for the last three regular seasons. The number in the parentheses indicates Martin's rank among the team's regular defensemen.

Games Played TOI/60 P/60 PDO CF% CF% rel Offensive Zone Start % QoC TOI%
2011-2012 73 (2) 17.1 (3) 0.76 (3) 97.6% (6) 55.2% (2) 1.7% (3) 54.1% (4) 28.9% (4)
2012-2013 34 (5) 18.0 (1) 1.27 (2) 103.4% (3) 50% (4) -2.0% (5) 47.3% (5) 28.8% (2)
2013-2014 39 (7) 17.4 (1) 0.35 (8) 96.6% (9) 49.2% (4) -2.5% (5) 46.2% (7) 29.4% (1)

Because extra skater only goes back three years, we have to use Behind the Net to get Paul Martin's numbers for his first season with the Penguins. Those numbers are in the chart below:

Games Played TOI/60 P/60 PDO Corsi ON Corsi ON Rel Offensive Zone Start % Corsi Rel QoC
2010-2011 77 (2) 16.44 (1) 0.57 (5) 98.9% (6) 10.19 (2) 6.7 (2) 51.5% (5) 0.608 (3)

I think there's two things worth highlighting. First is usage. Paul Martin led the team in even strength ice time three out of four years. That means he's playing big minutes and that the coaches trust him to make good decisions during those minutes. Additionally, he's played against top-level competition (especially the last two years) and has been given less than favorable zone starts in the two most recent campaigns. The fact that he's been continually rewarded with a lot of minutes against top competition means that he's doing something right.

The second point is that Martin has always driven play by out-possessing the other team. In the first two years, his numbers were elite, buoyed in part by his zone starts. They took a dip in 2013 and 2014 because he and Orpik were often placed in the defensive zone to take tough faceoffs in critical situations. But note that both this year and last year, Martin's CF% was higher (by a number of percentage points) than his offensive zone start percentage, which tells me that he's driving the bus despite the difficult minutes he's playing.

We can get an even better picture of his play if we dig into his with or without you (WOWY) numbers from prior seasons. I want to see whether Martin was holding his partner back or vice versa, and we do that by comparing how they do on the ice together with how they do when they're apart.

Martin's WOWY Numbers

In the 2010-2011 season, Martin spent the majority of his even strength time paired with Zbynek Michalek (nearly 800 minutes). Their WOWY chart is below and it's measuring CF% ("With Partner" means when Martin and Michalek were together; "Without Partner" means when they were separated). Please click all images to enlarge them.


That's a stark difference. Martin and Michalek had solid numbers together, but in the roughly 450 minutes apart, Martin was approaching a 58% CF while Michlek was trying to keep his head above 50%. The easy conclusion is that Martin was carrying Michalek. Once they were separated, Martin dominated play like an elite defenseman and Michalek appeared rather ordinary.

Context is important, though. We want to make sure that when they were apart, Martin wasn't getting easy minutes while Michalek was getting tough ones. If that were the case, then their difference in CF% wouldn't mean much. One could say "of course Martin does better: Bylsma is using him in a dramatically different fashion when he's away from Michalek." But in the 2010-2011 season, that wasn't the case. Martin saw tougher competition than Michalek and had a zone start percentage only a few points better. Because of this, I would discount the CF% numbers slightly due to the difference in zone starts, but that doesn't explain the nearly 8% swing when these two are separated. Martin was (and still is) much better than Michalek, and Michalek seemed to be holding him back.

Moving onto 2011-2012, Martin spent about 420 minutes with Michalek and 340 minutes with Letang. Their WOWY charts are below.



Similar story. In terms of Martin and Michalek, they worked well together, but when apart, Martin had a 56% CF while Michalek was below 51%. Looking at context, quality of competition between Martin and Michalek were pretty even (slight edge to Michalek), but Martin again had better zone starts. So that diminishes the CF% differential somewhat, but remember--we're dealing with a difference of more than 5% when Martin and Michalek were apart. That's only partly explained by zone starts. Martin was again the much better player.

As for Letang and Martin, the two of them together were dynamite. When separated, they both fell, but Martin was still above 54% while Letang tumbled all the way down near 52%. The context here is the same: Martin and Letang were even in competition but Martin had a better offensive zone start percentage by about 3 points. While that difference explains some of the possession differential, Martin still comes away looking like the better player.

As far as Martin's partners go, the last two years were pretty extreme. In 2012-2013, Martin shared the ice with Orpik for 482 minutes at even strength; his next closest partner was Letang at 40 minutes. In 2013-2014, Martin spent 544 minutes paired with Orpik at even strength; his second most common partner was Engelland, who he shared about 44 minutes with. So the sample when Orpik and Martin are split up isn't very big. It's less than 100 minutes over the span of two years. Thus, we shouldn't read a ton into these stats, but I wanted to take a peek anyways.



Despite the small amount of time spent apart, it's telling to see just how much Orpik craters when he's away from Martin. And it's not because of context. In 2012-2013, Martin and Orpik saw basically the same competition (slight edge to Orpik) and Martin's zone start was only two percentage points better. Nevertheless, away from Orpik, Martin was around 55% CF while Orpik couldn't stay above 40%. And in 2013-2014, Martin saw slightly tougher competition, and had a better zone start by only 0.8%. Though this is a small sample, Orpik looks like a massive hindrance.

To wrap this all up, we'll look at a really big sample: the combined WOWY stats for Martin's four years with the Penguins. The chart below contains every player who played at least 400 even strength minutes with Martin over the last four years and looks at their CF% with and without Martin.


Out of the 12 skaters in that graph, ten of them have better possession numbers when they're with Martin. Only Neal and Orpik have better numbers away from him, but the difference is so slight for each of them that its fair to call it a wash. The takeaway, then, is that Martin makes nearly everyone better. Despite different personnel, systems, and deployment over the last four years, the one constant has been superb play from Martin.

The Key to Letang's Success?

One thing I noticed in the data above is that Martin makes Letang much better. We saw it in 2012 when the two of them played together and approached a 57% CF. And we saw it in our four-year sample, where Letang is nearly six percentage points better in terms of possession when he's paired with Paul Martin.

We also saw it in the playoffs this year. Once Brooks Orpik got injured, the coaches paired up Martin and Letang to form Pittsburgh's super pair, capable of playing 27 minutes a night against top competition and still coming out ahead. Since Brooks Orpik was injured during the fourth Columbus game, the chart below looks at Letang's CF% in the first four playoff games (without Martin) and compares that to the rest of the games (with Martin).


That increase is enormous. Given that Letang's contract kicks in next year, and the fact that he almost certainly won't be traded, it's imperative that the Penguins find out how to get the most out of him. Fortunately they already have the answer: Paul Martin. Since making Letang as effective as possible with his $7.25M price tag is a priority, the fact that Paul Martin can do so further highlights just how valuable he is to this team.


On a micro level, the lesson should be that Paul Martin is a very good hockey player, and that he's been that way as long as he's been with the Penguins. He's been held back by defensive partners who don't handle the puck that well, and he's been tasked with playing some of the toughest minutes on the team during this time. With the numbers he puts up, it never made sense to characterize his play as indifferent, ridiculous, or ineffective.

The broader, macro point, is that some people considered Martin a "meh" defenseman because they focused on the wrong things. Martin isn't physically imposing and doesn't lay a lot of hits, but that shouldn't be an issue since hits have literally zero correlation with winning. Moreover, he doesn't block a lot of shots, which is actually a great thing since blocked shots are strongly correlated with losing a lot of hockey games. And while Martin does turn the puck over occasionally, that's going to happen when you play 25 minutes a night and probably touch the puck a couple hundred times. Focusing on those mistakes to the exclusion of everything else misses the forest for the trees.