There was a pretty interesting exchange in Dallas Eakins’ media availability today, talking about whether or not you need toughness on your defence.
SPECTOR: In the past we would say “If he doesn’t want to play that mean game you’re talking about, he should have some partners or someone else on defence who can fill in that role.” Are we still, in 2014, in need of someone back on your defensive corps who plays that game that you described?
EAKINS: What game did I describe?
SPECTOR: Well, the “I’m scared of you” defenceman. Do we still need that guy or are we past him?
EAKINS: Well, you know what? I’ve always loved having toughness on my teams. It’s the old school part of me. And the old school part of me is starting to get questions. When you look around the league now, the teams are changing their makeup. There’s teams that are playing a heavy game but not a lot of meanness. Just go, they check and they play high pace.
So I don’t have the answer for you. I could easily stand here and argue that “Yes, we need that.” We’ve got a guy back there that’s more than willing to fill the role in Mark Fraser and one side of me says “Absolutely, we need the toughness, we need toughness up front, we need it in the back end and the I go right over, two days later, and go the other way, say “We need puck moving guys, we just need to play fast and that’s it.”
And it goes back to, I just remember a conversation I had with Paul Maurice as an assistant coach. We were talking about defencemen, it was about half way during the year. He said “Go right now, I want to know how many hits Lidstrom has, right now.” And I went and I looked and then I thought it was wrong. It was like halfway during the year and the guy’s the best defenceman in the league, best defensive defenceman in the league and he had one hit. And I was like “That can’t be right. He’s had to have run into somebody else.”
But you know Mark, that’s the god’s honest truth, I’m not trying to get out of the question. There’s one side of me that says “Yes, we need to old school it and we’ve got to have those guys” and there’s the other side of me that looks at how some other teams are building and…
I’m not sure.
I know that there are a lot of Oiler fans who aren’t particularly happy with Eakins as the coach. He’s made the Journal’s front page for the weekend promoting their survey on who the Oilers need to fire or trade.
Aside: Taylor Hall? Really?
For what it’s worth, I don’t entirely get why the knives are out for Eakins. I get that this has not been a good year, one of eight. You just have to listen to the guy talk though – he seems like a guy who thinks about things, is rational and pragmatic. I haven’t agreed with everything that the Oilers have done this year, or everything that he’s said (just in general terms, I think I was more skeptical that they were playing well in the 10-4-3 run than the Oilers seemed to be) but there has to be a certain degree of “Find a good pragmatic coach and give him some time” in all of this.
Just on that 10-4-3 run…the table to the left sets out the Corsi% and PDO of players at 5v5 during the streak. I mean, it screams “Unsustainable.” There isn’t a team in the NHL that wouldn’t look great with a bunch of guys with PDOs of 103+. I’m actually more encouraged by the most recent three game road trip where they went 1-2-0 but managed to hang around Corsi% wise despite playing three in four nights than I am by the 10-4-3 run.
ANYWAY – where was I? Oh yes, “toughness.” It’s interesting to me that Eakins brought that up for a number of reasons. First, I’ve long been skeptical of the need for toughness, in the way that people talk about it in the NHL. I’m a huge believer that you have to be tough to play hockey but I’d characterize guys like Taylor Hall and Ales Hemsky as tough. They aren’t scared to go to hard places on the ice. They aren’t intimidated.
Just in general, I think NHL players are probably less intimidated now than they’ve ever been. It’s a bit of a consequence of hockey’s crackdown on violence, I think. This hit from 1991 always stuck in my head:
Keep in mind, the game had been cleaned up a lot since then. Today, Messier would get five games for that. The first time he did it. The fear of someone violently fouling you and injuring you has been greatly reduced. Guys like Hall and Hemsky can skate through the middle without the fear that a guy like Scott Stevens is going to turn the lights out.
So what’s left? Guys who play rough hockey? Take Mark Fraser, the Oilers’ current provider of toughness. The coaches (quite properly) won’t play him against anybody good. I loved this shift chart from Anaheim, where the Ducks had last change.
There’s basically one way to hide a defensive pair when you’re on the road and that’s to put them out as soon as the line you’re ducking comes off the ice. This is pretty much a textbook case of that. What’s the point of toughness and intimidation if you’re scared to use them against anyone good? Who are they intimidating then? The other team’s depth players? Sort of? Who cares. Guys like Joe Thornton and Ryan Getzlaf drive the bus and they intimidate coaches into playing good defencemen against them, not tough ones.
Here’s another example of the dance that the Oilers do with Fraser, this time from the 5-2 loss at home to San Jose. San Jose’s a bit trickier because they’ve got multiple lines that can hurt you.
Toughness that you won’t use against anyone decent (and where you’re giving up defensive/puckmoving skill for the toughness) seems pretty useless to me. Just as addendum to that, there’s been kind of an idea getting kicked around hockey analytics geek circles lately about the relative merits of soft puck moving defencemen. I tend to think that they’re undervalued unless they’re obvious stars. Petry’s been getting dumped on all year, despite being the guy who sees the best results when he’s on the ice and playing the other team’s best. It happened to Tom Gilbert. It happened to Tom Poti. Larry Murphy got chased out of Toronto. There is just an endless list of these guys.
This has been an idea that’s kind of fascinated me for a long time. Given that we don’t think that defencemen have an impact on shooting percentage and save percentage when they’re on the ice, what’s left? Their impact on the rates of shots for and against. Just from a logic perspective, who do you think is going to have the biggest impact on shots for and against? For me, it’s guys who can move the puck.
Along those lines, there’s a fascinating connection between the rate at which defencemen get assists on goals and their shot share. I gathered the data for the 188 defencemen who played at least 3000 5v5 minutes between 2007-13 and sorted them by the rate at which they get assists on goals. In other words, if you’re on the ice for 30 goals and get 10 assists, your rate is 33%. I then made ten groups of D, from best assist rate to worst. Here’s what the results showed.
There’s a lot of interesting information there. Note that the S% and SV% are virtually identical for every group. That’s consistent with the idea that defencemen don’t have a lot of control over the S% and SV% when they’re on the ice. Then slide your eyes down the SF/60 column. Defencemen who get assists on a higher rate of goals tend to see their teams take more shots when they’re on the ice. I find this fascinating.
Why might this be the case? My suspicion – which I don’t consider proven – is that defencemen who get more assists on goals aren’t getting them because they necessarily make the killer pass but because they can handle the puck and get it moving the right way. The skill underlying the higher shot rates when they’re on the ice would also underlie the rate at which they get assists on goals.
To tie that back to what Eakins was saying, I think that there’s at least some reason to suspect that his new age side is right. It’s one of the better questions in hockey analytics these days. Everyone wants to figure out which players are undervalued by the market. My suspicion is that “soft puck movers” are pretty high on that list and that toughness and defensive defencemen are on the overvalued list.Email Tyler Dellow at firstname.lastname@example.org