Merrimack College hosts New Hampshire this weekend for a two-game series at Lawler Rink, including senior night on Saturday. On Thursday, Merrimack head coach Mark Dennehy had his weekly availability with the media after practice as we talked this weekend, Merrimack’s culture, as well as analytics in hockey:
Over last few weeks, you have said you’ve felt you’re working harder. It seems like execution is catching up too. There were some mistakes last week that resulted in their goals. Is that accurate?
Dennehy: “Sometimes coaching is like catching sand in your hand. You’re dropping sand so you put your other hand underneath, then the sand is going through that hand so you’re constantly switching hands. We’re doing some things better on one front, then slacking off on another.
“We did a good job on the rush and we struggled in the defensive zone and face-offs and we’ve good there. We haven’t put a full 60 minutes together where we had all 20 guys humming. We just haven’t had it. We had some of our better players not have best weekends last weekend and they know it. It’s not from lack of effort or desire, sometimes it just happens. You get back to work and try to tune it up. We need to get more pucks to the net and score more goals.”
Obviously you’re trying to put it all together this weekend with playoffs around the corner?
Dennehy: “We’re running out of weekends. It can’t be any later than next weekend. I’m just stating the obvious, but thats the reality. We believe it’s there. We have had some good efforts lately. That game at Maine (on Sat.) we played really well. It was not good enough but we played really well and we haven’t had enough of those nights, especially in the second half.”
You said on Saturday you thought the culture was slipping away 18 months ago. But now, with the effort level where you guys want it, do you feel like you’ve stopped that ball from rolling down hill per se?
Dennehy: “I’m not sure we stopped the ball from rolling down hill yet. We identified the issue. I think that everybody involved in our organization have an understanding of what at least the coaches think is the problem. But that’s only half the battle and it’s usually the easiest half. Now we have to address that issue. We have to play harder, we have to block more shots. We’re working towards that.”
Does the size of next year’s class have any relation to that? To changing that culture? Getting new bodies in here, was that intentional?
Dennehy: “Oh yeah. We need to find guys who are willing to play the way we need to play. We think there are some here and we think we have some guys coming in. At the end of the day the proof is in the pudding. I haven’t blocked anyone’s shot, I haven’t back-checked anybody. I haven’t defended anyone. You have an opportunity, you make the most of it. Some of the guys here have, and we think some of the guys coming in will grab the bull by the horns as well. But that’s the future, we’re still concentrating on the here and now. We have some time here to make something of this season.
“We’ve been given a gift. We’re in the playoffs. We have two regular-season games left, both here at home. It’s up to us. Even though we’re working towards the future, the present is what we’re focused on. We’re focused on today.”
With it being senior weekend, what can you say about this class graduating?
Dennehy: “First of all, anyone who wants to honor the seniors, and I want to honor them, we do that at the banquet. During the season, our focus is on the game. We’ll recognize them on Saturday, but we’ll honor them at the banquet.
“They’re each, individually, wonderful people. They’ve invested four years of their life in Merrimack hockey. Our coaching staff is grateful in that regard. They all have strengths and weaknesses and they all bring something different to the table. They’ve all helped contribute to this program. When you get to this point as a senior, we’re thankful they’ve chosen Merrimack and those guys have left their mark.”
You’re taking part in the Sloan analytics conference. How did that come about?
Dennehy: “Well, a friend of mine works with the Boston Bruins and he is someone I call on when I have questions about some of the analytics we’re trying to put into place here. They’re a professional operation and they’re doing some neat things over there, and having access to that isn’t something they take lightly. He has talked to Donny Sweeney and Peter Chiarelli about some of the things they’ve shared with us, and it’s not watershed. I don’t know a tenth of what they’re doing over there, but because of my interest, when they asked Don Sweeney for someone in the college ranks, he was kind enough to reference me. To quote my wife, ‘I have a lot of opinions and I’m more than happy to share them.’ They asked me, I was more than willing. As a matter of fact, it saves me a lot of money because this is something I’d pay to go see and take part in. It’s neat. Hockey is just literally on the cusp of analytics. I think we have some of the worst statistics in all of the four major sports, and I mean the traditional statistics. They don’t tell nearly as much about our game as the new stuff.
“At the end of the day though, it’s a simple equation: get good players, get them on the ice, set good standards and get out of the way.”
Do you guys use advanced stats here? For someone like me, I love them, but it’s hard in college hockey because I can’t physically go out and chart every game for every team. But do you have guys keep some of those stats and study them? Do you use things like Corsi or PDO?
Dennehy: “We use Corsi. We have a a scoring chances plus-minus based on quality of shots. It’s hard. In the NHL they have a timeline. They have an army of statisticians and the standard is set by the NHL. Each team starts with an NHL timeline they can dump into a computer. We don’t have that.
“The other thing, in college, we don’t have a lot of data. Sometimes it can lead you in the wrong direction. I remember my statistics class and I struggled with it, because one of the examples they gave was, a drug test. If a drug test is 95 percent is correct, it doesn’t mean 95 out of 100 will be correct. You might get 10 false positives right off the bat, but over the long term, it will work out to be 95 percent. Well we don’t have a long sample set, we have 34 games.
“In the NHL they have 200,000 shots per year. If you have five years worth of data on shots alone, you have 1,000,000 points of data. Now you can start to make some predictions.”
I imagine it must be really hard in terms of recruiting. You almost can’t use any of it there because those leagues aren’t charting anything like that.
Dennehy: “No. I think we’re beyond the point now where it’s completely intuitive. What I mean by that is we’re beyond the point where you go watch a player and you can say, like Billy Beane, ‘he’s a five-tool player.’ But I also don’t think it’s completely analytical. It’s somewhere in the middle. It’s a combination. You see it with your eyes and the statistics line up.
“I’ve met very few players who didn’t score goals in juniors that come into college and score goals. But I’ve also met guys who have scored goals in juniors and then get to college and haven’t scored goals. So if you have to disseminate the quality of the goals, and we’re starting to do that.
“Goaltending is a position where they have the best stats. Goals against and save percentage. But even those aren’t great. Both of those are deemed by the defense in front of them and the quality of shots they’re seeing, so you have to go underneath those. The NHL teams are starting to do that and we’re starting to do that as well. When we rate our goaltenders, it’s not the quantity of shots but the quality of shots.
“To give you an example of something I’ll reference. Look at Northeastern. In the league they’ve taken 479 shots, they’ve given up 693. Now I go to Notre Dame. In the league, they’ve taken 606 and given up 529. So if I give you that information, who would you think is in a better position in the league? You’d think Notre Dame. Just on Corsi. Corsi is just shot attempts. But you have to start grading it out.
“We’re also different. We make so many more mistakes on the ice. The guys in the NHL protect the puck like it’s their job, because it’s their job. We have to be careful about using them because they might not always tell the story, but those are just two examples.”
So much of it, to me, is trying to measure possession. As simply as I can think to put it, if you have possession, you can score goals and the other team can’t. If you don’t have possession, then there’s the opposite. So it’s all about possession. Goals are almost random events, you need to possess the puck.
Dennehy: “But we don’t know who is possessing it. For example, I would argue that John Gaudreau needs fewer possession to be productive than the average player. So when he has the puck, the likelihood of something good happening goes up exponentially. Again, that’s where they might not tell the whole story. So it’s a combination of intuitiveness and analytics.”