Editor’s note: This story was written by former Flyer Chris Therien, who is one-third of the Snow the Goalie podcast team. 

===========================================================================

The big trend in sports today, not just in hockey and not just at the professional level, is the ever-widening use and misuse of analytics.

Personally, I can take or leave analytics. I’m not ashamed to call myself old-school in how I watch and break down games. That said, every NHL team has an analytics department. The effective ones find various ways to further illustrate the same sorts of things that coaches have looked at for decades.

I was curious about how analytics departments are used in real life by coaches in the NHL. To get those answers, I talked to two current NHL coaches and one former coach who has coached in the analytics age.

All three have head coaching experience at the NHL and AHL level and all three won the Stanley Cup during their playing/coaching careers.

When I interviewed them, I did not interject my own opinions. I simply asked them two questions:

  1. How much do you/did you use analytics as a coach?
  2. What data do you find useful in your job?

Their answers follow in the next sections.

I noticed two common themes. Typically, all three coaches used analytics as a supplement to the in-game eye test and subsequent video study. Basically, they use it to answer the question, “Do the stats match up to what I’m seeing?” Second, they do make use of stats to look at team-wide trends and as one piece of background information in preparing for upcoming opponents.

Here is what they said:

 

Coach 1

“There is lots of overuse and wasted time and data that means nothing. But I do think there is (a usefulness) to taking some stats that a coach thinks are useful to show some players weaknesses they can improve in their game. It gives them a framework to see what to work on and improve.

“To me, there are stats that show patterns in team systems. Coaches watch video and we get pre-scouts. So, we’d see it, anyway. But the data does give you some pointers on what you might see as you prepare. The data on where and how players are used can be a good head-start on the pre-scout meeting, for example.

“I think having the data is good, but to rely on it without a coach’s experience and common sense would be disastrous.”

 

Coach 2

“Things happen fast at ice level, so there’s a lot of different things I’m managing from the bench. There are things you’ll miss. For example, I might think we defended pretty well and we only gave up a couple chances from the high-risk scoring areas. The data might contradict that a little bit, and I find out we gave up several more than I thought. Then I’ll look at it on video and see what I missed.

“Usually, the analytics tell me what I already knew and just fills in a few blanks. Within a game, I really don’t make much use of it. I don’t want someone handing me a sheet of paper after the second period with the things they think are important.  We spent a lot of time in our own end when [a top superstar’s line] was out on the ice?  They had more shot attempts and three scoring chances? They’re carrying rather than dumping into our zone? No [kidding]. I already knew that. Give me a break!

“I already went in knowing what matchups I wanted and what matchups we’d have to manage around. If I don’t like how it’s going during the game, I’ll see that myself and try to adjust.

“I do make some use of stats for pre-scouts. The video will show me that we can forecheck a team’s D into turning pucks over but they have a couple guys who make a good first pass under pressure. And there are numbers that say the same thing, so there’s some info that confirms what I’m seeing.

“There are some in-house stats we keep that I might look at to see if something our staff has been working on with a player has gotten better compared to where it was at before.

“I don’t think any of it is a big secret. We all do the same things, because it’s pretty standard stuff. The real test to me, still, is how what I see on the ice matches up to the video and then the stats are kind of the [tertiary] part of breaking down the game. It’s usually backup confirmation for what you see with your own eyes.

“In terms of getting ready for the next game, we all look for little ways to be a little more efficient, and the numbers help a bit with that.”

 

Coach 3

“Stats can argue just about anything you want them to argue, depending on which ones you want to cherry-pick. To be honest, I think we all tend to pick and choose the numbers — not just in hockey — that match up with what we already thought. That’s kind of human nature. But, ultimately, there’s no substitute for having a feel for the game.

“I have seen stat sheets — after games we’ve won, with players who’ve had good games, where the stats would lead me to the wrong conclusion if that’s all I went by.

“Same thing the other way. I’ve seen stats where a guy has a [bad] game where he knows it and I know it but the stats say he was great.

“Go back and watch it on video, and almost always, what our coaches see there, the details of the game, might or might be not what the stats say. I still trust what our staff and I see.

“For instance, we have a guy who is a good defender. Good stick-on-puck work. In the right position. Blocks a lot of shots. Strong on the walls. Cancels out guys at net-front – all pretty reliably. By his analytics (though), he is way underwater. But he’s earned our trust.

“I believe in giving information to players. Some find it useful, some don’t. Most of it isn’t analytics but there are times where I’ve used stats a little.

“For example, one of our D was having issues with his gaps. We showed him video, [our assistant coach in charge of the defense] gave him some adjustments, and it improved to a large degree. We had video to show him of what he wasn’t doing right as opposed to before, and also showed him some before-and-after numbers of the last 10 games against the previous 10 that backed it up, too.

“(Now), what do I think about what’s online? I see it. I think you have to keep an open mind. Most of it (though), honestly, is just a bunch of noise.

“Sometimes I’ve been online and I’ve seen media people who bring up all the buzzwords and the trendy stats for the individual players they like and they don’t like. Bottom line, I will never make lineup decisions based on a lot of the things they often seem to bring up.

“There are guys where I’m expecting them to be strong on possession and shot suppression. That all shows up on video, anyway. There are going to be players who start a lot of shifts in the D-zone. They might or might not be great with the puck, which I already knew. It’s my job to know our guys’ strengths and weaknesses and put them in position to succeed.

“We’d all love to attack all night, but there’s two teams out and they’re going to have the puck, too, and get in your zone. You can’t just defend all night, but you need those role players who can hold the score for the next line.

“Not every line and certainly not every player is going to be able to so-called “drive the play,” which is a badly overused buzzword. (Wayne) Gretzky drove the play. Mario (Lemieux) drove the play. (Nicklas) Lidstrom drove the play. Patrick Kane and (Connor) McDavid and (Sidney) Crosby drive play. Erik Karlsson drives play. Other guys are good on the forecheck. They win battles and protect the puck, which is absolutely something you need in the team concept. But I wouldn’t call that driving the play; certainly not as an individual.

“There’s still room in our game for mostly defensive guys and energy guys; maybe not as many of them as before, but you still need a couple guys whose main strength is in one or two particular areas. Those are your so-called role players and it’s our job as a coaching staff to use them in situations where we need them to do the things they are good at.

“Less is usually more for most players. Play within yourself and within the team concept. If we defend well, our players go to the right spots and they can make the simple plays right most of the time, we’ll get possession in good shape. If they are panicking and throwing the puck away, then we’ve got an issue.

“For our skilled guys in particular, there are going to be times where you can make a play because the open ice is in front of the guy checking you and there are times where you make the simple and safe play and live to fight the next battle. That’s when the ice is behind the guy defending you.

“These are not tough concepts, and they aren’t new ones. Analytics hasn’t changed any of that. It’s Hockey 101. It’s always been better to have puck possession than not have it. (It’s) always better to attack than defend. (It’s) always better not to get hemmed in, but if it happens, keep the play to the perimeter, keep inside stick position, box out in front and give your goal a clean look.

“Again, we’re talking Hockey 101. There’s nothing new about any of that, except now the bar is set where you are aiming for three lines that can score and activating your D a little more. You have to know your personnel. I know who my best five-on-five guys are. I know who my best penalty killers are. I know who deserves power play time. I know who I want out there when we’re up a goal or down a goal late. I don’t use my computer or a highlighter pen on the numbers on a sheet of paper to make those decisions. I do look at some numbers. Maybe sometimes I’ll see something that make me take another look. So, I think there’s value in having the data I want to see.

“But at the end of the day, the decision comes from the eyes, ears, brain, and from the gut. I put more value in our coaches’ meetings and discussing what we’re noticing. I need that input.

“Analytics add some info. But, any which way, we have to know if we’ve been playing the right way, which is really the old-fashioned way of saying: good process.”

 

Closing thoughts

I found these discussions with the coaches interesting. I hope you did, too. Ultimately, every coach will use the tools at his disposal in whatever ways he sees fit. What’s helpful to one coach might not be helpful to another, regardless of the sport in question.

When current New York Rangers head coach Gerard Gallant was shockingly fired as head coach of the Florida Panthers back in 2017 before joining the Vegas Golden Knights, there were widespread rumors that he was let go because of his complete disdain for analytics. He then set the record straight for where he stands on the issue. His thoughts closely aligned with those of the coaches I spoke to directly.

“No, I wasn’t huge about analytics,” Gallant said at the time. “For me, analytics is certainly part of coaching, but it’s not the whole thing. In my mind, if I take a job, analytics is part of it for sure, 25 to 30 percent, whatever percentage you want to put on it. It’s definitely a tool. If you get the right information, you’re happy with that. Every coach uses analytics. We all go over the same stuff.”

Barry Trotz, then of the Washington Capitals and now with the New York Islanders, coached against Gallant in the 2018 Stanley Cup Final. He has said that he makes certain use of analytics but cautioned against playing what he terms “an analytics game.” He said it specifically in regard to not going overboard in trying to carry the puck into the offensive zone in situations where chip-and-chase is the smarter play.

Said Trotz to NHL.com, “Teams were really clogging us up and when we needed to chip pucks and have self-retrieval we were trying to dangle. I call it analytics death. There’s a time to carry and there’s a time to self-chip it. You’ve got all the speed so self-chip it, get on the other side of the guy, and hang on to it. We were trying to dangle guys.”

Trotz was right. The same thing goes for exiting the defensive zone. If there’s a clean breakout available rather than going high off the glass, that’s one thing. But there are times where the old-school play is better one.

It’s all situational, and it’s hardly anything new to our sport. Those options and decisions have been things players on the ice have dealt with for a long, long time.

 

Chris Therien played in the NHL for 11 seasons, playing all but 11 of his 764 regular season games and all but five of his 104 playoff games with the Philadelphia Flyers. He later went on to be  both a radio and television broadcaster and analyst for the Flyers for 15 years before joining the Snow the Goalie Podcast with Crossing Broad contributors Anthony SanFilippo and Russ Joy in the Fall of 2021. You can follow him on Twitter @CTherien6

[the_ad id=”103880″]