WINNIPEG — When Scott Arniel joined Rick Bowness’ staff in 2022, the Winnipeg Jets coaches believed the team faced two key problems.
The first problem was culture. Bowness spent that summer getting to know his leadership group, recruiting feedback from several players about what had gone wrong the previous season. Ultimately, Bowness took the captaincy from Blake Wheeler in year one, gave it to Adam Lowry in year two and put a lot of energy toward encouraging new leaders to emerge.
The second, more tactically obvious problem was Winnipeg’s defence. The Jets had given up the 13th most goals in 2021-22 — largely due to a lack of defensive structure and effort in front of their Vezina-calibre goaltender. Publicly available metrics showed that Winnipeg was a bottom-10 defensive team at five-on-five, a bottom-10 team on the penalty kill and that the Jets gave up way more shots than average from the middle of the ice.
But how was that happening? And what did Winnipeg need to change to fix it?
Arniel, Bowness and the rest of Winnipeg’s coaches addressed those topics in their first formal presentation to players.
“We used analytics to show them different areas — slot shots, loose puck retrievals, plays into dangerous areas off the rush and how many plays get into the slot,” Arniel said. “We were number 29 out of 32 teams (in inner slot shots against) so that was going to be our emphasis.”
Winnipeg’s coaches were armed with clips prepared by longtime video and analytics coach Matt Prefontaine. They combined proprietary data from Montreal-based analytics company Sportlogiq with video software called Sportscode and showed players how Winnipeg gave teams access to the middle of the ice, leading to slot shots against. Some of it was about in-zone coverage. Some of it was about backtracking. All of it required all five skaters to work together as a team to cut down on the most dangerous scoring chances.
The coaches’ presentation of data and video made an impression on Winnipeg’s players. Bowness, Arniel and their staff generated buy-in for the X’s and O’s changes that they wanted to make. Inner slot shots became a focal point for Winnipeg to stay in its defensive structure between the faceoff dots, to backcheck hard and to commit to outnumbering opponents in key areas instead of getting picked apart in the name of chasing offence.
“Bowness made us more aware (of analytics) than what you guys are led to believe,” Lowry told The Athletic at training camp. “Inner slot shots, that’s kind of become a good indicator for how we protect our side of the ice off of D-zone coverage.”
The Jets improved from 253 goals against the season Paul Maurice left to 224 goals against in Bowness’ first year, then 198 to earn the Jennings Trophy last season. Their commitment to the X’s and O’s that Bowness and Arniel taught them was obvious in the metrics they most liked to track: Winnipeg went from 29th in inner slot shots against in 2021-22 into the top 10.
“We improved after year one,” Arniel said. “Year two, we went to some offensive things. This year we’re going to have some areas that we’re fine-tuning now. What’s next this year? I can’t tell you. But our video will have analytics (in it).”
It sounds like such a happy story, doesn’t it? Team recognizes defensive problems, marries analytics to video to show what is important to change and then goes forth and makes those changes. In Arniel’s first season as head coach, Winnipeg hasn’t seemed to lose a step: The Jets are 7-0-0 and have given up the fewest goals in the NHL.
But what are the Jets’ next steps in their quest — not just to catch up but to get ahead of the curve? What happens when coaches or players think analytics have gone too far? And, despite this continued investment in analytics, why do the coaches keep playing a top line that gets outscored at five-on-five?
The Connor, Scheifele and Vilardi question
The Jets are undefeated and their top line of Kyle Connor, Mark Scheifele and Gabriel Vilardi has been outscored 6-2 during five-on-five play. The trio has won Winnipeg games, too, but those victories have been earned on special teams, at six-on-five and in three-on-three overtime.
Scheifele’s line hasn’t outshot or outscored its opposition — and didn’t do it last season, either — which leads to an obvious question: In a league where top lines are so good that getting 50 percent of the goals is below average, could the analytically inclined Jets be even better if they found a different combination?
I asked Arniel about Connor, Scheifele and Vilardi getting outscored together.
“Your lines change all year long,” he said. “Even (Leon) Draisaitl and (Connor) McDavid don’t stay together. When it comes to the big time, they go together, but at the end of the day, over 82 games, you’re going to change your lines.
“You’re going to have to trust your feel as a head coach. Is it time to make that change? Is it time to change it up a bit? And you’re right. There is chemistry built. A lot of history that you see with great lines is there’s always been two guys (who make it work). So who are those two guys? Is it KC and Scheif? I don’t know. Is it Gabe and Scheif? I don’t know. That’s part of what we’ve been trying to build. Injuries disrupt things sometimes, particularly with Gabe last year.
“And there’s the human element. The players like playing with each other. They have a feel for each other. They’ll make plays sometimes and the only reason they get away with it is because they’ve done it together a bunch of times … You’re comfortable when you go into a high-pressure game and you know what your linemate is going to do. Sometimes there’s that, too. That’s where I don’t get carried away with analytics.”
My interpretation is that the decision comes down to human factors. It seems clear that they play together because they like playing together. It also seems as though they’ve shown Arniel enough via the eye test to convince him to bet on them getting better results.
EBERLE WITH THE BACKHAND 🔥 pic.twitter.com/nFJagGJm0x
— TSN (@TSN_Sports) October 25, 2024
On Thursday in Seattle, Arniel sat the line for a shift after they were caught watching Jordan Eberle’s third-period goal. And now that they’re down 6-2 in goals as a line, Arniel will have decisions to make. Does a 7-0-0 start mean he can give them more time to find their game, particularly in the defensive zone? Lowry’s line is up 6-0, Vladislav Namestnikov’s line is up 4-1 and Scheifele and Connor are still point-per-game players thanks to special teams success.
“It’s all part of the process when it comes to decisions you make about your lines and your team,” Arniel said at training camp. “How did you play last night? The next day, when you have practice, how mad are you? And yes, analytics will be a part of that. We’ll be able to take a look at them and say, ‘OK, we’ve given up a lot of stuff here.’”
Connor vs. Ehlers could be Connor and Ehlers instead
Arniel smiled, anticipating the question.
“I know where you’re going with this,” he said.
My preamble was as follows. Nobody scores like Connor does. No Jets player has the hands, the speed and the finishing that Connor does. Does he short a backcheck sometimes or struggle to get the puck out along the wall? Yes. But you’re not going to tear that player apart because he is a human being working to get better at those things. You need his offence to win games.
Meanwhile, Ehlers’ defensive game isn’t any better in his own zone than Connor’s is but Ehlers is such a good player in transition that it helps keep Winnipeg out of its zone. But I understand the human element there, too: When Ehlers is zig-zagging in multiple directions, do his linemates have to eat hits in front of the net? Do they sometimes get confused based on the unpredictable nature of his play?
In the end, I asked Arniel how he planned to take those different playing styles, manage the human beings involved and get the most out of them for his team.
“That’s the relationship part of this game,” he said. “Those are probably the best two examples you can throw out there. Both of them have such strong qualities that they bring to the game. KC is in my top five guys in terms of goal scoring. His shot, the way the puck comes off his stick. Nik can’t do that. KC has some of the best hands I’ve ever seen when it comes to shooting the puck. Then, the explosiveness, the game-breaking things that Nik can do that maybe KC can’t do. Nik can take a 2-2 game and make it 3-2 in a heartbeat if he decides he’s going to take off and go.”
Arniel said it’s his job to teach board play and defensive zone responsibilities. It’s his job to make sure players understand when they need to be better in those areas. He also made a point of defending both players’ commitment to backchecking, particularly in recent seasons. Connor’s backtracking has been particularly noticeable early this season.
“Our top players all bought in,” he said. “And maybe some guys do some of those jobs better but you don’t win hockey games without that skill.”
He then spoke to a concept that drives part of his perspective as head coach. The NHL’s 82-game regular season is long. A quality regular-season performance, then, means finding several different ways to win.
“You need pure goal scorers at times. You need shutdown defencemen at times. You need fighters, at times. Specialty teams. You obviously need great goaltending for a lot of it. But every single game has its own quirks to it,” he said. “We could be in November, playing in Anaheim and it’s our fourth game in six nights and maybe we should be blowing them out of the water and we’re not, but maybe that’s the night Nik Ehlers goes all of the way down the ice, scores a goal, and wins us the game.”
EHLERS WINS IT‼️
Nikolaj Ehlers is the @Energizer overtime hero as the @NHLJets move to 7-0-0 on the season! pic.twitter.com/ngYfkygHyl
— NHL (@NHL) October 25, 2024
Sometimes I think Winnipeg’s coaches and management see Connor do things that other players can’t do — whether it’s his release, the way he creates space out of a cutback, or the way he processes the game fast enough to stickhandle through some of the game’s top defenders — and lose track of the results. His line needs to get out of its own zone for those marvellous things he does to pay off.
It also seemed to me — although Arniel didn’t say this — that Arniel viewed Ehlers’ game-breaking excellence as a November thing, while Connor and Scheifele are viewed as a go-to duo that can be counted on in the playoffs. Connor’s 32 points in 45 playoff games certainly trump Ehlers’ 14 points in 37 playoff games, and it seems to be that the organization is all-in on one and still figuring out how it feels about the other.
Arniel says he views both players as game-breakers and believes it’s his job to get the most out of them.
“That’s the part where analytics don’t matter,” he said. “It’s about the human element. Now it comes down to what can I get out of that guy and how can I get him to do it when I need him the most.”
Winnipeg’s next steps, plus what its coaches are afraid of
Winnipeg uses data in scouting, management and coaching. Arniel and his staff clearly sold the players on the importance of their defensive zone, using specific data like slot shots to drive home their points. Arniel is wise, though, to keep some analytics conversation between himself and the coaching staff. Players need to feel free to play.
“Sometimes you get shown all of these numbers and then you get caught — not every shift, but you get caught thinking — and now it takes out some of that free flow of the game,” Lowry said. “If you’re thinking too hard that ‘OK, I have to be here, I don’t want to give up these things’ but now you’re off in your read and they might get a chance somewhere else. I think it’s about a happy medium of incorporating (analytics) while still allowing the freedom, the creativity of players to make the reads and trusting them to make the correct ones.”
It’s a vital perspective — and one that data analysts have understood for a long time. There’s a difference between finding useful information that might help a team plan its strategy and the choices players make in real time as they play one of the fastest sports in the world.
It’s also important to be careful with how we use the word “analytics” at all.
“I know the word analytics, by some of the players, gets looked at a certain way and they don’t quite understand it,” Kevin Cheveldayoff told The Athletic in September. “So one of the missions is to try and explain how we intend to use it with the players, as opposed to just putting it out there and letting everyone digest it. Because there’s all kinds of numbers out there.”
In a sense, all analytics are about counting things that happen during a game, gathering information about those events and then trying to make smart decisions with that data. Whether it’s about how a team breaks the puck out of its zone, where a team likes to dump the puck or how different teams get into the slot for scoring chances, it’s just information. Sometimes calling it “analytics” can distract from the point while overusing complex charts instead of communicating to NHL players like the elite athletes they are just gets in the way.
“It comes down to how we use it,” Arniel said. “We have to make sure we don’t get caught up on just one thing. Your eye test is still there. Your feel. Your relationship with your players. Are you still feeling the same way about them after that stretch of games? Do you have that confidence in him? And then you have your analytics side, too.”
(Photo of Nikolaj Ehlers and Kyle Connor: James Carey Lauder / USA Today)
