When people talk about firing the coach, which is usually very emotional, they're doing that thing that Justin Bourne talked about in regards to the Leafs – decide who you want to add, not who you want to get rid of.

I wrote a big headline when Bruce Cassidy was fired about how Brad Treliving could save his job by firing Berube and hiring Cassidy. Not that he really could have, no GM on any team gets to make the decision to have the organization eat years of contracted salary by themselves. But that was the only idea that could have helped when Treliving was hanging by a thread. He did nothing – his forte – and he is gone. Bruce Cassidy's name is still on everyone's lips as the most famous guy who has the appearance of being good. Win/loss or Cups won or reputation or any of that is not a good reason to bring in a coach. It's how you argue about who should coach. Not decide.

The question first needs to be: should they change coaches at all. So I want to discuss how I see that in a way that does not require deep or even superficial understanding of measures that mean something. Again win/loss record means nothing. It's as wrong to use that as it is to listen to colour commentators claim GAA measures a goalie and not the entire team.

I will try my best to not have to start this with: In the beginning there was Corsi... Instead, I'll put my personal opinion of Craig Berube on the table. I was mildly appalled when they hired him even if I was not surprised, but I wanted to give him the benefit of the doubt. I first paid serious attention to him when he was about to be fired by the Flyers many years before. I used to focus solely on the worst teams in the NHL at that time (which is one of the reasons I'm not convinced by many hindsight views of the Coyotes) and the Flyers were demoralizing their good players and pushing them towards the exit. They've all moved on now except Sean Couturier and more than a decade on since he was fired there, they're getting a tiny taste of the playoffs.

I watched the Blues win the Cup with Berube as coach and I don't have this extreme negative feeling that's common to Jordan Binnington and I love Ryan O'Reilly, so it was a lot of fun, but it was pretty obvious they bored their way to winning while Binnington did the, well, Winnington. But I was willing to see if Berube plus the dynamic players on the Leafs equaled a slightly more balanced game to Sheldon Keefe's.

The meme-based response to Berube on this site has been to create a caricature of a red-faced screamer who hates everything good and spends all his time concerned only with the way the players perform their masculinity. I don't think there's any foundation in reality for that view and it's a barrier to really understanding the team.

My take on his affect was that he was trying to be boring and calm and boring in media hits to avoid having the heat turned up on the team. I find him funny, and I bet I'd like him as someone to have a beer with. I think he knows more about hockey than me, in ways I can't know it, that he has real and meaningful expertise. And I think he should retire.

The really wrong question being asked about Berube is the one that's posed as: Can the Leafs win with him as coach. Well, yeah. Demonstrably. The Leafs won their division with him behind the bench. But this discussion always digresses into ideas about how he plays a particular player wrong or he isn't right for the team but maybe he's okay for some other team. And that's the wrong conversation. The right question is: How should a winning team play?

What John Chayka has said, more than once so it's his go-to line on this, is that he wants the Leaf to be a versatile team that can win in many ways. A team that can do the things that will get them wins in multiple circumstances.

You may have heard Flyers players talking about Carolina after they got stomped to dust in their opening second-round game. They said things about how they studied the way that team plays and discussed how to counter it and then they went out and just couldn't do it. They couldn't do it a second time either.

What does it mean to be versatile? The ability to adjust to the context of who you're playing so if you are crashing headlong into the Panthers hard forecheck you have an answer, if you're being run around by the Habs, you have an answer, if you're facing the high pressure system of Carolina, you have an answer. And underneath your versatility is a foundation that is a winning system. When all the lottery balls are bouncing against you, you can still win.

I think there's another meaning there, and in reading the book Moneyball, I was reminded that the origin of baseball analytics deep in the past is largely financial analysis techniques brought to sport. Derivatives traders importing their ideas to breaking up the game into bits and measuring them. You will get a lot of finance bro language out of Chayka, by the way. What I think he says he's after is a team that spreads the risk around.

Okay some analysis. Expected Goals is a way of weighting the quality of shots with their location, type and various other factors to avoid just using the volume of them as the base measure (that's Corsi). The xG% is the weighted shots for vs against. The bigger that number is at five-on-five, the more likely a team is to win in the future. Because that is true, it has become a way of deciding if a team is "good" or not. Which is not how you decide if a team is versatile enough to win in multiple ways.

I found the 2024-2025 season a trial because for anyone who takes that single number as the good/bad blinking light, the Leafs were bad (if you picked the right variety of xG or Corsi) and did not deserve their place in the standings. And that's just wrong. The Leafs that year at 50% xG by the site I looked at first, could win playoff games, and they did. The took the Cup champs to seven games, made them work harder than anyone else managed. But at no time they did the Leafs spread their risk out. Their five-on-five xG was so mediocre they had to have good goaltending and above average shooting to succeed. There's no special award for winning while you're goalie in terrible. You aren't a better more morally sound team of higher quality if you win with a guy who is average.

It's just really risky to need very good goaltending.

The Leafs, with some power play struggles real and imagined, and a five-on-five result that needed shooting luck/talent and goalie luck/talent to break the right way to just hit even on the goals for and against was not a sustainable, reliable versatile system. It was unevenly risky and exploitable by a team that was bent on winning in the playoffs. The fact they were better at limiting quality of shots against than they had been the prior year with Keefe is not enough of a risk reducer to have made the rest of what they did a long-term viable process.

This is the question about Berube. Not can he coach Morgan Rielly to individual success or did he stifle Auston Matthews. It's not even just the zone exits – and Chayka also described the Leafs as having some opportunities to improve in terms of clean exits in the understatement of the century. This shouldn't be a thought experiment about erasing injuries and adding better players and returning to something like his first year as coach. That's better, yes, but it's not good enough. Lest we forget the Leafs hit 50% xG while tanking in 2015-2016.

Berube is and will try to create a game on the ice that he believes in. That he just thinks is right. And he's got a long track record in the NHL to draw on of what does and does not work. He's not the hapless fool many would paint him as, and he isn't Randy Carlyle making up fanciful tales for why getting massively outshot is okay. He knows it's not. But there is a point at which he will accept risk at five-on-five to keep the team defensively focused and count on the lottery balls to give the Leafs wins. I don't mean he intellectualizes it that way. I don't think he would see his system as risky at all. That's the problem. It seems safe.

For all his genuine expertise and understanding – related more than once in pressers – about the need for engagement, commitment, having the puck more, shooting more, and yes, physical play, Berube is not a guy who coaches towards achieving measurably good results in all the things that lead to winning in all game states and against all opponents. He doesn't think that way. For the entirety of NHL history very few coaches have, and they used their expertise and their intelligence to find patterns that worked. And then did all the other things coaches do like motivate and manage and put lines together and build skills. And the good ones were chosen by win/loss.

Those days are over. Hockey is a game where it takes years to figure out exactly how good a goalie is because so much of the game is random. Chayka's job is not to play in the realm of hope for lottery balls. His job is to work on the slice of the game that isn't random. Hockey is a hard job, winning is hard, building a team is difficult. It's difficult because the surgical slicing of baseball actions into derivatives works, and there isn't quite the same clean and surgical methodology in hockey. It's difficult because the nature of the game opens the door to lottery balls a hundred times in every game.

The Leafs need a coach who understands how to diversify their portfolio of options in a game, and also how to know if the strategy will achieve the goal, if the tactics will advance the strategy, and that what things seem like aren't always what they are. Maybe I'm wrong about Berube. Maybe he can be that guy, but he's only once put a team on the ice that had a good enough five-on-five game to be taken seriously. His Cup-winning St. Louis team. The rest of the time, his teams' mediocrity at five-on-five has left them mired in risk. It's pretty hard to look at last season and believe that's the outlier you throw out and not that long-ago Blues team.

Keith Pelley and MLSE decided to add John Chayka to this team. They have to let him decide who to add as coach.

Offseason