Writing about Wyatt Johnston has turned into my own Fast and Furious franchise. If I’ve got a lot to say about the 20-year old it’s because there’s a lot to cover. There is the “eye test”, to give us a sense of how his abilities stand alone. There is the historical record, to give us context about where belongs among his peers. And there are the fancy models churning their gears, to give us a projection of where he’s headed. Added up, and we’re looking at a special player who needs no introduction to Dallas Stars fans. And yet I feel a little unsatisfied. Why is Johnston next to Jamie Benn and Evgenii Dadonov such a sacred cow when Dallas has the best offensive depth this franchise has arguably ever seen?
If it ain’t broke, don’t fix it. If it ain’t broke, don’t fix it. It ain’t broke. If it ain’t broke, if it ain’t broke…
That’s what everyone keeps telling me. Hey; you might be right. You probably are right. Johnston’s post-deadline trio was a good one. Of hockey’s most potent trios with at least 200 minutes of icetime together, the Johnston line ranked 7th in goals-for per 60. They have a solid grasp of time and space, knowing when to stay wide and when to choke the defense, anticipating each other’s every offensive move, forging chemistry, etc. According to most fans, this line must stick together.
That was the undeniable conclusion when I posted the poll asking exactly this question last week. (And no, I can’t get used to that stupid midlife-crisis-designed logo in the top right corner either.)
Speaking of, help me write this bad boy. What's the ideal Wyatt Johnston trio, in your (measured) opinion?
— David Castillo (@DavidCastilloAC) July 26, 2023
Obviously, I’m here to dispel the notion that 14-53-63 is undeniable. But I always like to be mindful about what conclusions can be drawn from whatever argument I’m making. In this case, what follows will not be an argument to break them up, or call them bad. What follows will be an argument about where they fall short, and if there’s a better alternative. Nothing more. But first let’s start with something of a shocker. But before I get to that here’s a quick sidebar, because I’m a big believer in Words Mean Things.
When I use the word “offense” I’m not talking about just goals. I’m talking about the stuff of goals: all the ingredients baked into what makes them possible. The passes, the movement, the routes, the chances, etc.
So about that shocker: offensively, the Johnston line is just kind of…okay?
As you can see, they’re super dangerous in the high danger area of the ice, which is in keeping with everything we’ve seen. However, it seems to come at the cost of everywhere else. In fact, Dallas is two percentage points less likely to score with them on the ice according to Micah’s model. For some fans, this might sound silly. “I don’t get it. They outscored their opponents, and didn’t you even mention that they had a super high expected goal share?” Yes. And thanks for reading.
It’s true that they outscored opponents at a high rate — 17 to 11 in the regular season — and that they controlled the shot quality battle (well, according to other expected goal models). These are both good things. But offense is about more than scoring. In their 23 regular season games together, their shot attempt, or Corsi For, plus/minus was +7. That might seem good, but it’s extremely average. Of nine regular trios, that ranked seventh on Dallas’ squad, just below the Mason Marchment-Tyler Seguin-Denis Gurianov line. If they were so average offensively according to the model, then why were they so productive? And the answer is this: 15.37.
That was their shooting percentage. It was the third best shooting percentage among all NHL trios with at least 200 minutes of icetime together. It was so good that not only was it the third-best shooting percentage of all forward lines last season, you could put them in any season since the analytics era (2007), and they would still be top three. Granted, there’s a larger discussion about shooting percentage and repeatability, and certainly, the Johnston line profiles like a line we’d expect to create their own luck. Benn and Johnston are two excellent shooters, and Dadonov is an excellent playmaker. 2+2=4, right?
That’s the best case to keep them together: they’re not perfect, but they’re perfect for each other. I don’t have a problem with that logic. However, how is the belief that they will continue to defy the shooting gods any more theoretical than the counterargument that they can still be effective away from each other? 15 percent is a lot. If you’re bad at math like me, the number alone doesn’t give you any context. For perspective, their shooting hand was five percentage points more dangerous than the Hintz line from two years ago (a 10.5 shooting percentage) and three percentage points more dangerous than the Hintz line this past year (11.74). Benn-Johnston-Dadonov are good but they ain’t that good.
It would be one thing if the Johnston line had something else going for it, like strong playdriving (they are not: 48% in shots on net share, 48% in unblocked shot differential, and a CF of 50%), or strong defense (their 45.29 unblocked shots allowed per 60 was the second worst of regular forward lines), but they have none of that. What they have is offensive chemistry, and offensive chemistry alone. Is that enough to offset their other flaws? Possibly. Again, I don’t want to understate how dangerous they are. Just because a high shooting percentage is due for regression doesn’t mean it’s due for a dramatic fall. In some ways, their unlikely shooting percentage is also a signal of what they’re capable of.
“So what do you propose?”
I’m not gonna advocate for the original trio of Johnston with Benn and Ty Dellandrea but I do want to note that they were actually…slightly more dangerous.
Nothing crazy, but slightly better shift-to-shift offense. (The fact that both trios are so dangerous in the slot is a big reason why I’d argue that despite Benn’s gaudy point totals, it’s Johnston driving that line.) Dellandrea’s in a strange spot. He’s largely considered expendable right now if the healthy scratches during the playoffs (and the poll) were any indication. And for sure, there were times when he played his way into the doghouse with his unconscionable penalties during the playoffs. But Dellandrea also showed some genuine chops as a multidimensional threat. It’s worth remembering that Dellandrea was only one game shy of being a rookie. For as much grief as fans give Nill for trusting veterans over prospects, I’m kind of surprised so many fans have disregarded Dellandrea as a middle six option. Two points shy of 30 for a functional rookie seems good.
However, as mentioned, I’m not advocating for that line to get back together either. It’s just a useful illustration of what I think is the low down dirty truth of the group: Johnston is its prime mover. He’s the guitarist, and the leadsinger. In a vacuum, this argument doesn’t amount to much. “Ok we get it. Johnston is good. But so are Benn and Dadonov.”
Sure. But the question is not about how good they are together, but whether they can be better. The other, more important point, is this: if Dallas were your average team in no rush to compete, make history, or accomplish much of anything, then this would be fine — just a feelgood story about a young player being helped along with two elders getting their second wind. But any team this close to making history has to weaponize each and every strength, including what’s left in the margins. In the playoffs, those margins matter. If Johnston is the 60-point player I believe he will be next season, then what is it worth to maximize that player?
I love these WAR Roster Builders from JFreshHockey. Simplicity doesn’t automatically imply a lack of complexity. While these numbers are not arbiters of value, they shouldn’t be taken with a grain of salt either. Each number is configured by adding up how much each player is worth in the standings. However, how each number is determined comes from a comprehensive list of factors that include the ripple effect they have on both sides of the puck. I don’t want this point to get lost from the ease with which we can reference these: these numbers don’t reflect what happened when each team scored while they were on the ice (like the traditional but lazy plus/minus), but what happened from shift to shift. With that in mind, I thought it’d be useful to input various Johnston lines.
I tried the following. (Total value in projected wins is under “Proj. WAR”)
As you might have guessed, replacing Dadonov with Duchene projects to have an even bigger offensive impact. The line loses a decent amount of defense (Duchene’s never been the most defensively responsible forward), grading out as the worst trio in terms of defense. This might actually be the best argument for keeping the Johnston line together: there’s no other combination that sacrifices enough defense to offset the offense it creates, and vice versa.
However, the real question is this: if Dallas were to split up the old Johnston line, what would be the likely knock-on effect for Seguin?
This, to me, is where it gets interesting. While Duchene next to Seguin is predictably the best possible trio, Dadonov next to Marchment and Seguin profiles like the Johnston line, minus the bad defense. I know that might be a shock to some fans given how much of a whipping boy Marchment became, but over the last two seasons, he’s not the one giving up copious amounts of chances.
However, I don’t want to get too sidetracked.
To me the most important takeaway from the different Seguin trios is that they all grade out as having more projected net value as a group than Benn-Johnston-Dadonov. Seguin may seem like an odd candidate to be the glue on any given line (he didn’t have the big year that Benn had, after all), but he was quietly effective, and unlike Benn, not a defensive liability.
The reason why I worry over things like this (beyond this being slow season) is that I don’t like the theory-induced blindness to it all. When you accept that something has value (in this case, the 14-53-63 trio), that’s fine. But when you turn it into a tool in your thinking — i.e. 14-52-63 is a key part of Dallas’ depth and Johnston’s development — then suddenly it becomes harder to notice its flaws. I would argue that is exactly how players like Ryan Suter and Colin Miller came to be acceptable partners for Miro Heiskanen. Heiskanen is the blueline’s ultimate equalizer. “It doesn’t matter who you put next to him!” Right?
That’s the irony to the logic of “if it ain’t broke don’t fix it.” What was wrong with Benn-Johnston-Dellandrea? Nothing, especially early on. But most would agree it wasn’t necessarily optimized, yes? It’s great that Johnston is on a line that can threaten. But Johnston is now one of the big boys. The effectiveness of his line could determine how many more minutes he plays. In that context, the degree to which his line is effective will matter more than whether or not they are.
It’s not “good” that wins Cups. It’s “great” that does. Given Dallas’ depth, why not dare to try?