I suppose it’s time for the annual conversation about whether Auston Matthews and Mitch Marner should play together.
Whether they are together or apart, this debate has taken place every season across three different coaches. We have watched the main characters of this story for so long now, but I’m not sure the dynamics are always framed correctly.
Let’s start with the simple facts regarding the possible combinations of Matthews, Marner, Nylander, and Tavares.
The Matthews-Marner duo
First and foremost, the best possible combination the Leafs can assemble is pairing Matthews and Marner together. There should really be no debate about this; at least in the regular season, there is no real argument against it.
Since the start of the 2022-23 season, Matthews and Marner have played over 1,400 minutes together at five-on-five, and when they do play together, they almost exclusively play head-to-head against the opponent’s top line, or at worst, their shutdown line. In those minutes, they are controlling 55.40 percent of the shot attempts and 57.22 of the expected goals, and they have won their minutes with a ridiculous 87-55 goal difference. It’s a massive sample size of elite results.
Stylistically, it makes sense as well. Matthews is the best goal-scorer in the league, and Marner is one of its best playmakers. It helps that it’s a lefty and a righty who can pitch and catch with the puck, so everything is seamless. At the other end of the ice, they are both excellent defensively.
The Matthews-Nylander duo
When the team swaps Marner for William Nylander, the duo is still good but not as good. The numbers bear this out. If we stick with the same three-year sample, Matthews and Nylander have played just under 830 five-on-five minutes together, owning 52.73 percent of the shot attempts, 55.06 percent of the expected goals, and – notably – winning those minutes 57-31 in actual goals.
There are a few caveats worth noting about this comparison. The first is that when Matthews and Nylander are paired together, they don’t necessarily go head-to-head against the opponent’s top line. When the Leafs run it, they usually have the alternative Tavares-Marner duo match up against top competition. They did this most recently against Tampa Bay, as Tavares-Marner took on the Brayden Point matchup. In seasons prior, Sheldon Keefe did this, too, including last spring in the playoffs when Matthews and Marner were on separate lines.
Matthews-Nylander see easier matchups, control play at a lower level, and score slightly more.
Nylander, like Marner, is also a right-handed shot, allowing for seamless plays, as we saw recently when Nylander and Matthews sliced through the neutral zone together, leading to one tape-to-tape pass on the forehand that went for a breakaway and easy goal against Chicago.
Because Nylander is an excellent shooter, he can draw in defenders, creating extra space for Matthews. The opposition can’t cheat toward Matthews as much as they would with Marner on the ice.
At the same time, Nylander is simply not as good defensively as Marner and not really a player they should or would use in a direct head-to-head matchup night in and night out. Nylander is best viewed and utilized as an elite offensive game-breaker.
The Tavares-Nylander duo
So, if both combinations are so good – in slightly different ways – what is the issue?
Over the past three seasons, when Nylander is on the ice without Matthews or Marner, he’s dead even in his minutes with 82 goals for and 82 goals against. We can slice it any way we want, but it’s a big sample – over 1,700 minutes – and the bottom line is he’s not winning his minutes away from those two.
When Nylander is not on the ice with Matthews, he generally plays with John Tavares. Since the start of the 2022-23 season, Tavares-Nylander have been outscored by one – 52-53 overall – in five-on-five goals. That is not to suggest the pair has played poorly, as they have generally won their minutes in the run of play (52.36 percent of shot attempts) and in scoring chances (56.23 of expected goals), but it hasn’t led to any sort of distinct advantage where it matters most.
Naturally, this leads to this conclusion: If Matthews and Nylander are good together and Nylander’s not winning his minutes away from him anyway, why not just pair them together and run Marner with Tavares?
There is a logical thought process leading to this conclusion, but this is where the issue becomes more complicated.
The Tavares-Marner duo
On the surface, the Tavares-Marner duo has knocked their minutes out of the park. They are up 43-22 in actual goals (while together *and* without Auston Matthews) in over 835 minutes together since the start of the 2022-23 season. So far this season, they are up 7-1 together in goals. Those are significant margins in the Leafs’ favour.
However, controlling the run of play has not gone nearly as well compared to any other combination. Since 2022-23, they’ve owned just 51.83 of shot attempts and 51.87 percent of expected goals. This season has been particularly underwhelming in this department so far, as they’ve controlled just 44.5 percent of the shot attempts while getting repeatedly hemmed in their zone for long stretches. That said, they have kept the threat largely to the outside and are actually slightly above water in expected goals at 50.88 percent.
Controlling the run of play is important, though, especially knowing the Tavares-Marner duo has been leaned on as a matchup line. It’s one thing to suggest, “Sure, they haven’t been great territorially, but they are winning those minutes over a large regular-season sample, so who cares?” It’s another to think it will work out in the playoffs.
We’ve seen this pair exposed in the postseason for two years now. In the last playoff series against Boston, the Leafs ran with the Tavares-Marner duo in a matchup role; they were outplayed across the board and outscored 2-1. The single-goal difference is not the end of the world, but the fact that they only scored one goal as a duo is problematic for a team that has struggled to score. For as much heat as Marner took in the playoffs last spring, it’s interesting to note that without Tavares, he was actually up 2-0 in five-on-five goals.
In the 2023 playoffs, this duo was again outplayed territorially and on the shot chart, although they managed to win their minutes 4-2 overall. Even this comes with a caveat, though, as the Leafs started their second-round matchup with the idea of running Tavares-Marner against the Sam Bennett line, and they couldn’t even last a game. They were dominated in Game 1, and Keefe never returned to this matchup.
It’s one thing to deploy Tavares-Marner as a matchup unit against the Chicagos and Nashvilles of the league (or even Tampa Bay without Nikita Kucherov) throughout an 82-game regular season. It’s another for them to do it every other night against another very good team in the playoffs.
Ultimately, Tavares and Marner are skilled enough to be opportunistic offensively and smart enough defensively not to give anything away for free, but they don’t combine to control play well enough. In the playoffs, the importance of controlling the run of play is heightened as special teams opportunities can dry up and the margins become very thin against top competition.
The 2024-25 results so far and Matthew Knies’ emergence
So far this regular season, the story somewhat follows the same script.
The Leafs posted a great record when Matthews was injured, but their five-on-five play left a lot to be desired. Nobody in the league scored fewer goals at five-on-five than the Leafs after Matthews went down; the team scored nine five-on-five goals in nine games without its captain. The next lowest totals belonged to the Chicago Blackhawks and Calgary Flames, who each scored 15 (albeit in 10 and 12 games, respectively).
Territorially, the Leafs were a lottery-level team without Matthews, controlling just 43.87 percent of shot attempts. They did a good job of boxing out defensively, so their expected goals numbers were much better (and even slightly above 50 percent). Still, they couldn’t sustain any pressure and generally followed the lead of their de facto top line in that regard.
When Matthews returned, Craig Berube kept the Tavares-Marner duo together and started Matthews alongside Nylander. The team won the first two games, but they were out-attempted at five-on-five in five of the six periods – against a Tampa team that played the night before and was missing Kucherov and Nick Paul, followed by a Blackhawks team that also played the night before and is one of the worst teams in the league. After two listless periods against Nashville, Berube pulled the trigger and reunited Matthews-Marner, who immediately scored two goals.
In past years, I have sometimes argued that the Leafs should split up Matthews-Marner. Just last year, I wrote in favour of it around this time in the season as I didn’t think the juice was worth the squeeze. Around this point last season, Matthews and Marner weren’t tilting the ice enough together, and they were up only 14-13 in goals.
The circumstances have changed across the board so far this season.
One reason I thought the Matthews-Marner line struggled last season, especially early on, is that they lacked the forechecking support they previously received from Michael Bunting or Zach Hyman. Tyler Bertuzzi couldn’t keep up with Matthews and Marner from a skill perspective, and it didn’t work out as a line. Eventually, Matthew Knies was moved up, and while he showed flashes of promise, he was a rookie who scored 15 goals and 35 points while navigating the learning curves of a player’s first full NHL season.
The simple fact is Knies has taken a step this season. He’s playing over four minutes more per game and asserting himself physically and as an offensive creator. He’s shooting an astronomical 24.4 percent, but he already has 10 goals and six assists. Knies is pacing to push for 60 points, and the weight he added over the summer is helping him dominate games physically.
Knies started to show signs of this in the last playoff series, where he scored several big goals. He has seven points in 14 career playoff games already despite averaging just 14:06 in ice time.
But what about Matthews-Marner’s history in the playoffs?
A common argument against pairing up Matthews and Marner is their past playoff performances. They haven’t exactly taken over a playoff series together. That’s fair and true, but two things are worth pointing out here.
Firstly, they’ve won their minutes by and large. Over the past three playoffs, they are up 14-8 in goals. But they have fallen short primarily when the chips are on the table in series-deciding games, and there is no getting around it.
Secondly, as mentioned, Knies is emerging. If we review Matthews and Marner’s full playoff history together, Knies is arguably their best linemate to date. Michael Bunting’s act was exposed in both of his playoffs in Toronto. The team tried all sorts of other options – Nick Foligno, Wayne Simmonds, Andreas Johnsson, etc. – that were not on Knies’ current level.
And then there’s Zach Hyman. He is an excellent player, but he wasn’t productive in the playoffs in Toronto with 13 points in 32 playoff games as a Leaf (yes, his production with the Oilers in the playoffs is tough to look at as a result). Knies has already been more of a gamer than Hyman ever was here, and his game is only getting better. I’m not suggesting he’s a better overall player than Hyman, but what he’s providing the Leafs is better than what Hyman gave them, particularly in the playoffs.
Matthews has started this season slowly, and their results as a line are still basically the same as last season when Matthews scored 69 goals and Knies was a rookie learning the league. What happens when Matthews gets rolling?
Adding – or, in this case, developing – a really good player to the line makes a massive difference. The Knies-Matthews-Marner unit has the goods to be one of the best in the league, and that would be a significant advantage in any matchup or series the team finds itself in.
Tavares-Nylander’s left winger
Of course, this leaves the Tavares-Nylander duo on the second line. As noted, the pair haven’t really excelled at five-on-five. However, their left-winger situation has also changed the dynamic.
When Tyler Bertuzzi lined up with them last season, he was a forechecker and net presence for the unit, and the trio won their minutes decisively – 54.11 percent of the shot attempts, 59 percent of the expected goals, 16-12 in goals.
The Leafs don’t have Bertuzzi this season, but with Max Pacioretty in his place, they’ve posted great results – 57.78 of the shot attempts, 67.91 of the expected goals, and 5-2 in goals. It’s an incredibly small sample size with Pacioretty, but as a continuation of what we saw with Bertuzzi on the line, when they have an effective forechecker handling some of the dirty work (rather than Alex Galchenyuk or Alex Kerfoot), they have been much more effective.
Even with Bobby McMann on the left wing, Tavares and Nylander have shown well since the start of last season. In just under 100 minutes together, they’ve come out ahead on shot attempts, chances, and goals (7-3).
Only the sample size with Bertuzzi is notable, but we’re seeing evidence of a power-forward type of profile elevating the line across three different players by filling in gaps and complementing Tavares and Nylander’s strengths.
In Summary
I’ve spilled a lot of words to arrive at this conclusion: In past years, it has generally been fair to suggest Matthews-Marner is very good, Matthews-Nylander is really good, Tavares-Marner is good-to-really good, and Tavares-Nylander is middling. This made it reasonable to split Matthews and Marner up in order to raise the floor of the top two lines. As the supporting cast has evolved – and the three Leafs stars have rounded into their primes along with Tavares entering his mid-30s – I’m not so sure this is true any longer.
Knies’ emergence is a big deal, and so is finding a fit to improve the Tavares and Nylander duo.
I’m not suggesting Berube should lock into these combinations and never move them around again. There will be times when they need to shake it up or there are matchup dynamics at play that make it sensible to flip Marner and Nylander (or even split them across three lines). The team needs different looks it can can throw at the opposition and make adjustments with.
Part of what made the start to this Leafs season so frustrating is the team played several poor games – most notably against St. Louis and Columbus – where every forward line changed except the Matthews-Marner combo. They shouldn’t be exempt from mixing it up, and this recent run of play without Matthews should show the new coaching staff that they have options, including splitting up Matthews and Marner when it isn’t going well. But I’m not sure we’ve seen enough to suggest the team is actually better off if they split them up regularly.
Now, this could change as the season progresses. If Brad Treliving is able to add a quality forward, we’ll likely eventually talk about splitting up their talent across three lines, too.
However, based on today’s roster, there is plenty of reason to believe the Leafs’ top line can dominate play and head-to-head matchups. There is also reason to believe the team’s offensive second line can sustainably win their minutes to support them. As long as both lines consistently do their jobs, they should be kept together. If and when they don’t, Berube will need a counteralignment to turn to.
