Yesterday at the opening of training camp we got confirmation of the plan to try out William Nylander at centre with "Max" on his wing. To be clear, this means Max Domi.
Is this a bad idea? Well, yes and also no. If I make that classic mistake and focus in on Nylander alone, make him the POV character of my little hockey story, then I'd say it's a terrible idea. He has shown himself to be very obviously better as a winger than a top-six centre. He has more freedom to circulate in the offensive zone, he can use his speed for rush chances, and he's not tasked with as important a role in the defensive zone.
All of which is true of most players. There aren't all that many who are better as centres. I think Auston Matthews clearly is, and John Tavares has a lot of skills that make him better as a centre, and no one thinks Sidney Crosby should move to the wing. But in general, forwards do better at scoring if they can focus on scoring.
Nylander is not the POV character, and aside from Auston Matthews, few players on this team are locked into any role so it's not wrong to try it.
How do we measure Nylander as a centre, though. He only does it every year in training camp and then everyone puts the pumpkin spice away, he goes back to the wing, the Leafs have better offence, and we forget about it for another year unless emergency injuries bring him back for a game or two to the middle of the ice. He's never looked either unplayable or particularly great. He has seemed uncomfortable about where he should go in the offensive zone but I can't look at key information like his shot rate or location.
In his normal role, he isn't terrible on defence – although he and John Tavares had one terrible year recently that seemed like it was something beyond either of them as individuals. But he's no two-way guy, and that's why this idea seems doomed every year to a preseason experiment that doesn't catch on.
Craig Berube said Nylander has the puck moving skills to carry the puck up the ice for them, and well, I've made the 1-800-Zone-Entry jokes before, there is almost no one better at bringing the puck across the offensive blue line with control, and his rush game is also exceptional. None of that requires him to be the centre, though, so that explanation seemed odd.
But look at the roster as a whole, and you can easily see why this is going to be tried out again. Maybe just a little more cloves in the mix this time and everyone will like it... we'll keep it around all year.
The serious players will be split into two groups for training camp and each preseason game will have its playing roster drawn largely from one group – at least at first. This gives the centre options as:
- Matthews
- Tavares
- Kämpf
- Holmberg
Minten- Some guys who are AHLers
You can see how it's very easy to balance those groups if you move Nylander to centre, just like last year. And last year, it was very clear this was an idea coming from the GM that Sheldon Keefe gamely trotted out and then was flushed with joy that Fraser Minten was almost ready for the NHL and let him end it quickly.
With Minten out for weeks there isn't any obvious way to avoid a repeat of last season with the underpowered, underperforming third line with a rotating cast of players, including that period of Max Domi as 3C experimentation. The line had no identity – in the true sense of the word, not the weird hockey jargon way that means play hard.
Once preseason games are over, the calculation becomes a little different, and it's about the way lines interrelate. So now I want to talk about a new feature at HockeyViz. It's part of the SG model which attempts to calculate the value of all of the various player skills using one currency, called SG.
Since that guide was published the model has been updated with a measure of a player's tendency to gain or lose territory in their shifts. Auston Matthews – who is so elite at almost all skills that people can actually forget the extremes of his skill – has a positive impact in both ice won and ice lost. That is, that when he starts out in the defensive zone he goes north, and when he starts out in the offensive zone he stays in the north. His total of the two is .5. For context: This is better than Connor McDavid, who is at 0.3, and the same as Aleksander Barkov, likely considered the best two-way centre in the game. Matthews scores too much or he'd win the Selke every year Barkov doesn't.
What this means is that the line that comes out after Matthews' line is going to be better off for it. Which means for the Leafs, that their second line has by default been put in a position to succeed more than if they were just started at centre ice every shift.
William Nylander has a modest .1, all coming from not losing ice last year, a year where he played mostly with the much-maligned John Tavares. Tavares was at -o.1 at the end of last year, so slightly worse than Nylander.
This is one small segment of overall skill, but it's one that matters in choosing the order of lines, or in the Leafs case, deciding who is on the second line since the first is set.
The obvious secondary benefit of Nylander as 2C is Tavares as 3C. I think the Leafs might be aiming for a line of Max Pacioretty - Tavares - Calle Järnkrok. And they aren't going to be fast, but they are going to be able to score, be good positionally, decent defensively. They are big but not bruisers. At least they have a definable role: keep the intensity up, play hard, get some goals, stay out of the box, let the fourth line bang some bodies.
I'm assuming that the top-six wingers in this configuration are Mitch Marner, Max Domi, Matt Knies and someone. It will look like that someone should be Easton Cowan I'm guessing. Or Nick Robertson – depending on who scored in the most recent game. But don't count out Bobby McMann. Maybe Pacioretty has more to offer now after a full offseason than he did last year.
There is a whiff of a Vegas alignment here with a lot more real offensive power on the third line than most teams bother with. There's also a whiff of the kind of "roll three lines" concepts fans come up with that just look too spread out, too much like the result is less than the sum of the parts.
But one advantage of a Vegas approach is that the third line spends a decent amount of time playing a lot easier competition than the top line does. When some coach on the road is playing the matchup game, they are going to go hard at line two and suddenly there's two other options. When Berube takes Matthews and goes hard against someone like the Barkov line, suddenly there's two other options.
The attractions to a system like this are really obvious. I think they're one winger away from it being plausible, but what if Knies plays all the time like he did in the playoffs – h/t to someone whose name I've forgotten who brought this up recently in comments, but you are very correct, he was a horse in the playoffs, and not just to the eye. What if Cowan has dropped those junior habits? What if McMann has enough complementary skills to be a good fake top-six guy like Michael Bunting was for one season?
I'll be surprised if we're doing Nylander the centre in January, but you never know.
By the way, David Kämpf's ice won/lost rating was at .3 last spring. If you were wondering what it is coaches see in him – other than the actually very good defensive outcomes. If he's setting up Matthews, then this whole idea just got a lot more plausible. And in a very real way the roll three lines concept needs a fourth line of a particular set of skills that isn't about some guy you play for six minutes. The players are there to make a more sophisticated fourth line (they gave Steven Lorentz a real number, which is a tell he's got the inside track on a contract), but there is also the six minute guy sitting there eating more than his share of the salary cap chilli.
You shouldn't end every look at a topic with the same conclusion, but alas, I'm forced to by circumstance. Take Ryan Reaves and Max Domi off this team and spend four million on someone who could ride beside Nylander and add value all over the ice, and you have a concept touching plausibility. A little less pumpkin spice and a little more a real espresso.