The September tradition of playing William Nylander as a centre has given way to an October where the plan sounds a little less serious but not completely tossed aside. The most recent practice saw the use of a job-share between Nylander and Max Domi with John Tavares not participating due to injury. Who will ultimately play as 2C is not known yet. Naturally the question gets asked: what problem are the Leafs trying to solve?
My answer recently was about John Tavares and his ability now to be a second-line centre, and the probability of when that ability wanes further. I think now I was wrong about some of that, so this article is part mea culpa for letting the fog of random events interfere with my reasoning when I should know better. This article is also partly about Travis Yost's attempts to address this same issue:
Travis Yost
I think this analysis goes off-track by looking at Auston Matthews and Nylander together to try to determine if Nylander can live up to his star contract. There is an attempt here to oversimplify something – understandable because the information glut if you look at the bigger picture is very hard to communicate in text. Yost wants Nylander to show he can win the minutes away from Matthews, and puts that forth as a general problem of the Leafs. He uses a WOWY over three years to illustrate his points, which are mostly about Shooting %, but also says this:
Considering the teammate mix (and the intermittent horror show that’s been Toronto’s goaltending), it is hard to place much of the blame for the Maple Leafs underperformance beyond Matthews at the feet of Nylander.
Is there an underperformance beyond Matthews? Do the Leafs lose the non-Matthews minutes as a routine? I hope you said no, instinctively. But in case you didn't, this is the picture, using Natural Stat Trick's line tool for 2021 to the end of last season at five-on-five and including all four of the key forwards on the team. (You can go there and recreate this information for yourself if you want to see all the numbers, I'm not lumbering this complicated picture with a giant chart of data.)
Auston Matthews on his own, with none of John Tavares, William Nylander or Mitch Marner played 618 minutes over these three seasons.
- Fenwick For % - 55 (the share of all unblocked shots that were for the Leafs while he was on the ice)
- Expected Goals For % - 58 (the above weighted for shot quality)
- Goals For % - 64
This is easy to understand. Matthews, without any help from one of the other four key forwards, wins his minutes by outshooting the other team by a good margin, improving on that with quality of both shots for and limiting the quality of shots against and the goals for and against are even better. He doesn't need help.
Some things to remember as this gets more complicated: Even in 600 minutes of hockey, the goals are going to appear in a random way and there can be periods of unsustainably good or bad shooting or goaltending. Those periods can absolutely stretch over a full season or big blocks of minutes played. If all you look at is goals because you believe in the worst saying about hockey ever uttered – it's a results oriented business – then you will end up thinking things that aren't true. We already know Matthews is good in all three zones and has obvious shooting skill, we don't need that goals %.
Here's our other three two key forwards all on their own, without any of the others:
- Tavares: 574 minutes - FF% - 48, xG% - 53, GF% - 55
- Nylander: 730 minutes - FF% - 41, xG% - 46, GF% - 54
You don't get Marner on his own, because it's less than 300 minutes and is just too many tiny post PK shifts or fourth line shifts stitched together to be taken seriously.
For the purpose of asking if Nylander or Tavares have won their minutes unaided, the answer is mixed, and that Nylander result is relying totally on goals to get a yes answer. The Tavares minutes are relying on quality while at least keeping the shots close to 50%. What's primarily going on with Nylander's minutes away from the – I'm going to just say it – the core, is that he's had the happenstance of excellent goaltending and a high Shooting %. That's not really explaining why he's outshot so bad, though, so I flipped over to Moneypuck and used the line tool. (This is now a different xG model, but it's the quick way to find out who Nylander plays with when it's not the core.)
Nylander played some scraps of minutes in 2021-2022 with Kerfoot and various other people that were sometimes really bad. In 2022-2023 he played some more very minimal time (less than 50 minutes) with Sam Lafferty and Calle Järnkrok or Bobby McMann. All of that should be disregarded. Adding it up doesn't make it more meaningful. There's only one line with over 100 minutes in any of these three seasons that is Nylander sans core, and it's with Max Domi and Tyler Bertuzzi, and they were horrifyingly bad. Expected Goals % of 34.
Tavares, meanwhile has some small scraps of non-core minutes with various wingers, and there are none of significant length. I think we can surmise that neither Tavares as he plays now, nor Nylander, are guys you can strap some random wingers to and assume they'll prosper. Tavares might do better than Nylander. The evidence for that is very weak, though.
With the actual core four groupings, pick any two or three of them, have generally very good results, but you get a few questionable numbers. Tavares with Marner, but not Matthews or Nylander gets you this:
- 889 minutes - FF% - 49, xG% - 51, GF% - 61
The spectacular goals rate overshadows that Tavares and Marner, which most agree should work great as a pair, are merely good enough. They have unsustainable on-ice goaltending in those minutes of .934. The Goal differential is all an illusion. It's not a huge amount of time, and the only passable shotshare could be due to other issues outside these two players, but it makes you realize why no one but the fanbase thinks switching Nylander and Marner is a good idea.
Now for the mea maxima culpa, Tavares and Nylander without the other two have:
- 1,559 minutes - FF% - 53, xG% - 55, GF% - 46
Their on-ice Save % over those considerable minutes is .896. And yes, I know you think that's because they were always hemmed in, and really bad at defence, and you saw it, and it was so frustrating. I did too. Every player gets hemmed in, and is sometimes at fault for a goal against. But Tavares and Nylander outshot their opponents and they improved that some more with shot quality for and against. They did not personally convert 55% xG into 46% Goals. And I should have known that my feelings about watching them and seeing the puck go in the net too often, particularly in 2022-2023, were making me think they didn't work well.
Their Expected Goals Against per 60 is 2.56, which is about average for the NHL over this same period of three seasons. The Leafs as a team are at 2.48 or ninth in the NHL. Tavares and Nylander are not outstanding defensively – and the other players on the ice with them get their fair share of the credit for results – but they are very good overall and most definitely win those minutes.
Yost is all about the on-ice Shooting % though, a tricksy number that isn't a true measure of any one person or the group of people. It's 8% for their minutes together. Yost posits that a player wins minutes by shooting well enough to offset the defence and goaltending, and that, frankly, is bunk. A forward line as a unit needs to possess the puck more than the other team, outshoot the other team, shoot from good locations while limiting quality against, and then they need the hand of fortune to give them the percentages worthy of their effort. Yost has it backwards.
All of the core four players in combination with each other have at least good numbers for FF% or xGF%, most are very good They even manage to be not bad all on their own. Only Auston Matthews is a one-man show, though, and he elevates everyone.
Do the Leafs lose their minutes without the core, though?
- 4,523 minutes - FF% - 50, xG% - 49, GF% - 50
That's not actually bad. It's staying even, and with quality goaltending, you can do something with a depth that manages that. But it's not helping, and as we know there are lines in the bottom six that don't work well at all.
The key number is the 4,523 minutes which works out to over 17 minutes a game. That's a lot of time to be treading water waiting for the talent to show up and elevate the team. The more minutes that are spent in that state, the harder it is to weather, not just goaltending issues, but the fickle fates which give you goals when you don't need them and keep you from scoring when you really do need it.
It would be nice if players really could just be "clutch", could score on demand, when it's needed, and try the game to right score. But that's bunk too.
The idea that a player, to be worthy of his salary, must have finishing talent and the ability to turn ordinary players into high-scoring wingers – to drive a line with two passengers of dubious merit – is the plot of an action movie. But in the actual NHL, that kind of talent is very rare. What the Leafs have in their other three forwards not named Matthews are individuals who deliver all the prerequisites for success – outshooting their opponents, outdoing them in quality and getting goals from very high shot rates and shot quality that is very good.
In other words, I don't think any of the three are particularly exceptional finishers. And if you want a hero narrative, go to the movies.
But for the Leafs to be better than they are now – overall – less time at 50% shotshare is clearly necessary. Some of that is down to a poor quality defence who weren't helping the cause. But it also leads right to the idea of spreading out the core over three lines. Not a new idea, maybe not possible to pull off with the supporting cast available, but the motivation is understandable.
I was off the mark on John Tavares with William Nylander in a general way. There are real issues with his personal shooting and possibly the way he was playing offensively over the past few years – shooting from in tight, but not scoring over a very long period. It is a sure thing that he's going to lose some of his ability to keep the puck travelling in the right direction as he ages, and playing him less is likely a good idea.
But the real problem to be solved is reducing the time spent, not losing the minutes, but not winning them either.
