On Thursday night, the Toronto Maple Leafs surprised much of the hockey world cruising to a 5-2 victory over the first-overall Nashville Predators, a team that hadn’t lost in regulation or overtime in over a month and had just four regulation losses in Calendar Year 2018.

Perhaps the most repeated stat in this game by the Toronto-area broadcast team was that Toronto’s most polarizing defenceman, shutdown bruiser Roman Polak, had been leading the team in Time on Ice for most of the contest. Now, in fairness to the Sportsnet crew, particularly the intermission panel, they were pretty transparent about the fact that they were doing it to ruffle feathers, rather than to make a point.

But I’ll use it to make a point. Don’t worry, the point isn’t to slag on Polak; there are different occasions for that. In fact, one could argue in a vacuum that he earned those minutes last game; he was near the top of the Leafs defensive corps in on-ice shot differential, his only glaring “mistake” involved him being stuck in a bad spot because of others, and he drew a penalty. Honestly speaking, it was one of his better games this year.

But that’s not why he got the minutes. At least, not entirely. The minutes were situational; the Leafs led 47 out of 60 minutes in that game, never trailed, and were on the penalty kill six times. It was a perfect storm for the 31-year-old, who plays under a coaching staff that has its fallback rotations, particularly when they’re looking to protect a lead or come back from a deficit. Instead of just telling you this, though, let me illustrate it for you, by showing you the Leafs’ usage of players in various score situations, through both time on ice and zone starts.

Right away, we see some patterns here. Not massive, wild swings, but a few notches nonetheless. You get a general idea of who the Leafs trust to play in the defensive zone to protect a lead (Ron Hainsey and Nikita Zaitsev), what they do when the game is tied (shorten the bench just a bit to emphasize the top two pairs), how they adapt when they need to catch up (Connor Carrick starts to play!), and what they see in Travis Dermott (a player they can use anywhere).

Generally, it’s about what you’d expect. Hainsey and Polak being used to protect leads, Carrick and Borgman, used to pull back into games, and the Rielly, Gardiner, Zaitsev, Dermott quadrant getting Swiss Army Knife minutes. With that said, having just six defencemen to work with means that these gaps tend to be a lot closer; guys get tired, especially the special teams contingent, and sometimes you’re throwing a player over the boards because a percieved-lesser choice that is fresh is better than a percieved-greater choice that is out of breath.

With that in mind, it’s probably no shock that the patterns are much evident in forwards. The Marleau-Kadri-Komarov shutdown line (now featuring Mitch Marner most of the time) was a go-to as a lead protector throughout the year, with the Hyman-Matthews-Nylander trio not being far behind. Van Riemsdyk-Bozak-Marner (now Brown or Komarov) have been looked to much less in those situations, while the fourth line of Martin-Moore-whoever that started the season almost never got a look while leading (or lets be honest, down, or tied) to start many shifts.

Even with these numbers being rate-adjusted, Moore and Martin are never sent out when the puck is in the zone most important to the score situation. More so than any other regulars on the team, those two are sent out either in neutral-zone faceoff situations, or on-the-fly, when other options are exhausted.

On the other hand, the numbers flip a bit when Toronto is trailing. The Hyman-Matthews-Nylander line gets the most minutes, likely because the best player and scorer on the team is in the middle of that line. But that seems to come on-the-fly as well, with the Bozak line getting the most offensive zone starts when trailing, and the Kadri line getting the most defensive ones, likely focusing on both trio’s strengths and purposes.

While the sample is small and early, there is perhaps a shift to be seen in fourth line usage based on some of the lesser-used, recently integrated forwards. Kasperi Kapanen’s total zone start rates look closer to the rest of the team’s average than Moore and Martin’s were, and Tomas Plekanec’s shutdown abilities have been put to the test pretty frequently since getting here.

TOI% Interlude

Something I experimented with while making these charts was seeing what percentage of players’ time on ice was in what score situation, to see if that would give an even more blatant picture of how certain players were used. The results were very muddy, which after some thought, makes sense; a lot of players on the list have played enough minutes and games to get an idea of their own individual performance, but the team’s results and records would vary a ton from player to player, meaning not everyone was on the shortlist to go over the bench at the same time.

There is enough to show about three lines of consistent data, though, as Toronto’s forward core has stayed spectacularly healthy for the second consecutive season (sorry, Josh Leivo). Nine players have missed three or fewer games up top, and their spread looks as follows:

Again, some easy typecasting here. Kadri, Brown, Marleau, and Komarov as the lead protectors. Van Riemsdyk, Bozak, Marner, Nylander, and, once again, Brown (a product of being integrated into other lines as relief while with Martin and Moore, most likely) as the rally squad. A little tighter in tie-game situations, save for Brown.

But hey, let’s go back to that original point about ice time and records. You know, the one about how Polak played so much on Thursday because the team was winning the whole time.

Often, in Toronto or any other market, you hear people bring up a team’s record with a certain player in the lineup. That’s usually a bunch of noise, though I did include it in the last tab for people who were curious. I doubt that Connor Carrick and Travis Dermott’s presence are all the Leafs need to be a 119 point team, and I’d imagine that Josh Leivo having the worst record out of the bunch has a lot to do with the fact that 10 of his 16 games played are a direct result of him having to replace Auston Matthews.

But we’re going to better address a more nuanced version of that, and that’s record tied to ice time. You hear things like, “The Leafs win more when Komarov plays more! That means he’s a key piece!” or “Their record is great when Van Riemsdyk plays 13 minutes or less, must be a rest thing”. These are common sorts of takes, but they ask the wrong question. Are the Leafs winning or losing because the players are playing less or more, or are they playing less or more because the team is winning or losing?

That’s what the chart directly above aims to illustrate, with the prior knowledge of what we discussed previously. I’ve done two things here; I’ve divided up the season of every regular player (really stretched it with Leivo, I know) into three based on ice time, and also looked at their ice time in wins and losses.

You’ll notice a few things. For example: when forwards like Komarov, Brown, and Kadri, play more, their records are way better. When forwards like Van Riemsdyk, Bozak, Matthews, Nylander, and Marner don’t play very much, that’s when they have better records. On defence, Toronto is the greatest team to ever play hockey (a 147 point pace, undefeated in regulation pace!) in Ron Hainsey’s top 24 games of ice time, but have a points percentage lower than the Red Wings in his lowest 24. In Roman Polak’s lowest 24, they’re, fittingly a team on a 46 point pace.

So what does this mean? Does it mean that the key to a Stanley Cup is to run Komarov-Kadri-Brown with Hainsey-Polak for 25 minutes a game? Should the likes of Van Riemsdyk, Nylander, Matthews, and Gardiner be sheltered instead? Not at all. It just means that Toronto’s lead-protectors will likely have their most active nights when they’re, well, protecting leads, and that their top scorers are used to get them back into games that they’re trailing. If a JVR is winning a ton of games where he plays 12 minutes, its because he and his teammates have accomplished their end of the bargain early, allowing for Kadri & co. to close things out and grind their way towards the 20 minute mark.

So do you have a conclusion or what?

Usually, a post like this comes with a suggested improvement. This isn’t one of those. I could tell you about how I feel about lead protection (Spoilers: I’m on Team “Keep Scoring”), but these charts probably don’t contribute much to that conversation, unless you’re suddenly alarmed by the fact that the best Leafs team in over a generation seems to have a set objective of “get the superstars to put us in a position where we can put the game in Hainsey and Polak’s hands”.

More than anything, this is just a collection of observed information, to outline how these usage discrepancies happen. Perhaps it gives you something to consider the next time you’re wondering how Leo Komarov played 20 something minutes in a tight game, or why you didn’t notice Tyler Bozak in a blowout win. Alternatively, it might help explain why close regulation wins seem to end in a stressful fashion, compared to the begrudging feeling of entertainment that comes at the end of close losses.

---
Thanks for reading! Hopefully you enjoyed this post. If you did, don't hesitate to share it on Twitter or Facebook; having more readers will help the site grow. As well, consider a subscription if you're interested in reading additional work that isn't available to guests.