It’s been a while since I sat down and did some high school math level analysis of the Leafs, and that’s exactly what I did last night. I was curious about what parts of the rink Leafs players are thriving in and how tilted are the Leafs towards being an offensive team.
I decided to start by putting the 5v5 for and against stats for Corsi, goals, and expected goals into league-wide percentiles (with a minimum of 200 minutes played) and then looking at the trends around the for and against stats while factoring in time on ice to identify any excessive sheltering. (For example, Nic Aube-Kubel had strong defensive numbers, but was heavily sheltered, only playing in the 7th percentile of 5v5 ice time.)
I broke the players into the following groups with the following definitions:
Advertisement
Ad
  • 200ft player: They would need to be above the 70th percentile in both for and against stats, as well as being in the 70th percentile for time on ice. Basically, you can’t be a 200-foot player if your coach is afraid to put you out there.
  • Offensive player: You are in the 70th percentile or higher in “for” stats. There isn’t a time on ice restriction here, but certainly there are offensive players who you can make an increased case for sheltering. John Tavares had the lowest against numbers of the Leafs’ offensive players, followed by William Nylander. That is probably a big part of the case for splitting them up and/or the reluctance to add someone like Nick Robertson to their line and truly making them a line you only want to play at one end of the ice.
  • Defensive player: Pretty much the exact opposite of the offensive player, but with the requirement that they were at least at the 20th percentile for 5v5 ice time. That at least gets them into 3rd line utilization.
  • Low event player: Any player whose events rates combined are some of the lowest in the league. There is no time on ice restriction here, but it should be noted there is a huge variance between Calle Jarnkrok as a high utilization low event player, and Adam Gaudette as a low utilization low event player.
  • Bench player: These players didn’t really excel in any area. They are pretty much replacement-level, bubble players. Maybe it was a bad year on a bad team that skewed the results, but on the surface, they should be on the outside looking in.
Now I’m not saying that 200 ft. players are the best, and just because it is the group that features both Auston Matthews and TJ Brodie, it seems like it could be intended to be taken that way. It’s also encouraging that Mitch Marner was very close to consideration for that category, as was Timothy Liljegren. There’s definitely good reason to be excited about both Liljegren and Sandin, and Marner being classed as a 200 foot player isn’t going to cause anyone to bat an eye.
Another thing that is important to note is that David Kampf and Mark Giordano could just as easily fall into the low event player category as they do the defensive player category. The fact that they are low event defensive players makes me love them even more, and the fact that Giordano and Jarnkrok both have their numbers heavily influenced from being on a lottery team last season shows a lot of grounds for improvement, the preseason eye test should confirm some of that.
Advertisement
Ad
What I’d say my biggest question that comes from this is how should this inform lineup card construction. The Matthews line had all three forwards very close to the 200 ft category, and that’s probably something the Leafs want to emulate this season. The high octane offense of Tavares and Nylander might need some cold water rather than more gas though, and that could be a case for Jarnkrok over the likely more gas option of someone like Nick Robertson. The Leafs also probably want to find a way of steering Nick Robertson away from having too many low event linemates, as he deserves a chance to produce.
The breakdown of where the players excel is heavily skewed by how they were deployed, and I’m sure if someone wants to Ph.D. of math this rather than High School math this, they will be able to better account for Muzzin and Holl being identified as stronger offensive options than defensive. The reality of last season though was with the heavy defensive workload they took on last season, they didn’t perform as well as in previous years, while still being adequate in the offensive zone.
Advertisement
Ad
It may also be interesting to see what this translates into when it comes to third period deployment. Do the Leafs play less of Tavares and Nylander when protecting a narrow lead and do we see less of Kampf, Engvall, and Aston-Reese when they are trying to catch up? We’ve certainly experienced the latter at times, but Keefe has often continued to give Tavares and Nylander their regular shift when protecting a lead.
While nothing here is particularly groundbreaking, it’s an easy way of binning players and when looking at the league wide numbers it was a great reminder of what a special player Patrice Bergeron is. Of the entire league last year the players that were above the 90th percentile in both the offensive and defensive categories were limited to just five players: Patrice Bergeron, Brad Marchment, Charlie McAvoy, Matthew Tkachuk, and (wait for it…) Mason Marchment. I’d like to believe that was a Barkov contact high.
Advertisement
Ad
Depressing Mason Marchment mentions aside, the Leafs are as we expected them to be, lopsided towards offense, but with the new additions from this offseason there is a bit more of a balance being achieved.