The Leafs have locked down third place, have two essentially meaningless games to play, and Mike Babcock is putting out what looks like his playoff lineup. With no certainty at all who the Leafs will play in the first round of the playoffs, there’s only one thing left to do: Put the fourth line under a microscope.
The first time I saw someone today state that Andreas Johnsson should be in the lineup instead of Leo Komarov, I got irritated. I picked a fight with the PPP player analysis department, and Arvind, who does not let you get away with merely kvetching, said, “Well what would you do?” I answered him, and then I had to think about how much I liked my answer and about why I was so irritated.
Travis Yost came out with his take on all this which is reasonable here:
It’s possible that playing with a veteran like Plekanec or a lightning-quick skater like Kapanen is helping Johnsson’s transition. It’s possible that he’s just eating up soft minutes against inferior depth lines around the league. And it’s true that Johnsson’s limited action has come against some weaker competition in general – his most common opponents by ice time are stacked against guys from Buffalo and Montreal because of the way the schedule fell.
But he sets that in opposition to a WOWY chart, which you can see at the link above, of the Leafs with Johnsson and the Leafs without Johnsson, and boy howdy, does it look impressive. And now I know why this discussion is irritating to me.
First: Andreas Johnsson has played 83:37 of five-on-five hockey with the Maple Leafs, spread over nine games. This is too little time to tell us anything. You cannot judge a player by nine games, and I do not care how All-Star his AHL play was before. This is an eyeblink in an NHL season, and the fact that Johnsson has such good results is a hint that he’s a real player and makes me very excited for the future.
If your honest-to-goodness first response to his play in nine games is to expect that Leo Komarov should sit out so he can play, I ... well, I get irritated. Yost doesn’t say that, by the way. He ends on a vague note saying that Johnsson has better results than some players up the lineup and should be given consideration.
Name names! Show your work!
I’m kidding. I know he has a limit on how long his stories can go, and superficially, I think he’s right on that conclusion. I have no complaint with his overall point that Johnsson, with the above caveats, is looking pretty damn good.
However: There is nothing that gets me more annoyed (there totally is) than that kind of WOWY where a role player is compared to the whole rest of the team. Usually this sort of quick and dirty evaluation method comes out when you want to make a fourth liner look bad. Because nothing helps the look of a depth player’s stats like comparing them to Auston Matthews and James van Riemsdyk.
In this case, Johnsson’s tiny sample looks super compared to the Maple Leafs as a whole. But the underlying point of that WOWY isn’t really Johnsson’s results, it’s the unstated and unnamed others who are bad, and who he should kick out of the lineup with his paragraph-long resume of NHL success. It’s the stats that aren’t shown that sells the idea that Johnsson is better than the alternative, and most people’s minds will immediately jump to Leo Komarov as the unnamed alternative.
Let’s jump there.
Leo Komarov versus Andreas Johnsson: We need to look at their results side-by-side.
But, wait. Isn’t that just the same as comparing a fourth liner to Auston Matthews? Hasn’t Komarov played most of the year against top lines on his off wing with Nazem Kadri? Hasn’t he been tasked with something dramatically different to what has been asked of Johnsson?
The answer, as you all know, is yes, he has. And there are not very many players with a quality of competition measure as skewed as Leo Komarov’s. Patrice Bergeron plays less time against top lines, so does Aleksander Barkov or Sean Couturier. I don’t think you can, or should, ignore that and just to a straight-up one-to-one compaison.
As you also know, Komarov has played on the fourth line lately. So I’m going to do a thing I don’t like and deliberately throw out information about Komarov’s season and cut his sample size down dramatically. But I think it’s warranted in this case.
Komarov vs Johnsson
Player 1 | Player 2 | TOI | CF | CA | CF% | HDCF | HDCA |
---|---|---|---|---|---|---|---|
Andreas Johnsson | Tomas Plekanec | 47:06 | 48 | 43 | 52.75% | 8 | 5 |
Andreas Johnsson | Dominic Moore | 4:15 | 0 | 0 | 0.00% | 0 | 0 |
Leo Komarov | Tomas Plekanec | 26:00 | 15 | 24 | 38.46% | 3 | 7 |
Leo komarov | Dominic Moore | 141:52 | 143 | 131 | 52.19% | 21 | 19 |
What we get here is a view of Leo Komarov that’s a little different from his total numbers of 45 per cent Corsi and 48 per cent High-Danger Scoring Chances. If you combine Komarov’s fourth-line time with two different centres, you get a Corsi percentage of just over 50.
You also get to see just how small the numbers are with regards to Johnsson. One or two shots can skew those percentages wildly. That’s why small samples are unreliable. Random streaks of luck can make them lie easier than a big pile of data.
I hope you’ve also noticed that Johnsson’s fourth-line time is nowhere near all of his minutes played. So where did he spend the rest of his time? The very first thing I said when I read Yost’s caveats was that the most important one is that a decent chunk of Johnsson’s time was spent with this man:
Now, back to Arvind making me say what I’d do with the lineup.
I said that I think Komarov is a valuable player, and by that I mean his ability to at least hold to something like even in high-danger chances against the toughest competition, but I don’t want him on the power play. This is where Johnsson has immense value. Kasperi Kapanen has only played PK in the NHL — brilliantly, I’d add. And the attraction to playing the two of them instead of the seemingly poorly performing Komarov is understandable.
But if I’m putting Komarov in, then I want Johnsson in there on the power play. The choice isn’t Komarov or Johnsson, and we’re letting a left wing designation make us see them as equivalent players for the same roster spot when they aren’t. The choice might be Johnsson or Kapanen. But that’s not how I’d necessarily leave it.
I’d give a very long and hard look to another player who isn’t adding personal offence, who doesn’t add to the power play, but who has value as a defensive player like Komarov does. I’d look at Connor Brown and ask if his pros outweigh his cons and if his minutes are justified. I assume Yost had him in mind too in his article conclusion.
Why is Connor Brown on the Bozak line? He hasn’t been terrible there, but his personal scoring isn’t going to add much. He needs to function as a very good zone exit guy to add value, and if that’s the purpose of him there, that might be a good roster decision even if, one-to-one, someone else looks better by overall shot metrics.
I would consider Johnsson on the Bozak line, though, and Brown on the fourth line. I’d consider this even more if Johnsson had been on the team for months, not nine games. But here’s the only fair way to compare Johnsson and Brown:
Johnsson vs Brown
Player 1 | Player 2 | TOI | CF | CA | CF% | HDCF | HDCA |
---|---|---|---|---|---|---|---|
Andreas Johnsson | William Nylander | 26:39 | 36 | 24 | 60.00% | 10 | 4 |
Connor Brown | William Nylander | 36:39 | 35 | 34 | 50.72% | 6 | 8 |
The teams the Leafs were playing when Johnsson was on Nylander’s wing for a very tiny bit of time were either bad or playing end-of-season hockey, and you shouldn’t let that thought exit your mind for even a second. If you compare Johnsson to any regular roster player with a background assumption of all else being equal, you’re wrong. All else isn’t equal.
But that looks very interesting to me, nonetheless.
Could you make the Leafs better by putting one of Kapanen or Johnsson up the lineup in place of Brown and having a fourth line of Komarov - Plekanec - Brown? I think so. I think that would be particularly valuable to do for games where you want the fourth line to take some tougher matchups. This is just fanciful thinking however, or at best, premature. This might be next training camp’s big question.
The decision between Komarov, Kapanen and Johnsson, which many people who buy into the mythos of Mike Babcock think is already made, might come down to who the Leafs are playing. Which type of depth they are facing, and does the opposing coach line match or not. Do they want to beef up the power play in the playoffs or do they want Kapanen and Komarov and their PK skills ready to prevent goals — the Leafs might not actually need more scoring, no matter how much Johnsson’s offensive numbers make you long to see him score a big goal in the playoffs.
I keep thinking about that seemingly weird decision to put Johnsson on the right wing against Buffalo on Monday night. Maybe that was just to make room for Matt Martin. Maybe Kapanen is banged up and a game off was warranted. Or maybe it was a chance to see Johnsson on the right side.
One other thing to note: Johnsson’s other big stats booster is a few shifts with Auston Matthews where they were 5-1 in Corsi. One injured centre in the playoffs means a winger spot opens up in the top nine. Maybe on Matthews’ right hand.
All this flexibility is a nice problem to have going into the playoffs. That we know for sure.
All numbers are from Natural Stat Trick and are five-on-five unadjusted.
Comment Markdown
Inline Styles
Bold: **Text**
Italics: *Text*
Both: ***Text***
Strikethrough: ~~Text~~
Code: `Text` used as sarcasm font at PPP
Spoiler: !!Text!!