It’s just one game.

Arvind has already talked about how this one game made him a little uneasy, but not quite worried about the Leafs:

Is it too early to worry?

I’m not a worrier. But the Leafs were absolutely bad in this one game the way they usually are, while being good, for the most part, in their normal ways.

Frederik Andersen was excellent, and by several measures I looked at he saved at least two goals over what an average goalie would have done in that game. Some of the offensive forwards we wanted to cheer for seemed to not be on their game, and the third period, in particular, was dismal. And yet the Leafs won the game.

You can’t look at one game and extrapolate that into a season. But you can look at the game in detail and see if what you thought you saw was real. In my recap, I said I hated the Nazem Kadri line, but then right after the game people started telling me Corsi numbers, and I shrugged and said, stubbornly, that I still hated them. The Score and Venue Adjusted five-on-five Corsi For percentage champs in that game were: Kapanen, Kadri, Leivo, Dermott and Brown. That’s the Kadri line and one interesting extra forward at the top of the pile.

I don’t look at Corsi numbers from individual games all that much. I like the team heatmaps to show the strength of the offence, the pace, the location of shots, etc. I don’t really like single game expected goals data, as I find the cumulative shards of slim chance shots piling up unconvincing. Sometimes it’s fun to look at and see who has the highest quality of scoring chance that night, however.

See the problem? Josh Bailey took it for the Islanders in OT, but on the basis of the probabilities, Carolina should have run away with that game long before then since they had the overwhelming advantage in total shots.

What any sort of data we can look at after a game does is tell us what actually happened in the game, minus our feelings about the players and our hopes and dreams and parade plans. The value of something like tracked passing data is that it’s accurate, that it shows you how often each player made the decision to pass the puck in a way that led to a shot, how often they chose to shoot, and how often, by inference, they weren’t really doing much at all.

No set of single-game statistics will really tell you what that player will do the next time around, and yet the temptation to look at the numbers from one or a whole bunch of games and say this is the measure of the man, not just the measure of what he did that one time is immense. Let’s try to curb that temptation.

What a player does on the ice is influenced first by what he can do, what he believes he can do, who he plays with, what the coach expects, who he is playing against, and so many other things, chief among them luck.

So, with all of that in mind, who did what for the Leafs in that one game?

From Corey Sznajder (see below for more information on his work):

This is a chart of choices made with the puck. The passes here are those that immediately precede a shot. Bear in mind, that to make the choice to pass or shoot, you have to have it in the first place, and this chart might lead you to correctly think that Auston Matthews didn’t have it much. His Corsi data backs that up.

But I want to test my irritation at Kadri that was belied by the Corsi data. And there he is at the head of the class, doing 11 things to the puck that mattered. Three are his own shots, and the rest are passes that led to shots.

Whose shots, though? Josh Leivo has one shot and three passes so now we’re up to seven passes that went somewhere. Connor Brown shot the puck once, and did not make one single pass that led to offence. So who did the shooting?

Now it’s possible that all of Kadri’s success came in his weird, short shifts with other forwards, so I looked at his CF by teammates, or all shots taken by anyone while Kadri was on the ice with that one player. For Brown it’s 13, Leivo: 11, Kapanen: 5, Johnsson: 2, Marner: 2.

So, really if he was passing to a strange forward, it wasn’t very often. I fear he was passing to defenders. He played most with Jake Gardiner and Nikita Zaitsev, and a lot with Morgan Rielly and Ron Hainsey. Three of that four handled the puck a lot.

What I think we are seeing is a line that controlled play a lot, spent a lot of time in the offensive zone, and made plays with a very low percentage chance of succeeding, largely based on who passed the puck to whom, and who then shot it.

Corsi, Corsi, everywhere, and nary a shot that mattered, bar Kadri’s own.

It’s not enough to get zone time and pepper the opposing team with shots. The Hurricanes did that against the Islanders and lost, and they lost largely because they only managed a few more really good quality shots than the Islanders did, while tossing up trash from bad locations. Given enough time, volume will get you goals, but volume plus quality of shot gets you glory.

Back to the Leafs. Connor Brown did less useful stuff with the puck than the 4C, or Tyler Ennis, who is pretending to be a top-six winger. He did less than Zach Hyman, who you don’t want to do anything but pass. He did less than Andreas Johnsson, who played a lot less time on ice, and he did a hell of a lot less than Kasperi Kapanen, who looks like the top-six guy in this crowd.

All Leivo did was pass, although, by the eye-test, he played a feisty and smart game most of the time. So given all of that, give Kadri someone to pass to! Give Kadri at least one real, honest to goodness quality shooter. It’s fine if the focus of the line is drawing the other team’s toughest customers to a standstill, but don’t handcuff Kadri with no one but Nikita Zaitsev and Ron Hainsey to shoot for him.

While we’re here, let’s look at the zone entry chart:

And you all thought Mitch Marner didn’t do anything! But note that it’s Kadri himself who carried the puck in, while Brown just dumped it in the corner and got on with the forechecking. Kadri’s line did that a lot. They did it really well, and the end result was some tough and tenacious board battles that let to ... shots by defencemen.

That’s better than shots against, for sure, but it’s not offence that matters. And at the risk of giving in to temptation, I don’t think Kadri can produce offence that matters when he’s three-to-four times the man his wingers are night after night.

Give him one good winger. Just one. Please.


Some of the basic Corsi information came from Natural Stat Trick, the expected goals information is all from Moneypuck as indicated on the images.