The Leafs have a lot of speed, a lot of offensive skill and some real troubles in their own end. We can see that easily enough even from a few games. But 21 games of data on shot rates is enough to start predicting the future with some confidence, and it is enough to get a much clearer picture of who this Leafs team is than we got after only 13 games.
All data used is score and venue adjusted five-on-five, from Natural Stat Trick. Shots always means all shots, or Corsi, unless otherwise stated.
Luck, Randomness and Goaltending
If you cast your mind back to last year, you likely remember no goals and bad goalies. The truth is, while the scoring woes persisted almost evenly through the year, the goaltending started out fine and got really bad later. The full year result was something on the low end of good for save percentage, and bad for shooting percentage. This year, the results are a different story so far.
Comparing to last year, and the year before, also known as the time of great upheaval, the Leafs team shooting percentage is just totally normal. There might be individuals still who have unsustainably high numbers, but overall the team isn’t suffering from no luck at the net. There are a couple of forwards who likely won’t maintain their low numbers either. But there is no schedule on that sort of regression. There is no due or overdue here. It happens when it happens, and we have no idea what is a normal shooting percentage for a very large number of players on the team because they’re rookies.
Now for that save percentage. This year’s terrible number so far is actually better than the entire second half of last year. So if the next three quarters of this season are better, and there doesn’t seem to be a lot of reason to think the October troubles will continue, then by the end of the year, well, to be honest, we’ll be where we’ll be. This is not easy to predict given the young team and those October troubles.
Shooting and Goals
In the bigger picture, we can look at the Corsi For percentage, the Goals For percentage, and the component parts of the CF%: the Scoring Chances and the High-Danger Scoring chances. This data uses the same formula to calculate scoring chances as I used last year, so I stuck with it.
That improved GF% stands out. Also notable is that the CF% is low, just at 50% in fact, and yet the scoring chances and high-danger chances are very heavily weighted in favour of the Leafs. This is something that should be looked at with a great deal of skepticism. By slicing the total shots up into smaller piles and grouping them by type, the information is less reliable. It’s not more detail, it’s an impression of the way the team plays and how well they do at maximizing opportunity and limiting opponents’ chances.
About that GF%. If the save percentage has been terrible, the shooting percentage average, and the GF% is almost at 50%, perhaps we haven’t really seen yet what this team can do. More on that when I look at their Expected Goals For percentage.
Separating the Shots into Component Parts
This is all shots, scoring chances and high-danger chances, split into for and against. This is where we see where that just barely 50 percent shot differential comes from.
I don’t think anyone is surprised that the shots against are way, way too high. But the overall pace of play, which is the total shot rate for and against are very large. The Leafs have never played at this pace before. Almost no one has.
Is this, in fact, a problem? Can you have too much pace and end up with high shots against? Maybe, but the eye-test says the shots against has a different origin. A poor zone exit and a tendency to indulge in long periods of defensive passivity in the face of opponent’s pressure seem to be the issue.
Here we see that the more favourable split in scoring chances and high-danger chances is coming from higher rates both for and against than have been seen in the past, but the proportion is better.
This seems to imply that while the Leafs allow a lot more shots against than they have recently, even in the last year of Randy Carlyle’s tenure, they are defending more effectively in their own zone. And yet, at times, it doesn’t look like it.
It might be the “at times” here that is confusing, but the numbers and the eye-test are at odds. There are occasionally serious lapses in defensive positioning and some basic execution of simple defensive plays that is very poor, and it may be that, due to the wide-spread rookieitis on the team combined with some defensively poor veterans, those lapses are really bad. It’s not uncommon for people to equate good defence with low shots against, but what you actually do while in the defensive zone can’t be terrible, or even low shots against will beat you.
Those shots against are too high though. Just because you keep them proportionally to more low-danger chances than most teams do, doesn’t mean you won’t give up low-danger goals. It’s a highwire act that can come crashing down very easily.
Arvind explained expected goals very well, but in brief, it is a much more sophisticated weighting of shots for and against than the crude splitting into scoring chances and high-danger shots I showed above. This is a few less games than the other charts.
As a companion to the piece I just bumped:— Sean Tierney (@ChartingHockey) November 23, 2016
The Toronto Maple Leafs.
Break-even possession. SCF% is strong. GF% is a little low. PDO's even. pic.twitter.com/wnTGLEMkps
You want to look at the fourth column, which is the expected goals for percentage and is well above 50 percent, and then look at the actual to its left which is much lower.
Some of this difference is in that save percentage number we looked at first, but some of it is just luck and randomness on the scoring and the goals against. In terms of underlying systems, what the Leafs are doing is what leads to success. The reality is a little different for a lot of reasons, some of them simply chance.
With reasonably average and stable goaltending results and no other improvements, this team is going to improve its GF%, which is what you win games with.
I feel like I’ve said that before, and I have. In the first half of last season, where the Leafs got points at almost exactly the rate they are this year, they were sitting in a spot at the bottom of the large number of bubble teams in the Eastern Conference or at the top of the bad teams league-wide, however you want to look at it. They would have been drafting at 10-15 overall, and they were a lucky splurge of winning away from a wildcard spot.
And we know what happened next.
This year is even more complex. The Leafs right now are in that same bubble team spot in the standings, but the expected path for rookies is up, up, up. This is not a team of veterans expecting to be traded, who are learning how to play decent hockey, some of them for the first time in their NHL careers. This is a team full of players who haven’t even started to climb the aging curve. And then there is the goaltending to try to predict.
I don’t know where they’re going! I don’t know if they will tighten up the shots against or get worse. I don’t know if they will keep up that offensive pressure or slow down. I don’t know if this team is going to crash and burn or end up sniffing a playoff spot.
I’m guessing no on the playoffs. I just don’t think I can buy a team with four second pairing defenders as a playoff team. But it has been Nikita Zaitsev and Connor Carrick who have most improved their defensive execution as the goals against have dropped—not a coincidence—so maybe they have much higher to go yet.
Rookies! Who knows what they’ll do next. We’ll look at all of this again at the halfway point of the season.