Who are the elite teams in the NHL, today?

Pittsburgh, obviously enough.  Back-to-back championships and two top-five players will do that.  Washington, probably; even if you think their playoff failures point to something deeply wrong with them, they’ve won the President’s Trophy the last two years.

After that?

The next tier of the league is a lot less clear.  Chicago’s depth is made out of paper mache and they just lost one of their two best defencemen.  Anaheim’s best forwards are in their 30s and they’re headed by the most infamous Leafs’ coach of this century.  (He hasn’t changed; they’re still worse than they should be in shot differential.)

Montreal has good shot differentials, and Minnesota has bad ones they outperform with PDO; one’s counting on a very raw 1C (Jonathan Drouin) and the other on a very good 1C who is nonetheless soon to turn 33 (Eric Staal.)  And they’re both leaning heavily on aging defencemen who used to play for Nashville.

Right, the Preds—who might have the best defence core in the NHL, and who also finished 16th.  Tampa, who got killed by injuries and missed the playoffs entirely.  San Jose, who have an average age of about 93.

And Toronto.

Who have the field wide open in front of them.

The Muddle

The NHL has said for years that a key purpose of the salary cap is to engender parity, and while that soft-pedals the more obvious motive of saving the owners a ton of money, it certainly seems to have worked.  The bulk of the league looks more closely bunched together than ever.

In a recent article, TSN’s Travis Yost looked at whether Corsi has lost its predictive power.  He concluded that shot differentials still clearly matter, but he also found that the distribution of shot differentials was tighter than at any time in the last five years.

You can see this for yourself, if you’re so inclined.  To take 5v5 adjusted CF%, as per Natural Stat Trick: half the league (the teams ranked from 6th to 20th) was within 2% of each other, ranging from 49.5% to 51.5%.  Further to that, the two teams zooming off the top end of the chart are two who are widely agreed to have been pumping bad shots on goal, and they weren’t especially successful doing that.  The Boston Bruins barely made the playoffs and went out in Round One.  The L.A. Kings missed entirely.  Shot differentials matter, but no one’s running away based on them.

Dom Luszczyszyn has commented on how odd his projections have been: 21 teams are projected with a higher than 50% chance of making the playoffs, since after the clearly awful teams things get very tight, with a bunch of middling ones.  A look at the implied probabilities based on betting odds tells you the Pens are good and the Oilers are getting better, and then, well, who knows?

You get the picture.  It’s a Wild West out here (and a nearly-as-wild East.)

The Leafs and the Atlantic

Zooming in a bit: is there a team in the Atlantic you’re sure, in your heart, is really better than the Leafs?  I think Tampa is, if they stay healthy, but last year they didn’t and they consequently weren’t.  Montreal won the prize last year, then lost their second-best scorer and got significantly worse at left defence.  If the Drouin-to-centre experiment goes like the Galchenyuk-to-centre experiment went, well...they’re going to need vintage Carey Price to repeat as division champs.

Hold that thought.

Toronto has, at least offensively, a top-three forward group in the NHL.  They were second in the league in adjusted CF60, second in 5v5 scoring chances, fifth in goals, and second on the power play.  They lost no one in their top nine and added Patrick Marleau, who is still an effective player whatever you think of his contract.  Even with that, three of their top five scorers were under 21, and none is older than 28.  They didn’t ride an especially elevated shooting percentage, certainly not one that’s doomed to regress.

They have a defence group that, while not dazzling, is awfully good at moving the puck.  If Gardiner can stabilize the top pair, and Hainsey help the second?  If their injury luck holds?  If their shootout luck starts to approach normalcy? Well, this is getting pretty rose-tinted...but do you want to bet against them being the best Atlantic team?  If they do it, do you want to bet against them winning two rounds in that division—as many as any Leafs’ team has won since 1967?  And after that, well...who knows?

This is very much the scenario of hitting all the green lights on your way to work, where your boss spontaneously gives you a raise.  But in the NHL of 2017 there’s no definitive red light that goes in front of a team like this, no overawing, titanic class of opponent that’s clearly so far above we can’t see them.  Pittsburgh is better than we are; they’re also injury-prone as hell, had worse shot differentials than Toronto, had a playoff defence that was somehow notably worse than ours, and were a goal away from losing to the Ottawa Senators.  It is very easy to talk yourself into Toronto competing for or even winning the East, and I would argue it’s more realistic than it’s been any year since 2004.

Of course, there’s another scenario, and it’s the one Sabres fans are dreaming of.  If Frederik Andersen gets seriously hurt, suddenly this team is going to be in very tough just to make the playoffs.  If that spectacular rookie class was overachieving—and boy, when they did this well, you have to at least consider the possibilty—regression is going to bite like a rottweiler.  This team was almost as bad defensively as it was good offensively; we’re hoping that improves with experience, but what if it doesn’t?  I honestly believe you can look up to the top of the league without seeing anyone totally untouchable for the Leafs.  You can also look down and see a hell of a lot of teams nipping at their heels.  If the Sabres finally calm down that defence and get a healthy year of Jack Eichel, if Boston rebounds with that superstar top line, if the Sens don’t regress as hard as I expect—and I never picked them to make the Conference Finals, so what the hell do I know?

What I’m trying to get at with all this is that the error bars around a prediction should be huge this year.  Predicting is a mug’s game at the best of times, and even whiz kids like Dom the Lush Chicken have to accept misses go with the territory.  This isn’t me trying to cover for being wrong (I’m wrong all the time anyway, who cares?)  It’s that I genuinely think all this parity and the unique position of the Leafs mean they have almost as wide a range of possibilities in front of them as any team in the NHL.

So Where Do They Finish?

Last year, I made a very detailed historical study of what last place teams did the following season, and concluded the Leafs would finish with 87 points.  The Leafs did not care for my prediction because they had dope rookies who decided to wreck shit, and they finished with a cool 95.

Still, looking at that gave me an idea of the conventional rebuild path for the successful teams, which is something like:

  • Last year bottoming out
  • Transition year moving up the standings
  • Contention or near-contention
  • Cup/

This is a very imperfect model, based mostly on Chicago and LA (Pittsburgh is weird with their double #1 picks), and the Leafs defied history with their unusually successful transition year.  But I think both from history and from normal aging curves, the Leafs can expect a reasonable amount of progress.  Plenty of teams don’t take the next step, it’s true, and this could all go straight to hell. The Leafs’ offence looked real, and damn, it looked good.

I tend to believe Mike Babcock can get the most out of his teams, and he’s a consistently excellent possession coach.  It would not take a huge defensive improvement for the Leafs to make a considerable jump, given their scoring.

What about regression to the mean?  What weird numbers should come back towards average?  Well, the Leafs had career seasons from Nazem Kadri and Tyler Bozak, and we don’t have a baseline to judge whether Auston Matthews can sustain that shooting percentage—though we have shooting data that says he may well do it.  I also don’t know if that power play will be this good again.  But there are other examples running in the opposite direction—Nylander and Hyman are two—and no one really was so far above his norm that you can’t trust it.  Most of all, though, this team got killed in one-goal games.

No, seriously, they were 30th in the NHL in 1GG, with a winning percentage of .378.  That’s somehow even worse than the Colorado Avalanche, whom I would have expected to be the masters of every kind of losing.  Obviously that ties into the Leafs’ absurdly bad shootout luck and their penchant for blowing leads, but one-goal games tend to average out over time; there’s a lot of luck involved in them.  I think you can make a persuasive case that this team was at least a little better than its 95-point finish, even last year.

The last kind of regression, though, is injury luck.  The Leafs had absurdly good health last year; no key forward missed more than five games and most of them played all 82.  Connor Carrick was the only significant defence injury before the very end.  The website Man Games Lost estimates that the Leafs were the fourth-healthiest team in the NHL last year, accounting for the quality of the players who missed time.  Some people attribute this health to the Leafs’ youth; all I can say is that young men get injured playing hockey too, and luck is a crazy thing.  I have to expect the Leafs are going to get dinged up a little worse this season.

Put all that together?  I think the Leafs were a top ten team on talent by the end of last season, I think their luck ought to even out a little, and I think their talent will tell.  I’m picking them to finish with 100 points on the button next year, second in the division behind a healthy Tampa Bay Lighting.

But keep an eye on those error bars.

How many points will the Leafs get this year?

105 or more189
84 or less14