Four minutes per game! That seems like a lot. There’s a reason for that. The Colorado Avalanche had the most power play time in the NHL last year at 483 minutes. That large total minutes available affects how many minutes per game a player has. The Avalanche also played one unit almost exclusively, which the Leafs did not last year, and so far haven’t this year.

No one plays more power play minutes than Alex Ovechkin, and no one should really need that explained. Second to his 358 minutes last year was Avalanche centre, Nate Mackinnon, then Capitals defender John Carlson and then Tyson Barrie with 315. Sam Girard, the second defender by minutes on the power play for the Avalanche had only 165. Girard is 21 years old. The Avalanche were playing a rookie as their second unit power play defender.

The Maple Leafs, last year, used Morgan Rielly on the first unit and Jake Gardiner on the second. They did not play anyone under 21 on defence. Now, I don’t think points is how you should do the math here, so I want to look at Individual Expected Goals (ixG, the probability that individual shots will be goals) instead.

Last year the player with the most ixG on the power play in the NHL was John Tavares with 14.24. He played 145 fewer minutes than Ovechkin, who had 13. It’s obvious not all power plays or players are created equally. You won’t find any defenders in the top 50 ixG, though. To more realistically judge their contribution, it’s worth getting into more complex numbers because they need to be measured by their isolated impact on the guys like Tavares or MacKinnon (fourth last year with a mere 11 ixG).

Evolving Hockey’s RAPM is a way of measuring the individual performance of a given player in an environment where many factors can impact their raw results. I took a look at the power play Expected Goals per 60 RAPM, and you still get Tavares on top with 1.459. The best defender last year by this measure is not Barrie or Rielly, it’s Justin Faulk at 10th, followed by Mark Giordano, who both had .96. Rielly is 15th with .913. Barrie, by this measure is 58th. Not overall, just within defenders. He’s worse than most team’s second unit guy.

There’s two things going on here. The Avs power play wasn’t all that good, they just had a lot of them, and Barrie is more of an individual shooter and a “see the ice” guy than he is a power play guy. He’s not terrible at it clearly, but he wasn’t earning his money that way, even if his reputation says otherwise, and the sheer number of Avalanche opportunities got him 25 points (only 2 goals) on the power play last year.

But maybe last year was just a fluke. His career has been up and down more so than most players’ usually are. What about a long multi-year look?

Three seasons from 2016 through the end of last year gives me (and I had no idea, I just thought it was the right thing to do to check) Morgan Rielly as the best power play defender by this measure. Barrie is 64th for defencemen, worse than Girard, so the Avs weren’t out to lunch using their rookie.

The best player overall over that period is the man the Avs got for Barrie, Nazem Kadri. Rielly is 15th overall. One thing, I will say is that isolating out individual impact can only go so far, and the Leafs clearly have had a very good power play for years, with very good players on it.

It’s frankly a little absurd to even ask if Barrie should take some of Rielly’s minutes on the power play given their historical performance, once you do the actual math. It’s way too soon to judge if Barrie on the Leafs is a better threat on the power play than he was on the Avs, but he’s got a really tough job ahead of himself to prove it. Even with a full roster, the Leafs can’t cobble together much of a second unit.

At the moment the Leafs have a 60/40 split in available power play time between Rielly and Barrie. Barrie played about 75% on the Avs last year.

Whatever led to the desire to find yet another roster choice controversy, this one seems really wrong-headed. Mo is the man. There can’t be a credible argument against that. If he wasn’t already playing too much, I’d be calling for him to play full two-minute power plays like John Carlson does.

If this is the reason for this idea getting floated:

And it likely is, that’s a terrible way to judge a power play. I have criticisms of the Leafs power play, but success percentage calculated over 13 tries at anything shouldn’t be taken seriously. Go make 13 tweets, and see how many are good. Never use success percentage anyway. At the barest minimum, you have to look at GF/60, and even then not over 13 power plays. Employing recency bias on purpose isn’t the way to go unless you’re trying to spin something.

This idea that the Leafs should start making changes for the sake of changes, that’s just going to lead to more unfulfilled hopes from fans who don’t want the coach doing any math, good or bad, they just want him to flap in the wind trying something new every day that they know in advance they’ll hate. The process General Manager isn’t going to do that, no matter how many clever ideas are trotted out. Maybe Babcock should just change his suit instead, to satisfy the need for changes for the sake of them.

I think Tyson Barrie is underused on the Leafs. I think he should play more five-on-five in ways he can have offensive impact, and I expect his power play stats this season to actually look better than in the past. We’ll find out in time if that’s true, even with the challenges he faces with a unit that’s him and William Nylander and three fans out of the stands. But please, no more manufactured roster controversies.

If you want a constructive idea for fixing that second unit, I like this one:

Analysis