Last week, on the radio, Gord Miller said a thing.
"You can't sit there with a spreadsheet, and think you're smarter than Marc Bergevin. You can't." Gord Miller on #TSN690 just now.— Carey's Glove (@chile_pepper) November 2, 2016
This sparked the standard Twitter banter. Plenty of nerds and bloggers (including me) snarked about it, while the crowd who wants to take those nerds and bloggers down a peg luxuriated in a +1 from the MSM. This whole discussion has been had a thousand times, and while it’s fun to get a one-liner in for your side, there’s nothing novel about it.
For Leafs bloggers who lived through the Carlyle/Nonis era, all of this seems kind of quaint. We no longer have much doubt as to whether it’s possible we know something more than bad NHL coaches and GMs, because we watched as a group of them ran our franchise straight into the ground. Here is a long string of PPP writers and commenters being right about the Carlnonis group screwing up. If we want to trouble ourselves establishing how a bunch of people with spreadsheets can be smarter than NHL front offices and coaches, there you have it.
The Gord Miller song and dance, though, got me thinking.
Randy Carlyle is possibly the most incompetent NHL coach currently employed, and Dave Nonis’ tenure running the Leafs’ FO is one of the most stupendously bad of any GM this decade. The work done by the first generation on this blog and elsewhere, in the face of even dumber and more dated resistance than there is now, was brilliant. (I can say that without immodesty, because none of it was mine.) But they were also writing in opposition to a miserably inept regime.
This Leafs group is different. We have a GM who’s in the Hockey Hall of Fame and a coach who probably will be. We generally think they’re right.
So what do we do when they look like they’re dead wrong?
Am I Smarter Than Mike Babcock?
Of fucking course not.
Mike Babcock is both reputed to be, and actually, one of the best coaches in the league. I wrote a bunch of reasons why here, but the short version is that he’s coached teams to basically every championship in men’s hockey. Our knowledge of underlying numbers suggests he makes his teams better. He’s revered. He was the subject of the priciest bidding war for a coach in NHL history. He’s forgotten more about hockey than I’ll ever know.
I love having Mike Babcock running our team. He did impressive work with a skeleton crew last year and I think he’s going to lead this team to a great improvement this season. There’s even an outside shot at a playoff spot.
Just a couple of things, though, Mike—
Hunwick, Smith, Hyman
The thing about Matt Hunwick is that he is very bad at hockey now.
To be clear, I don’t mean he was always bad at hockey (he was not), and I don’t mean that he is bad at hockey for any league other than the NHL (he isn’t.) I also don’t mean that he’s not a great guy (by all accounts he is.) He just can’t do his job.
Both last season and in the early going this year, he was by far the worst regular Leafs defenceman in most metrics and was bad by pretty much all of them. Unlike some other unloved Leaf players, where you can find something to make a devil’s advocate case, Matt Hunwick was unplayable.
And yet Babcock continued to play him.
Hunwick is out with a nominal injury (“to his hockey”, was the Twitter joke everyone made.) Hopefully Babs has observed the Leafs’ defence is functioning slightly better without him than it did with him. But you can see a similar problem, ongoing, with Ben Smith.
Ben Smith is a very good AHL centreman. He was waived by the Colorado Avalanche this year (who are probably a worse team than the Leafs) and promptly took the 4C job from Peter Holland. The fourth line, rather mysteriously, has been submarined in EV shot differentials since the change, though it’s of course still very early.
Smith, though, has a little more going for him—according to Babs, he was explicitly brought in to be a PK centre. He does win faceoffs. So far as I can tell, Smith was a meh penalty killer from 2013-15 with Chicago—the only two NHL seasons where he played more than 20 games—but perhaps he’s right. I doubt it—I really think we should put in Holland as of immediately—but there might be something there. Hold that thought.
Zach Hyman is in a happier category than the first two players, because he very much looks like an NHL player. He doesn’t produce all that much offensively, but his possession numbers are superb in the early going, and he has the kind of grit and grind game you appreciate a little more every time he edges out an icing. I like Zach Hyman a lot. Do I think he should be playing 1LW to Matthews and Nylander? Well...granted his linemates do a fair bit of their scoring on the powerplay, the fact is, Zach has one assist in eleven games with two of the best young offensive players in the NHL. If that doesn’t improve very soon, there can’t really be a case for keeping him in the job. There just can’t.
So there are three Mike Babcock decisions: one I’m dead certain is a mistake, one I strongly doubt, and one I’m unsure of.
Mind The Gap
Insofar as I know anything much about hockey, it comes from a few places. I watch it religiously, I played it ineptly, I read about it all the damn time. I have War-on-Ice bookmarked, and Corsica bookmarked from when War-on-Ice shut down this year, and Natural Stat Trick bookmarked for the 30% of the time Corsica won’t load. I have a copy of Hockey Plays and Strategies that I thumb through each time the Leafs get hemmed like hand-me-down pants. I try to follow smart hockey Twitter people and then make sophomoric comments on their tweets. I even squint at Micah McCurdy’s hyper-complex visualizations and figure out what they’re saying...uh...well, every now and then.
One thing I know better than anyone is how much that leaves me not knowing. There’s a gap in my understanding a mile wide. Some of those things come from ignorance and inexperience, some from distance. An elite professional on Babcock’s level has decades of expertise, from big picture planning to little things a fan would never notice. He’s also inside the team, working the practices and hearing the dressing room. The gaps in my knowledge are where Babcock rightly earns his millions, and to some extent no one is ever going to know what’s in there.
So far, this is about what Miller or any of the millions of traditionalists would say about us: there are things you don’t and can’t know, so X-out of Excel and shut your laptop.
The lesson of Carlyle (or Tortorella, or Mike Keenan, or whoever), though, is that sometimes the gap is a landfill. Distance can give you perspective, and being on the outside can free you from traditional bias. When the results are consistently awful, as with Carlyle, you don’t have to worry too much about what’s going on in your knowledge gap, because whatever it is is generating failure. The stuff we can see is enough to damn the stuff we can’t. It’s a mystery where a corpse is stinking up the crime scene; you’re looking for cause of death, but you’re pretty sure something’s wrong.
With Mike Babcock, the gap generates success, most of the time. Many of the things he’s doing I can see, some of them I can’t, and most of them work. For Babs with the Leafs, it feels like the early stages of a ten-thousand-piece jigsaw puzzle. The picture looks pretty good—so when a piece doesn’t seem to fit, how hard do we squint before saying saying it’s wrong?
So far under Babcock and Lou, when they’ve done something I haven’t understood, I’ve tried to put myself into a mindset where I make the case for the decision the Leafs made, and see what I turn up. I did this for Byron Froese and for Leafs fourth-rounder Keaton Middleton. I wasn’t totally convinced in either case, but both times, I was able to find enough that I at least thought the decision made more sense after I finished than before I started.
I tried to do this for Matt Hunwick, and I got nowhere. I can’t see an angle where his play is good for this team. And it drives me nuts.
Because Mike Babcock is a good coach, I try to find a reason that makes sense to me. After all, he played Matt Hunwick and Morgan Rielly as a pairing last year, even though it forced Rielly onto his off-side. Rielly is a core piece of this franchise; maybe Hunwick was teaching him things that are useful long-term, even if Hunwick wasn’t capable of performing at a high level himself. It’d make sense, right?
The only problem with this is we have no idea whether it’s true or not, and we never will. (And for what it’s worth, Babcock has given quotes saying Hunwick earned his spot through his play, which suggests he was seeing something the rest of us can’t.) The truth is, if we’re devising theories without evidence to save every instance of Mike Babcock’s decision-making, maybe that’s a sign we’ve gone too far in trusting him.
The only benefit of the doubt to be given Ben Smith is because penalty killing is a little trickier to adjudge in shot differentials and tiny samples, and because he’s only played a few games. But there’s no number by which he looks good, and he’s rapidly approaching Hunwick territory. I definitely can’t see a case for playing him over Holland, and that’s who he’s displacing.
Hyman is trickier for me. I started noticing him more because of what Mike Babcock said, and I liked what I saw. I trust that Babcock has some knowledge of what he’s doing with role variation on his lines, and there’s an intuitive logic to it. If you want, you can say Hyman as 1LW maintains our scoring depth, and that plenty of the good stuff he’s doing shows up better in advanced stats—what, you nerds don’t like Corsi now?—than it does on the scoresheet. Even granting that, I can’t accept that he belongs where he is if he can’t produce more than this.
There’s no question at all I’ve given Babcock more benefit of the doubt in all three of these cases than I would have given Carlyle. I think that makes sense, because Babcock is a much better coach. But the numbers that tell me he’s better are the same ones that tell me he’s wrong about Hunwick and Smith, and the most obvious stat of all makes me question him on Hyman. At a certain point, there isn’t enough that can be going on in the gap to make excuses.
The easy response to little decisions like this is “Who cares? It doesn’t matter.” Not in a big way. Hunwick is a 6D, after all; Smith a 4C. They’re two of the least significant roster choices, and while they almost certainly make us worse, it’s probably to a limited degree. Who plays LW on our top line? Well, it’s a rebuilding year. Who pairs with our best defenceman? Well, Gardiner-Polak has worked decently enough early that I didn’t use it for this article.
As people who write, think, comment, and bicker about the Leafs, we’re continually trying to strike the balance between credit where it’s due and sensible skepticism. It’s easy to say—I assume we all would—that Babcock is a very good coach who sometimes makes mistakes. But how far something goes before it gets called a mistake is a moving target, and one we’re sometimes happy to miss. I want to believe Babcock is doing everything right, and Lou is doing everything right. So do a lot of other people, to judge by the convoluted explanations I saw for the Matt Martin signing. After the horrors of the Leafs since 2005, we want to believe this new FO understands every problem we had with the old FO, and that Babcock understands and avoids every mistake Carlyle made. Except every now and then, Matts Martin or Hunwick crack those paradigms, and the jigsaw puzzle starts to look less attractive.
On balance, the picture’s still good, especially behind the bench. The systems that have made the Leafs a shot-generating machine; the tactical savvy that helped shut down Connor McDavid last week. And the willingness not to be hidebound—Babcock has played a very young lineup, with limited NHL experience, and trusted them more than many coaches would. If he has faults, you put them against his virtues, and his virtues win in a knockout.
But if we’re going to engage seriously with this team and how it’s doing, we’re going to have to engage with Babcock’s limits, and to figure out as best we can what’s happening in the gaps. So no, I’m not smarter than Mike Babcock, and I’m never going to be. I just want to keep trying to shrink the gap.