Thomas Harley Is Having a Quietly Excellent Start to the Season

Today, I wanted to point out something that might have escaped you as much as it has me, thus far. In terms of both underlying numbers and raw data points, Thomas Harley may well be having the best season of any Stars’ defenseman.

Let me just admit for the record that this really did catch me off guard. Until the last month or so, I hadn’t been exceptionally impressed when watching Harley’s game this year. But in recent weeks, his and the Stars’ play has seemed to be rising, and it’s reached a point where, a third of the way through the season, Harley’s play (and Heiskanen’s slower start) have become too noticeable not to, er, notice.

And of course, we all know that Heiskanen is a world-class defenseman and the most important skater on the team. Even suggesting that another blueliner might be having a better start seems almost sacreligious. I know all these things, and I’m not trying to be provocative, just informative. Because Heiskanen and Harley are tied in team scoring, but Harley has done so despite fewer minutes and half the power play time, all without sacrificing defense in the process.

Actually, it’s better than just relative team performance. In fact, Harley having one of the best seasons of any defenseman in the NHL when you adjust for rate. The caveat in front of all this, of course, is that rate stats this early can be deceptive, especially when it comes to third-pairing defensemen. None of us is trying to jump the gun and declare that any heavily sheltered defenseman would actually be a top-pairing beast, if only the mean coach would just give him a chance. Much heavier usage does not at all ensure similarly better performance in the NHL.

But Harley is neither third-pairing nor heavily sheltered, and his numbers aren’t deceptive. He’s been playing 22 minutes a night, which has surpassed even Esa Lindell(!) so far. And that’s with many of those minutes being of the most important kind. The Stars have three heavily-used defensemen in Harley, Lindell, and Heiskanen, and then they have their Other Ones.

Okay, here’s the fun stuff: In 5v5 scoring, Harley is only 4 points behind Cale Makar for the league lead, albeit with more secondary assists to Makar’s tons of primaries. He’s also top-10 in the NHL in points scored per minute at 5v5 among defensemen.

Makar, of course, also piles up points on the power play, where Harley has scored none himself (somehow?). But I’d also point out that Heiskanen only has three assists in over twice as much power play time, so perhaps the “somehow?” is answered by “it’s the Dallas Stars this year, for some reason.” Table it for now, because we’ll talk about the power play in a bit.

Back to the league at large, Harley’s underlying numbers are likewise outstanding. Per Natural Stat Trick, among the roughly 100 defensemen who have played at least 450 minutes at 5v5 this year, Harley is 2nd in the NHL in actuals goals per minute (GF/60) while he’s on the ice, and 3rd in the NHL in Expected GF/60. So it’s not like he’s just lucked his way into great offense happening when he’s on the ice; the process says his numbers are what you would, ahem, expect. Harley is just making life easier for everybody else on his team from the get go.

He’s also 22nd on that same list in his rate of expected goals against (xGA/60) though just 33rd in actuals GF/60 (behind Lindell, by the way, who has radically outperformed his expected goals-against). That’s why Harley is 4th on that same list in xGF% overall (and 6th in actual GF% overall). He’s not just giving up chances in order to create them, bringing chaos to all her encounters; rather, he’s tilting the ice, with over 63% of all goals scored during his 5v5 time going into the other team’s net. And when almost two out of every three goals are happening at the correct end of the ice, you’re probably doing something right. Many things, in fact.

How much is Harley tilting the ice? Well, now we get to look at some charts that show where shots happen (and don’t happen) when he’s on the ice.

These lovely dealios show spots on the ice that get dark the more drastic the difference from league average, So, red means more shots than league average, and dark red means way more than league average; blue means fewer shots than league average from that location, dark blue way fewer.

So you want more blue–a negative number–on defense, and more red–a positive number–on offense). But of course, where (and how) the shots happen is also a factor in calculating the xGF number, and we all know NHL game crew folks don’t always code specific shots correctly. But across nearly 30 games’ worth of data, I’m confident these charts are speaking the truth, as is Micah McCurdy, who is smarter than I am, but only because he lives on a beautiful Island while I’m in Irving, Texas. But I digress:

Yeah, that’s what you want to see if you’re a forward, or a goalie. Harley might not be getting quite as loaded up with minutes as Heiskanen (and who ever is?), but he’s absolutely crushing the minutes he is getting, which is also why he’s tied with Mason Marchment for the team-best +15 in plus/minus, if you care about conventional things.

Oh, and to return to the power play: if you want to start asking for Harley to take over the top power play, the answer is that whatever power play he’s on has already been the top power play by default this year, as the Stars have scored at a higher rate with him on the power play vs. Heiskanen, despite Harley’s lack of points himself. Here’s an ugly screenshot from that link, in case you’re allergic to clicking:

But what, exactly, has been happening with Harley on the power play vs. Heiskanen? Well, because, I think, of what you can see in these final charts:

(Special Teams charts on HockeyViz are orange/purple to distinguish them from the red/blue of 5v5, just so you know, but otherwise they work the same way):

As you can see, Heiskanen’s power play time has seen more shots in that Jason Robertson area, curling in from the right point and shooting, whereas Harley’s time has seen pucks consistently funneled to the net, or passed through the high-danger area right in front of the goal, as Duchene, Seguin, and Marchment have been so good at doing this year, at 5v5 and the power play alike.

I have to think that’s a conscious decision designed to play to the respective strengths of those players. Heiskanen has generally been playing with Robertson, Hintz, Johnston, and Benn, while Harley has largely played with the Duchene line and Stankoven. But if you’re really looking to shake up the power play, I might see what happens if you put Harley with Robertson & Co. for a while. At the very least, I don’t see how it could hurt, if only to jolt some folks back into their historical form and get the power play a bit more enlivened against aggressive penalty kills.

In conclusion, Thomas Harley scored two goals the other night that he called “muffins,” and maybe that’s because he’s been feasting on the opposition all year. Right now, he and Heiskanen have identical scoring lines of 4 Goals and 10 Assists, but in Harley’s case, he’s done it with a lot fewer advantages. Maybe it’s time to think about giving him the biggest advantage, at least for a couple of games. But regardless, just having Thomas Harley has been a huge advantage for Dallas, and I suspect that will continue to be the case for a long time.