Why ‘Computer Boys’ aren’t reason Panthers flopped (Trending Topics)

It’s gonna come as a big surprise to you, but the reason the Florida Panthers struggled isn’t because they went with an analytics-first philosophy.

The decision to fire Gerard Gallant was an understandable one at the time — he basically complained to the media about the team’s roster construction, which will get you canned from a lot of jobs — but it didn’t work out in the long term. Now that they’re about to hire a new coach for the third time since 2014, and Dale Tallon is the GM again for some reason, apparently it’s time for a referendum on how the Computer Boys Era went.

[Follow Puck Daddy on social media: Twitter | Instagram | Facebook | Tumblr]

The shocking take from the old guys in the hockey media is that analytics didn’t work. The shocking reality is that a bunch of factors came together to really hurt their performance in a number of different ways, not the least of which was that Tom Rowe ended up being worse at coaching than Gallant.

Not to say Gallant was a genius or anything (he was perfectly adequate) but the team felt like it wanted to take the next step after investing heavily in solid player acquisitions this summer. Rowe’s work didn’t allow them to do that, nor did all those other factors, so here we are at another crossroads, and another season without the playoffs.

The most important thing for the Panthers in terms of overall success, or lack thereof, is obviously the way in which percentages swung for them over the past three years.

Pretty simple: They shot 7.9 percent in 2014-15, when they finished with 91 points. Then they shot 9.8 percent last year. This year they shot 7.8 percent.

And in 2014-15, their goalies stopped .915, which rose to .921 last year, then fell back to .915 this year.

It’s not always easy to explain why shooting percentages come and go — though I’ll talk about that in a minute — but the goaltending issue is easy to diagnose. Two seasons ago, Roberto Luongo was great (.921) but his backups were terrible. Last year, both Luongo and Al Montoya had phenomenal seasons (.922 and .919). Then this year, Luongo took a step back (.915), which happens sometimes when you’re 37, while James Reimer remained good (.920), and Reto Berro was horrible in seven appearances games (.876, and he picked up a loss in five of them).

It should be said that .915 is still a little above the league average, but .920 as a unit is really hard to do without one goalie getting up into the .930s or so. To illustrate the difference, let’s say the Panthers only got .915 goaltending last season; that costs them 15 goals and about five points in the standings. Think anyone is flipping out about a 99-point Florida team?

Hitting 100 is a big deal to people. It signifies something in a way that even 99-point seasons do not. It’s silly but it’s true.

As for shooting percentage, well, there were a lot of factors at play there, as you might imagine. But to open the discussion, note that just about every big-minutes player from last year’s team saw at least some decline in their shooting percentage this season. Jagr, huge hit. Barkov, smaller hit. Huberdeau about the same, but he missed a ton of games, and whoever they used to replace him just couldn’t measure up.

The Jagr-Barkov-Huberdeau line was of course the big goalscoring engine for Florida last season. They spent more than 602 minutes at 5-on-5 together, and scored more than two-thirds of all the goals they got on the ice. They were able to generate tons of scoring chances by working together in a unique way, and just pounded opponents. Did they get lucky during that campaign? You bet they did; in terms of expected-goals, they “should have” scored about 28.5 goals together. They actually scored 36. Huge difference.

Because of the big chunks of the season both Barkov and Huberdeau missed (21 and 51, respectively) this huge line only got about 284 minutes together this season. That’s a massive decline, and it means that a lot of time normally distributed to the team’s top offensive players had to go, instead, to lesser options. But even in those minutes this big three was together, they had their ups and downs. They generated fewer scoring chances. They allowed more.

Altogether, the big line in Florida took only a small hit in expected-goals-for per 60, but saw a lot more expected-goals-against. And instead of getting things to fall for them in terms of shooting luck, they instead were just in the area they “should have” been. And they got a little lucky in their own end. In the end, they ended up with a goals-for percentage of 59.1, down appreciably from 67.9. And combined with the far lesser minutes they actually got as a group, the impact on the Panthers’ fortunes overall was, shall we say, non-trivial.

But the real problems were deeper down in the lineup. The Panthers’ top-six was actually pretty good. Not great overall, but if you consider how much they dealt with, injury-wise, it’s impressive. Meanwhile, their bottom-six was a disaster. They got badly outscored, as you might expect, and for all the complaining Gallant did about how he didn’t always get to use guys he wanted in those roles (like Shawn Thornton or Derek McKezine) it’s worth noting that Rowe also used them, and they got outplayed pretty much universally.

That is, on some level an organizational failing, simply because there were a lot of guys on the roster who aren’t good enough. If Chicago can identify and sign good depth players (either from across the NHL, through the draft, or via amateur free agency) there’s no reason Florida can’t do the same. But they didn’t. And in fact, a lot of the older guys who dragged the bottom of the roster down were legacy contracts from when someone else was in charge.

Can you guess who? Here’s a hint: He was just named general manager again.

Not that the Computer Boys didn’t make their mistakes, too. Leaving aside the inexplicable McKenzie extension, I’d argue too many dollars and years went to their current guys coming off unsustainably strong 2015-16 seasons. In the UFA market, you gotta pay what you gotta pay, and if Yandle (41 points), Demers (solid middle-pairing guy stuck with problematic players), and Jon Marchessault (30 goals) collectively cost you $11.6 million, I think that sounds just about right.

But it came because Marchessault was a steal. The other two, I would argue, simply lived up to their deals, rather than exceeding their expected values, though relative to their ages, that’s not a good sign for the long-term future. The good news is Marchessault is still super-cheap for next season, and even if he won’t hit 30 goals again, he’ll deliver goals significantly above almost any other $750,000 player in the league.

What people seem to forget is that this team dramatically over-performed, appearing to arrive a little ahead of schedule in the rebuild. This was a team that still had too many veteran role players who aren’t good enough to make a real difference as the league continues to change. The Panthers have plenty of guys coming in over the next year or two who can probably fill those roles more ably and, if they get the right kind of coach (Dallas Eakins?) they might be able to kick up more reproducible positive results.

The problem this season was, ultimately, defense. Ekblad had concussion issues that held him out of a bunch of games, and Rowe put him back in the lineup before he was ready. The impact of losing Brian Campbell cannot be understated, though Yandle made up a good chunk of the difference. Michael Matheson scans as a pretty good defender going forward, but he was a 22-year-old rookie this season getting the third-most minutes on the team. Mark Pysyk was probably not good enough as a No. 5 D, but the patchwork of Alex Petrovic and Jakub Kindl as your No. 6 is going to hurt you every time.

The good news is Pysyk’s and Kindl’s contracts are up and the Panthers have money to spend upgrading the blue line, with relatively fewer concerns up front (not having Shawn Thornton for coaches to coo over is going to provide more help than it should).

This is a team of which too much was expected. It had a lot of flaws, but you have to Trust The Process and assume things will be figured out. The vast majority of moves this team made in the past few years as it got smarter indicate they know what they’re doing. But injuries to incredibly important players, weird coaching problems, no longer getting all the bounces, and issues at the bottom of the roster all came together in a horrific head-on collision.

Doesn’t mean this is suddenly a bad team. All those things happen to everyone. Just usually not all at once.

Ryan Lambert is a Puck Daddy columnist. His email is here and his Twitter is here.

All stats via Corsica unless otherwise stated.