In defense of Andreas Johnsson

Follow me on Twitter @ToddCordell | @InfernalAccess

Be sure to join the Discord channel to talk hockey, and everything else, with me and fellow subscribers.

---

Andreas Johnsson wasn’t as productive as I, or anyone else, expected him to be in his debut season with the New Jersey Devils.

He scored just five times and picked up 11 points in 50 games, which equates to fewer than 20 points over a full 82-game schedule.

Disappointing? Absolutely! But I don’t think the points were telling with regards to his overall game.

He was much better than most give him credit for and, in this post, I’m going to dig into why.

On-ice numbers

• The objective in hockey is to out-score your opposition. If you can’t do that, you at least want to hover around even. Johnsson did that. Despite absolutely nothing going for him all season long, he still finished only minus-1 at 5v5 in 50 games. The Devils were out-scored by 21 in that gamestate so having an every-day player come in ~even is a positive. Among players who finished the year on the Devils, Johnsson’s on-ice goal share ranked him 5th; sandwiched between Jack Hughes and Janne Kuokkanen.

• ‘Really Todd, you’re leading with goals?’ You’re right. Goals alone can absolutely be misleading. The objective, again, is to have more than your opponent. But since they are such rare events – especially compared to shots or scoring chances – they sometimes aren’t indicative of how a player actually performed. In this case they aren’t…but not for the reason most would think. I believe it is Johnsson’s negative goal differential, although by *one*, that is misleading. He deserved better than that.

The Devils controlled 52.34% of the expected goals with Johnsson on the ice, slotting him behind only Jesper Bratt and Hughes up front. They fared even better in terms of standard scoring chances, finishing with a 54.74% on ice share. Only Hughes bested Johnsson there.

Put simply, the ice was tilted in New Jersey’s favor with Johnsson on the ice. Yes, he squandered what felt like hundreds of chances. But even so, he still posted better underlying numbers, and on-ice results(!), than the vast majority of the team.

Ty Smith produced points and was -8 at 5v5 (although that was also undeserved). Miles Wood produced points and was -6 at 5v5. Pavel Zacha had a bunch of points and was -14 at 5v5. So while those guys found the scoresheet regularly, and Johnsson botched seemingly every scoring opportunity, the latter still garnered better on-ice results. And the underlyings suggest he deserved to.

• Let’s say Johnsson flat-out sucked offensively across the board. We’ll pretend that he didn’t produce because he had no chances whatsoever; not because he couldn’t finish any. He still brought value to the table defensively.

The Devils gave up only 47.19 shot attempts per 60 with Johnsson on the ice. Only Hughes (surprise!) did a better job at suppressing shots this season. New Jersey allowed just over 21 chances per 60 with Johnsson on the ice. Again, Hughes was the lone year-long Devils player to best Johnsson in that regard.

But Johnsson didn’t just limit shots and chances for the opposition. He limited goals. The Devils gave up just 2.15 per 60 with Johnsson out there. Exactly zero (0) year-long Devils fared better in that regard. Zero.

Bratt? 2.30. Hughes? 2.46. Yeezy? 2.88. Nico? 3.44. Zacha? 3.58. That’s right, Johnsson was an excellent shot/chance suppressor and his efforts rightfully led to very few goals for the opposing team.

Individual numbers

• ‘That’s all great, Todd, but he had 10 points.’ That’s a very astute observation. He did, in fact, have 10 points. But despite what most think, it’s not because he was the most useless player of all-time and played like garbage. A lot of the issues were percentage driven.

Individually, Johnsson generated 6.86 scoring chances and 0.57 xG per 60 minutes of 5v5 play. But a year ago he averaged 6.53 scoring chances and 0.53 xG per 60. So, he generated more for himself this season than last. And he produced at an 18 point pace (compared to a 40 point pace) while his 5v5 efficiency was cut almost exactly in half.

If the volume was there, why wasn’t the production? Again, percentages.

In 2020-21 Johnsson shot 7.69% across all gamestates. He scored on nearly 16% of his shots in 2018-19 and 10.26% in 2019-20. Even if we admit that 2018-19 will be an abnormality, it’s probably fair to say he’s a better finisher than he showed this season.

• Perhaps more importantly than a personal lack of finishing, nobody on the Devils could score when Johnsson was on the ice.

The Devils converted on only 8.38% of their all-situations shots with AJ out there. That’s a far cry from the on-ice shooting percentages he posted in the two years prior (10.49% in 18-19, 9.98% in 19-20).

It may seem like we’re splitting hairs talking about such a small percentage difference. We are not. A 9.98 on-ice SH% would’ve led to six more goals while a 10.49 on-ice SH% would’ve led to eight more goals. And that’s over just 50 games.

To put into perspective how much of a difference that makes, eight goals was the gap between Tampa Bay’s 5v5 offense and New Jersey’s. Eight goals was the gap between Boston’s 5v5 defense and Calgary’s.

• ‘How does, let’s say, eight goals explain the difference between an 18 point pace and a 40 point pace?’ Great question, I’m glad you asked.

Beyond the fact Johnsson nor his teammates could finish anything when he was on the ice, he also suffered from a big dip in IPP (individual points percentage). That is, quite simply, the percentage of goals scored a player picks up a point on.

Johnsson came in at 40%, picking up eight 5v5 points on 20 5v5 goals. That is *extremely* low for a forward. Mikhail Maltsev (50%) was the only other Devils forward to come in below 55%.

In the two years prior to this Johnsson’s IPP was 65%. Had Johnsson finished in line with that average, he’d have picked up 13 points on 20 5v5 goals.

That’s still nothing to write home about, I know, but it’d put him on level terms with Jason Zucker, Anthony Cirelli and Patrik Laine, to name but a few.

Parting thoughts

Johnsson was an awesome defensive forward for the Devils this season and, by all accounts, his offensive struggles can be pinned more to an unusually, and unsustainably, large dip in percentages. Be it individual shooting percentage, on-ice shooting percentage, or IPP, Johnsson took it on the chin this season.

Simply put, the Devils scored fewer goals than they *should* have with Johnsson on the ice and he picked up fewer points than he *should* have on the goals they did score. That’s why his production fell off a cliff; the shot and chance generation numbers were still all there.

That’s why I remain bullish on what Johnsson can contribute moving forward. Maybe he’s not the 20+ goal, ~50 point guy he looked like he could be a couple years ago. But he could certainly get back to the ~40 point range; while providing plus-defense on top of it.

I would not be handing him over to Seattle in expansion, and I certainly wouldn’t be selling low on the trade market.

I’d keep him and enjoy the regression next season.

---

numbers via NaturalStatTrick.com, Evolving-Hockey.com, and JFresh’s Tableau page.