I’ve noticed something in Leafs’ fandom this summer. No one seems upset that the Leafs signed Tyler Ennis — his contract is one-year, it’s cheap, no one ever says anything bad about him. But no one thinks he’s making the team in a meaningful way, or at all in some cases.
I did see one sarcastic comment that there will be fans who will be upset he isn’t played instead of other fourth liners, which is a phenomenon I call the Seth Griffith should pretend to be a centre fallacy.
But hold up a little. Ennis isn’t Griffith. Why exactly do so many think he is going to be sent to the AHL or be a press box extra? Why, in other words, is he joined up in so many minds with Adam Cracknell and Josh Jooris as callups, but not roster players?
On the one hand, few people’s hearts are lifted at the idea of Tyler Ennis on their team these days. He’s not got the cachet of some of the unexpectedly popular but not actually very good players. I assume that if P.-A. Parenteau announced he was coming out of retirement, that would excite people more. Ennis is also on his third team in three years, and he’s not what he used to be.
In order to know who he is now, we need to look at his past.
Who was he?
We know the Buffalo story, but sometimes it’s easy to forget the scale of the disaster. When Arvind looked at the best offences since 2004, Buffalo’s 2006-2007 season was second. They scored 300 goals that season and had 113 points. And then they ditched all their star players and ended up slipping to 90 points the next year.
In the summer of 2008, they drafted Tyler Ennis 26th overall.
Ennis played for years in Buffalo, beginning in 2009-2010. In his time there, the team finished with 100 points in his first season where he played 10 games, and then it was: 96, 89, 48 (lockout year, last in their division), 52, 54, 81, 78. So, he had one full season where the team wasn’t slow-motion crashing into an iceberg. It’s not really surprising that his career high in points came in the 2010-2011 season where he had 20 goals and 49 points.
What is a surprise is that he nearly duplicated that result in 2014-2015 — a year his team came last — so he has shown the ability to be the one good player on a very bad team. His goals dried right up after that in an injury plagued season, but there is some indication that his time in Minnesota last year wasn’t the last way station on the way to the AHL or a failure that necessitated a buyout.
He has not, by the way, played in the AHL since 2010.
His last Buffalo season
The year before last, 2016-2017, was his last Buffalo season. They were bad but not desperately awful. The ship was on it’s way up from the bottom of the sea, or so it seemed.
Jack Eichel was entering his second season, and the Sabres had just paid a boatload for yet another player haloed in the warm sunny glow of the John Tavares effect. To be fair to the Sabres, Kyle Okposo might be one of the best players to ever be overpaid for having played with Tavares. Ryan O’Reilly was in his second year with the Sabres as well, so they had some good players.
Coming off a season where he only got in 23 games, and seemed to get re-injured after every comeback, Ennis played 51 games, had 13 points and only scored 5 goals. As a solid and predictable 20-goal scorer, this was a huge disappointment. He couldn’t outscore Cody Franson.
A look at his shot locatios shows that he was shooting from the Matthews zone at five-on-five (as in don’t shoot until you can see the colour of the goalie’s eyes). But on the power play, he was shooting a lot from way over on the halfboards which was a totally different pattern to his former glory days.
He’d never been a great personal goal-scorer on the power play, but he’d always been a meaningful contributor with a lot of points. In this new structure (Dan Bylsma was in his second season as Sabres coach and should rightly have been expecting to get fired at any moment), Ennis had three points, all assists, and he played a lot less minutes. The new guys had bumped him off the first unit, and the new system didn’t suit his strengths.
Meanwhile, at five-on-five, he was suffering from a very low shooting percentage; at only 6.58, it was well below his career average, particularly if we leave out the season before where he likely should never have played. His rate of personal shooting, however — his Individual Corsi For per 60 minutes — was bang on his normal amount. He is extremely consistent, and shoots about 13 iCF/60. To contextualize that: On the Leafs of last season, he’d be well back of the phenomenal James van Riemsdyk and the excellent Auston Matthews, but he’d sit just behind Nazem Kadri, William Nylander, Mitch Marner, and just ahead of Patrick Marleau.
In other words, he shoots at a rate more consistent with a second line winger than a depth guy, and he does it consistently.
In that final Buffalo year, he played with O’Reilly more than anyone else, and given his shot rate, his shot location and the excellence of the playmaking ability of his centre, I’m inclined to think he was really unlucky more than he was failing to bring the skill he’d had pre-injury.
I’m not going to talk about on-ice Corsi right now, I’m going to switch to his Minnesota season and see what went on there first.
Minnesota was supposed to be a redemption story
The Wild seemed like a great fit for Ennis. They were a team desperate to score goals, riddled with injuries, and had added Eric Staal to good effect. Ennis should have just shown up and prospered. Something happened to the team last year, however.
While in the past, the Wild had been a blandly mediocre team in the west, they never seemed to get better, and they were becoming a bubble team just as the Western Conference got weaker. Their team shot pattern had hovered at league average for years, and league average means you don’t shoot enough from higher-danger locations to get enough goals to be a dominating offence. They always needed hot goaltending to get them anywhere.
Last year, their shot pattern totally changed to heavy shooting at the goal mouth and much less from the perimeter. The problem is, they were still average at the pace of shooting from the big central to high slot area or from the circles. They’d improved their highest danger shooting, but at the expense of almost everything else they did. None of the defenders shot the puck much ever, which is not a terrible plan, but the offence rested entirely on the forwards who could gain the goalmouth. They had one of the lowest overall shot rates in the NHL, but very few of their shots were blocked.
This pattern shows up in Ennis’s time on the ice, but then, that’s his style of play. He personally shot mostly from in tight last year, and while he was on the ice, the team’s highest danger shot rate was much higher than when he wasn’t.
Ennis played mostly with Eric Staal, Charlie Coyle or Matt Cullen, and that’s the gamut from second line to fourth on the Wild. The offensive pace he created with Eric Staal, one of the highest-rate shooters in the NHL, was Leafs-like. The Wild are rarely that, um, wild, so this was highly unusual.
Ennis was the third-highest rate shooter among regular players on the team, but when he wasn’t with Staal, well, things fell apart. With Cullen, the offence disappeared, and with Coyle, the defence when to hell. Coyle had a very bad year just in general.
Ennis played with them both about as much as he played with just one at a time. All three together had results that looked like a Leafs line you would try to keep out of the defensive zone — huge offensive pace, terrible defensive pace to go with it, but the percentages look favourable, so the weirdest Bozak line comparable ever assembled. But when Ennis was with just Coyle or just Cullen, they saw all offence dry right up, likely because the third man on those lines was usually Daniel Winnik or Marcus Foligno.
Ennis’s last two seasons were more similar than they were different when you dig down. He played with good offensive lines and bad depth in about equal measure. He averaged out to third line minutes, but actually played much higher in the lineup some of the time. He kept to his game, producing a very high shot rate from high-danger areas while being better than average defensively.
The problem is, he never scored in Minnesota either. His shooting percentage there was a full percentage point lower than it had been in Buffalo. His on-ice shooting percentage rebounded from the terrible number in Buffalo to something like a normal one, so his assists came up, but 22 points in 73 games is not going to impress.
He didn’t play enough power play time to even bother looking at it.
Is it possible that a career 10 per cent shooter has become someone who scores like a defenceman because of injuries? Or is this two years of bad luck? That question might tell us both why the Leafs signed him and why his contract is so cheap. I’ll bet even he has no firm idea of the true answer.
Shooting percentage is not a repeatable skill. That means, what you do one year doesn’t define your ability level or predict the next year’s result because there is a lot of variance year over year in all players’ results. It is a skill, however; bad shooters don’t end up in the NHL very often, and that makes it easy to fall into the trap of deciding that there is an explanation beyond the random winds of fate for unusual results.
You can get fans passionate about William Karlsson’s shooting-percentage-fuelled Vegas season who will swear he will be that good forevermore. You will get people who never want to hear anyone say career-high shooting percentage about Nate MacKinnon last year. They just want to give him the Hart, and fair enough, goals are fun; having someone score heaps of them on your team is exciting, and every prospect who rocks an absurd shooting percentage is always going to be the next JvR. That’s just human nature.
But for many fans, when a player has a run of low-percentage shooting, obviously it’s his fault. He needs to stop squeezing his stick too hard, or he’s shooting too much or too little or from the wrong spots or right at the goalie. I saw him that one time, aiming for the crest on the goalie’s jersey, and trade that bum, who wants Nazem Kadri on this team anyway?
Remember that? Did Kadri stay forever in his shooting percentage trough?
If Tyler Ennis had shot at just eight percent his last two seasons, which would still be a career low, barring the injury-filled year, his primary points per 60 minutes would have been:
- 0.95 in Buffalo up from 0.84
- 1.06 in Minnesota up from 0.87
Patrick Marleau’s Primary Points per 60 minutes last year was 1.18.
Players who managed less then 1.06 on the Leafs were: Tomas Plekanec, Kasperi Kapanen, Dominic Moore and Leo Komarov.
I only added 4 and a bit goals to Ennis’s two seasons of play to get to this adjusted number; he didn’t play a lot of minutes. But it’s enough to take him from a player who scores less than a bad defenceman to someone who is a very good fourth line winger, given his excellent defensive results. The margins between a good NHL season and a bad one are just that thin. But Ennis would not be the first player to put in a stretch of bad shooting years, and the less you play, the easier it is for variance to have its way with your results.
That calculation used eight per cent. If I use 10 per cent, closer to his career average, I turn him into a player who could outscore Marleau at five-on-five.
Ennis isn’t going to score 20 goals even if he gets regular minutes and his shooting regresses to his mean. He won’t get the power play time to make that happen, even though I think he’s better on the power play than Connor Brown. As a cut price winger who can take the centre spot in an emergency and who consistently executes a style of play that is exactly perfect for the Leafs, he’s ideal.
And by cut price, I mean, 10.4 percent of Marleau’s cap hit and 31 per cent of Brown’s.
The rest of the picture
Ennis played with such a strange mixture of players his last two years, that I hesitate very strongly to just look at his on-ice Corsi results and draw any conclusions.
By using Corsica’s RelT Corsi and RelT Expected Goals, the results are adjusted by the quality of his teammates and then compares to the team’s average.
For a more detailed explanation, see the story on Adam Cracknell:
In Minnesota, Ennis was above team average in RelT Corsi, and most of that came from excellent Corsi Against results. When you refine that to Expected Goals (which weight shots by location and type) his Expected Goals For is excellent, while his Expected Goals Against is also good.
So he performed well defensively when measured either just by shot volume or by type of shots allowed.
In Buffalo, the picture is similar, with a little less offensive excellence and a lot of defensive quality in limiting shots against, and limiting the highest danger shots against. Being better than the Sabres defensively is a low bar to clear, however.
That’s not exactly the common profile of a small, zippy winger who likes to shoot the puck a lot. And it’s very much not Marleau’s profile. Ennis has significantly better Corsi Against and Expected Goals Against results than Marleau does these days.
So where will Ennis be played? I highly doubt it’s the AHL. I also can’t see sitting him in the pressbox in favour of a borderline AHL call-up like Trevor Moore or Mason Marchment. If the Leafs played an old-style top-six and bottom-six structure, he’s absolutely be in the running for a third line winger job. Given the Leafs’ structure, it’s more likely he’d grab a fourth-line spot, particularly if the fourth line also has Kasperi Kapanen and a fast-skating centre on it.
And if his shooting does regress to his mean? He’s going to look overqualified for that fourth line job really quickly. If it regresses to the point he’s actually producing more than Marleau? That’s an interesting and awkward situation to navigate. What a nice problem to have, though, that the depth guy might be too good.
This story relied on Hockey-Reference, Hockey Viz, Natural Stat Trick and Elite Prospects. All RelT or Expected Goals stats are from Corsica.