Evaluating Willie Desjardins’ Lineup Decisions
7 years ago
Ice time isn’t given. It’s earned. That’s an adage that’s been repeated across the hockey world, and Vancouver is no exception.
It pertains especially to younger players, and Canucks’ President of Hockey Operations Trevor Linden repeated the common refrain earlier this year pertaining to sophomore forward Jake Virtanen:
“As long as I’ve been part of the game, especially at the pro level, there’s an element of earning your time… I think a lot of players would say, “Give me more time and I’ll play better’, and the coach would say ‘Play better, and I’ll give you more time.'”
That’s a fair stance to take, especially with a young player. It’s not as though the assessment isn’t without merit, either. During his tenure with Vancouver, he was the team’s wort forward by shot shares and struggled to produce offence, even in a limited role.
When it comes to a player like Virtanen, it’s easy to see why the front office, and Willie Desjardins by extension would want a player to prove himself before being gifted an abundance of ice-time at the NHL level. It becomes less clear what exactly the thought process is for Willie Desjardins when we begin to look at the players whose lineup spots are not in jeopardy. Players like Mike Chaput and Jayson Megna haven’t exactly been held to the same standard, which raises questions about whether or not Desjardins is really doling out ice-time based on merit.
In Vancouver, coaching decisions will always be heavily scrutinized. Second-guessing occurred under Alain Vigneault, when the Canucks’ success was at it’s zenith, and it happened during the mess that was John Tortorella’s single season as the Canucks’ bench boss. Lineup and deployment criticisms have been a recurring theme in the Willie Desjardins era as well, and those criticisms reached a fever pitch on Sunday afternoon, when the Canucks announced that points-per-sixty leader Sven Baertschi would be a healthy scratch against the Washington Capitals.
Gallons of digital ink have been spilled on the subject of Desjardins’ questionable player deployment, whether it’s which line goes out after an icing or a TV timeout, who’s playing when the team is behind, or the insistence on keeping the Sedin twins and Brandon Sutter together.
All this got me wondering about just how bad Desjardins has been at making decisions related to personnel. It sure seems like he’s especially bad at it, but we’ve seen the way Alain Vigneault loves Tanner Glass, so it’s fair to wonder how Desjardins compares to his peers.
(Spoiler: It’s not great.)
EVEN STRENGTH
We’re going to talk about Brandon Sutter again. I’m sorry. It’s just so much worse than I originally imagined. The Sedin-Sedin-Sutter line, Vancouver’s most-used line combination this season, is currently sporting an atrocious expected goals-for percentage of 36.16%. xGF% is a metric that’s designed to predict goal-differential based on the quality and quantity of shots a player or combination of players creates versus what they give up. The reason I’ve chosen to focus on this is that it should, in theory, be something that even an old-school coach should be concerned about and have an eye for. At even strength, the Sedins and Sutter have performed worse by this metric than the Dorsett-Cracknell-Prust combination from last season.
Here’s a list of lines to get more than 150 minutes of even-strength ice time that have carried a lower xGF% since 2007.
That’s it.
Just five, in almost a ten-year span. Four of those lines feature at least one career fourth-line forward, and the fifth was a combination on the 2014-15 Colorado Avalanche, a team with a notoriously permissive system.
In other words, keeping this line together has been a historically terrible idea, and one that’s basically unprecedented for a team’s de facto first line. The only other team whose most common line combination is also it’s worst by xGF% is the Philadelphia Flyers.
When we look at pure shot-attempt differential, things remain ugly. Sutter and the Sedins are 40th out of 49 lines with over 100 minutes of even-strength TOI in score adjusted CF%, and last in goal-differential and scoring-chance differential. It’s clear Desjardins has had an obviously better option available to him, considering that most of those categories improve by at least 7 percentage points when Sutter is replaced with their next-most frequent linemate, Loui Eriksson.
This issue extends beyond just playing Sutter on that line, however. Among players currently on the roster, Sutter is last on the Canucks in terms of shot-attempt differential. He leads the team in total time-on-ice. When compared to other teams, this is an anomaly.
87 Forwards have played over 350 minutes at evens, Sutter is 82nd in score adjusted CF% , and 83rd in P1/60. So if you think Willie’s over-reliance on Sutter is overly heinous, that more or less bears out when you look at the underlying numbers. Sutter is the only player in the NHL this season to lead his team’s forwards in ice-time while simultaneously coming dead last in even-strength shot-attempt differential.
So, if you’re still using Alain Vigneault and Tanner Glass or Mike Babcock and Matt Hunwick as a defense for this kind of decision-making, you can stop now. We’re on a completely different level.
When looking at defense pairings, things look better, but still not great. Twelve defense pairings have played more than 200 minutes at 5v5. Among those pairings, Biega-Bartkowski and Hutton-Gudbranson are second and fourth respectively in ice-time at evens, but are 11th and 8th respectively in xGF%. This is especially relevant this season because Gudbranson-Hutton is on pace to be Desjardins’ second most common pairing by the midseason mark, in spite of showing poorly by both the eye test and the underlying numbers, and exhibiting almost no chemistry at all. It’s important to note the effect injuries have had on the defense pairings, however, so I’m willing to hear arguments on why you’re willing to give him the benefit of the doubt.
That being said, there’s still the fact that Matt Bartkowski is still the single-season leader for ice-time while trailing by a goal. So we can say with almost complete certainty that over-reliance on sub-par players at even-strength has definitely been a pattern under Desjardins.
PENALTY KILL
The Canucks’ PK has traditionally been quite strong, and that’s continued under Willie Desjardins. That doesn’t mean he isn’t undermining the team’s success by consistently trotting out the wrong players, though. Luca Sbisa and Erik Gudbranson have allowed the highest rate of unblocked shot attempts at 5-on-4, but have received the bulk of those minutes. Conversely, Ben Hutton has allowed the least, and also had the smallest share of the 5v4 minutes. In Hutton’s case, I’d worry that’s sample-size related, but he was their best defenseman by this metric last season as well.
In this regard, the Canucks are an anomaly. The majority of teams are giving the highest 5v4 minutes to their best or middle-of-the pack shot suppression pairs. Only five other teams use their worst pair by FA/60 (shot suppression) as their main penalty killers: Buffalo, Calgary, Ottawa, Washington, and Los Angeles. In the case of the latter two teams, however, the spread in difference between shot suppression rate and ice-time is so small that it almost isn’t worth noting.
The stubbornness isn’t a good look considering the other options available, but again, injuries are playing a role. I imagine once Edler and/or Tanev return we’ll see a decrease in ice-time for Sbisa and Gudbranson. There’s also the fact that the Canucks’ PK is currently sitting at sixth overall, so you can consider this a minor quibble. Still, it’s indicative of a deeper trend of failing to recognize which players are best suited for which roles.
POWER PLAY
I’ll try not to belabour the point about Brandon Sutter, but it’s clear keeping him on the first unit PP is submarining the Canucks’ ability to produce offence. He’s been the Sedins’ worst winger at 5v4 since Steve Bernier, so it’s clear that’s not really working.
As bad as that is, this is an area where the Canucks’ coaching staff is closer to the pack when it comes to lineup decisions. They’re still giving the most ice-time to their best point-producers, and Loui Eriksson is starting to come within striking distance of Sutter’s time on the power play with the Sedins. The team’s issues when up a man are most likely systems related, which is somewhat ironic given that that’s been an area of relative strength at evens and on the PK.
THE IMPORTANCE OF PERSONNEL DECISIONS, AND UNCONTROLLABLE FACTORS
When taking all of this into account, one might get the impression that Willie Desjardins is one of the NHL least effective coaches, but that isn’t necessarily the case. This may look like a deep-dive, but it’s really only a cursory glance. There are so many facets to evaluating coaching that making any hard-line conclusions is ill-advised. Decisions relating to personnel are only a small piece of the tactical puzzle, something J.D. Burke was quick to point out in this week’s mailbag:
Desjardins’ tactical mishaps (lineup decisions and deployment alike) are blown greatly out of proportion. The reason for that is the Canucks don’t have the players to mask those types of decisions.
Hear me out on this one. Other coaches make lineup decisions consistently as bad if not worse than Desjardins. The Rangers healthy scratched Kevin Hayes (a considerably better player than Baertschi), but it’s less noticeable with the bounty of players ready to fill that role. When the Capitals put Jay Beagle on their first line, they still had Evgeni Kuznetsov and Niklas Backstrom on their second and third lines. The Canucks have Jayson Megna.
Tactically speaking, there’s so much to like about Desjardins’ system. Especially his adaptability. If the Canucks are bottling it up this season, it’s because it’s the only logical response when you’re constantly the lesser team.
Whatever the case, Desjardins is not the problem in Vancouver. Not by a long shot.
I remain skeptical that there are any coaches as prone to awful lineup decisions as Desjardins, but J.D. raises a two key points here. The first is that, by and large, we as fans and media have a tendency to overvalue small personnel and deployment decisions over systems, even though the latter has a greater impact. The reason we do this is because of what’s called the availability heuristic, which is a fancy name for the human brain’s tendency to make decisions based on what can be easily recalled. In other words, it’s much easier to remember all the times that Derek Dorsett has come on as the extra attacker than it is to remember small in-game adjustments that have helped the team close out a game.
The second point is perhaps even more important. The Canucks really just don’t have the horses to compete on a regular basis. Yes, the Sutter-Sedin line has been historically terrible and yes, the Gudbranson-Hutton pairing should probably be broken up. But Desjardins is in a bit of a bind here. It’s unclear if Desjardins truly believes his questionable combinations and pairings have played well, or if he’s been forced into bad lineup decisions due to injuries and poor depth. The optics of deploying players like Chaput, Skille, and Sutter in your top six are terrible, but these players are still going to have to play at least ten or so minutes a night anyway. Is the drag they have on a line’s performance worth spreading your skilled players up and down the lineup? It’s exceedingly difficult to know for sure, but when your most-used line is also one of the most inept in modern history, it’s hard to give you the benefit of the doubt.
I remain unconvinced that Desjardins is the problem in Vancouver. I just don’t think he’s the solution, either.
