Preseason Predictions – Canucks 2016-2017

Last year I ran a projection for the Canucks’ season. I’ve done the same this year and hope you’ll review the results.

Please note, this is a mathematical exercise. I try to remain objective and neutral as best I can.

Here’s the process:

I begin by using past NHL seasons (or NHLEs as applicable) to determine an estimate on the points-per-game for each player on the listed roster. I then extend that to the most likely call-up candidates within the organization who will fill the approximate number of man games required both at forward and defence.

Next, I will estimate, again based on an average adjusted to circumstance over previous seasons, the number of games each player on the listed roster is likely to play in that season, topping it up to within a reasonable margin with the call-up candidates mentioned above.

That gives me a rough estimate of the total number of points the team is likely to score.

I analyze each skater’s strengths and weaknesses and try to describe their likely and ideal deployment.

I then estimate, using similar methods, the games played and save percentage for the goalies. I estimate, based on previous seasons and adjusting approximately based on the relative strength of the defense corps, the number of shots that each goalie will face in a game (I have noticed that backup goalies will often face roughly 1 fewer shot per game than the starter as teams may play more cautiously in front of a backup). Calculating those shots against the estimated save percentage, I can then deduce a goals-against-per-game.

I then rank the estimated number of goals scored based on where that total would land in the standings from the previous season. I do the same with the goals-against per game.

Using the goals-against and save percentage, I see what teams had that same number and record their winning percentage.

I apply that winning percentage to the number of games each goalie is estimated to play and calculate the points that would result.

Taking that information, the rank of goals scored and goals against and finally making small adjustments for depth as it relates to the potential for injury, I arrive at an estimate of where I believe the team is most likely to finish the season. I emphasize most likely because these are meant to be estimates and approximations only and there are myriad unforeseen circumstances that can affect a team’s performance.

I typically take a wide assortment of statistical categories into account when researching each team, including shooting percentage, points per game, games played, TOI, IPP, PDO and more recently DFF% relative to competition (thanks to Woodguy and GMoney for their work in establishing this metric and their help in explaining to me), and expected goals for and against.

We’ll begin with the majority of the data I’ve collected, courtesy of Behind the Net, Corsica.Hockey, BecauseOilers, Oilers Nerd Alert, EliteProspects and NHL.com. We proceed alphabetically by the player’s last name.

Here are my projections for the Canucks roster by way of point estimates based on a points-per-game scale.

We’ll go through the roster one-by-one, beginning with the forwards and proceeding alphabetically by last name.

Sven Baertschi – Often deployed as a depth scoring option, Baertschi actually seems to do relatively well against the elite level competition and his IPP (the percentage of points scored during his time on the ice which is directly attributable to his actions) is quite high, all things considered.

Where Baertschi lacks, and has for some time now since breaking into the league, is in his two-way development. He cannot yet be trusted to play a defensively responsible game so as to be rewarded with more ice time while simultaneously struggling to score enough to offset those defensive lapses. It is something of a catch-22 as Baertschi’s calling card to the NHL was always going to be offence but he cannot yet garner enough ice time to collect the points needed to justify his defensive lapses precisely because of those lapses.

He is ideally suited in the depth winger role, yet curiously enough it is against that same tier of competition that he appears to struggle the most. I believe this is partially the responsibility of the player but that it is also in part due to some of the deployment decisions by Desjardins.

Be that as it may, I have estimated that he plays 69 games this season, totalling 31 points.

Alex Burrows – Through a miserable season last year and amidst rumours that the Canucks had attempted to move him at some point, Burrows managed to play a very effective game against middle tier competition. His IPP does not recommend him as a first-line option, nor do his DFF or expected goals numbers, which identify his play last season as better suited to a 2nd line role, ideally 3rd line had the Canucks enough depth at center to deploy him as such.

I have projected that Burrows plays 70 games this season, recording 27 points.

Derek Dorsett – A poor hockey player woefully mismanaged might be a little harsh in describing Dorsett’s last season, but it isn’t too far off.

Dorsett struggled against all three levels of competition last season, his IPP rate has him contributing offense at about the same level as a coin toss, his shooting percentage is low, he gives up far too much in the way of offense against and possession, he plays a greater share of minutes than Baertschi and Burrows, and all the while his PDO suggests that this is only a little less than what can reasonably be expected from the player.

The maladies that currently afflict the Canucks, and seem likely to for the immediate future, can be illustrated to some degree in the circumstances surrounding Derek Dorsett.

I have estimated that he plays 71 games this season, totalling 16 points.

Loui Eriksson – Likely signed because of his connection to the Bruins’ during Benning’s time there, this deft move is lamentably notable because it represents something of an exception to the majority of other decisions recently.

Eriksson is a very good player who can effectively challenge and succeed against all levels of competition, contributing positively to offensive chances and possession from the wing. Ideally he would be deployed in a 2nd line role, but with the Canucks’ current roster situation he ought to see time playing on the 1st line with the Sedins, even if it means moving to his off-wing.

The other options are playing him with Sutter or Horvat, which will likely reduce his offensive contributions.

I’ve projected Eriksson to play 80 games, with 66 points.

Emerson Etem – Another player struggling with his deployment assignments and what is being asked of him. He plays a greater share of minutes than Baertschi or Burrows, predominantly against the bottom and middle tier of competition, where he appears to be passable though not necessarily worth retaining should other options present themselves.

He cannot contribute extensively to the offense and does not do well against soft competition. He is not a good possession player and gives up too much offense against.

I’m estimating he plays 50 games, 15 points.

Brendan Gaunce – A young player growing into the league on a bad team is nothing new to an Oilers fan such as myself. Gaunce has some good things to take into the season. Based off of a small, 20 game sample size, he appeared to do quite well against the elite and bottom tiers of competition he faced and showed well in both the DFF and expected goals metrics.

His shooting percentage was almost non-existent and that is something to watch closely once he has amassed a larger number of NHL games.

Deployment is key, as it always is with young players, and will be another thing to keep an eye on this year.

I’m anticipating 46 games, 16 points.

Mikael Granlund – Granlund is another player like Baertschi whose calling card is ostensibly offense and yet he is undermined by it.

His IPP is the lowest of any forward of the three teams (also Calgary and Edmonton) I examined in this year’s exercise. At 35.3 it is closer to that of a defenseman than a forward, especially one with a reputation of creating offense. 

He is primarily deployed against bottom tier opposition and yet he cedes to them offensive chances and potential goals at an alarming rate.

I suspect there is a small part of this that is the result of teammate matchups and roster balance, however Granlund’s abilities are so far underwater at this stage that he cannot escape blame for his struggles here.

I’ve estimated he plays 65 games and scores 20 points.

Jannik Hansen – A 2nd line winger who finds himself playing 1st line minutes and competition can be a poor mix, but to Hansen’s credit he appears to be managing. Some of that is due to his linemates in the Sedins.

Hansen’s IPP demonstrates that he contributes offensively at a good rate, to the extent that I would say he knows enough to be able to play effectively with the Sedins (both their IPPs are higher, suggesting that can get along well enough without Hansen, but I have doubts that cuts both ways).

His DFF and expected goals suggest that the sooner Hansen can convert his ice time primarily to middle tier competition the better, provided he does not suffer too significant a drop off in quality of linemates.

I’ve projected he plays 73 games, recording 28 points.

Bo Horvat – When it came time to decide who was going to play behind the Sedins and Brandon Sutter went down to injury, Bo Horvat drew the short straw.

He played more on average than Burrows, Baertschi, Dorsett, Gaunce, Granlund, Sutter, and Virtanen. The majority of that he spent against elite level competition, a match up that surprisingly he tended to win. He struggled against the larger demographic of middle tier competition, and mightily so, to the extent that he lost both the expected goals and possession battles.

At this stage of his development he would be ideally situated playing middle tier competition, sheltered by capable, experienced veterans, and having one or two skilled veterans on his wings to assist.

I’ve estimated he will play 75 games, scoring 35 points.

Jayson Megna – Something of an off-the-map acquisition, Megna doesn’t have a lot of playing time to draw upon from last season, as evidenced by his peculiarly high IPP. His DFF suggests he was primarily deployed as a 2nd line forward, and while the numbers there aren’t terribly inspiring, his expected goals differential is remarkably high.

All of these may be the result of too-small a sample a size in the end and as such my projections are going to be a bit vague.

I generally err on the side of caution, regardless of the player, so I’ve projected Megna to play 36 games, registering 7 points.

Daniel Sedin – One half of the straw that stirs the drink that is the Canucks, Daniel Sedin is (still) an astonishingly effective player against all levels of competition.

His IPP is perhaps not as high as one might expect, given how he and Henrik have traditionally dominated entire shifts in the offensive zone, but his DFF is strong across the board and he is one of only four Canucks to record a positive expected goals differential from last season.

I’m projecting him to play 79 games, 65 points.

Henrik Sedin – You can more or less copy and paste everything from Daniel’s entry into that of Henrik’s. The biggest differences appear to be that Henrik more fully dominated middle tier competition in the DFF scale and registered negatively in the expected goals category, perhaps reflecting their time apart this past season due to injury.

Virtually identical projections to his brother, 79 games, 66 points.

Brandon Sutter – His season shortened due to significant injury, Sutter’s numbers still do not reflect the player that it appears Benning believes him to be. He does not drive the offense according to his IPP, though his possession numbers are positive on a nearly-entirely negative team and that is something to build off of.

His DFF numbers suggest that his deployment is not entirely aligned to his strengths, that is to say, he is sent out as a player who can go toe-to-toe against the elite and middle tiers of competition, the latter of which he appears at least capable, but his real strength lies in facing the bottom tier of competition. He is one of the few Canucks players to come away with a positive expected goals differential, so perhaps there is still yet some greater potential in this player, but likely only to the extent that he may graduate to becoming a passable second line center.

I’m estimating he will play 81 games, registering 28 points.

Jake Virtanen – He may be the Canucks’ best young player on their current roster with regards to his potential (Hansen is by way of current ability, but he is quickly moving out of that group and into established NHLer).

Over a decent early sample size of 55 games, Virtanen looked good by way of his matchups via the DFF numbers and his expected goals differential, keeping in mind he played as a rookie.

His skill set is a good match for the Sedins and it is unfortunate for him that he could get a chance to play with them only so late in their careers.

I’m projecting Virtanen plays 60 games, netting 13 points.

Joseph Labate –  8 games, 1 point.

Michael Chaput – 8 games, 1 point.

Alexandre Grenier – 4 games, 0 points.

Anton Rodin –  Coming over as an older player, and playing the wing position required by the Sedins, along with a proven history of putting up points playing against men gives Rodin a reasonable chance to find some measure of success this season. I’m estimating 30 games, 13 points.

Alex Biega – Injuries on the blueline last season forced Desjardins to play some defenders in positions they may not yet have been able to necessarily handle. Biega was among the most-sheltered of the Canucks’ D last season but registered the worst expected goals differential of any defenseman in the three teams I examined with the exception of one: Luca Sbisa.

Biega’s ice time was more or less evenly split amongst the three tiers of opposition, his only positive result being against the middle tier.

He does not contribute offensively, nor does he appear to be particularly developed in his defensive game, although I suspect some of this could well be the result of his partner. Nevertheless, Biega’s PDO coupled with his overall results from last season suggest that his ideal role is as a 3rd pairing shutdown defender, and a very replaceable one at that.

I’ve projected him to play 55 games, recording 7 points.

Alex Edler – One half of the Canucks’ defense corps pillars (Tanev being the other), Edler is rare amongst all the blueliners examined within this exercise in that he had a relatively high IPP for a defenseman and scores well in the DFF metric across the range of competition tiers. His expected goals differential is in the negative but this is more likely the result of being one of the only true NHL-worthy defensive options on a roster that is notable for a lack of balance and depth.

I’ve projected him to play 67 games this season, netting 30 points.

Erik Gudbranson – Like the Brandon Sutter acquisition, the concern I have with this player isn’t necessarily his abilities but rather the perception (or mis-perceptions, to be more accurate) of those abilities and his potential by management.

With the Canucks losing Dan Hamhuis this off-season, acquiring Gudbranson does, in some way, offset that loss, although an examination of the roles of the two players last season shows they were quite different.

Hamhuis and Gudbranson have both been used in a shutdown role, however Hamhuis contributed to a positive possession game consistently whereas Gudbranson was only a positive Corsi player once in his last four season, and that was against weak competition.

This isn’t to say that Gudbranson is a bad player, just that his ceiling is likely as a 2nd pairing shutdown defender if paired with a skilled, puck-moving partner who can play the more dominant role within that pair.

The problem arises if the Canucks keep Edler and Tanev together (adviseable) and that they lack that 3rd defenseman in the rotation of four to pair with Gudbranson. The most likely result is that they ask him to do things he is likely not capable of and both player and team suffer as a result.

Trust me, I’ve seen this before.

That being said, I’ve projected Gudbranson to play 69 games this season scoring 9 points.

Ben Hutton – Hutton enjoyed a nice rookie season, recording 25 points during an otherwise abysmal team campaign. His boxcar splits though raise an eyebrow – 25 points, 24 of them assists. This suggests that luck played a part in the result and as such caution over expectations is recommended.

Hutton’s ice time was spread relatively evenly across the competition tiers, and outside of some struggles with the middle tier (something we should expect with a rookie defender) he shows rather well. His expected goals differential is very poor, but nowhere near as catastrophic as that of Biega, Sbisa or Tryamkin, instead closer to that of Gudbranson, a relatively established veteran.

Despite my warnings about luck playing a part in his offense, I have estimated Hutton to play 75 games and register 23 points as I expect he will receive a healthy dose of power play time.

Luca Sbisa – Easily the most roundly despised Canuck on the roster right now, unless they ever get around to putting Jim Benning on the ice for a practice. Sbisa has a very limited range of skills and those appear to be both decaying and miscast, compounding the agony not only for the player (who surely knows that he is struggling) but also for the team itself.

How does Sbisa rate? His DFF is atrocious across the board with the exception of his time against elite-level competition. Does this mean the Canucks should specialize in sending him out against the best of the best? No. I believe this result is something of an outlier based on the Laurentian Abyssal depths of his results against middle and bottom tier competition.

Now for the bad news.

Sbisa has a relatively similar IPP rate as Gudbranson, his DFF and expected goals differentials are approximately half that of the former Panther – meaning Gudbranson may not be that great an improvement on Sbisa. Also, Sbisa’s PDO was a very high 1030, meaning he actually managed to overachieve to some extent last season.

Perhaps when the page turns on Benning’s time with the organization, Sbisa will, unfortunately, becomes something of a standard bearer for the era.

I’ve projected Sbisa plays 65 games, netting 12 points.

Chris Tanev – The other bright light for the Canucks on defense, Tanev provides a nice counterpoint to Edler’s puck-moving skills. Arguably the better defender of the two, Tanev also rates very well across the board in the DFF metric, showing himself to be a more-than-capable top-pairing defender against elite, middle and bottom tier competition.

Despite his reputation of being more of a defensive option, his IPP share is higher than that of Edler, perhaps giving further proof to the old adage that offense flows from good defense.

Perhaps what stands out the most for Tanev in this examination is that on a team that was roundly outperformed in the possession metrics and generally slaughtered in the are of expected goals differential, Tanev managed to be a positive possession player and his differential is a hair’s breadth away from break-even.

The Canucks have two top-pairing defensemen and Tanev is arguably the better of the two.

I’ve projected him to play 68 games and record 16 points.

Nikita Tryamkin – A large defender who put in a dress-rehearsal on a rookie campaign late last season, Tryamkin’s sample size for this exercise is going to cause some issues. However, what we can say is that Desjardins’ deployment of him – 39% of his ice time was against elite-level competition – is unadvisable.

His overall results bear some similarity to Luca Sbisa in terms of his PDO, expected goals differential and average time on ice. This does not, in and of itself, condemn the young player, but rather suggests that the Canucks proceed gradually with his development and keep expectations modest.

I’m estimating he plays 40 games, posting 4 points.

Phil Larsen – Former Oiler and recent acquisition, Larsen was added to play a role on the powerplay and add some offensive punch. His defensive game is not a strong point and the Canucks do not boast a defense corps that can afford to relinquish much in that direction no matter how desperate their pursuit of offense.

I’ve estimated he plays 42 games, netting 11 points.

Troy Stecher – A coveted FA signing, Stecher steps into a sparse prospect cupboard on the back end. An older player, he is likely closer to the NHL than the other names on this list.

8 games, 2 points.

MacKenzie Stewart – 1 game, 0 points, unless Desjardins suddenly sees fit to dress the hockey equivalent of a gravel crusher for a longer stretch.

Andrey Pedan – Can fill in spot duty as heavy demolition equipment if Stewart goes down to long-term injury.

1 game, 0 points.

Jordan Subban – Some offensive punch, but needing time to develop his defensive game. Being undersized makes this an absolute priority. He has some nice potential, but is a long-term project.

1 game, 0 points.

My estimation of goaltending is almost entirely based on precedent, which from everything I’ve read on the subject, is almost entirely insufficient to try and predict goaltending.

So it would appear that we have something of a conundrum.

Be that as it may, the best we can do to predict tomorrow is to base it off of today (or in this case, what occurred last season).

To that end, I’ve worked out the following:

Ryan Miller – 47 games as starter, 0.915 sv% with a 2.8 GAA based on seeing an expected 33.5 shots per game.

Jacob Markstrom – 35 games as starter, 0.914 sv% with a 2.8 GAA based on seeing an expected 32.5 shots per game.

The Canucks lost a few key roster players last season to injury, but when weighted against the league average, were far from the worst off. Yet they suffered one of the worst seasons in recent memory.

They received excellent goaltending from the tandem of Miller and Markstrom and yet the remainder of the roster was unable to capitalize.

They have replaced Radim Vrbata and Chris Higgins with Jayson Megna and Loui Eriksson, filled the void left by Dan Hamhuis with Erik Gudbranson.

Beneath it all the fundamental issues with the Canucks’ roster remains. They are not nearly skilled enough throughout the roster, those players who are skilled are at either end of their career trajectories, and the defense is terribly undermanned despite the presence of two stalwart players in Edler and Tanev.

The goaltending may continue apace from last season, but it is just as likely that it will not and therefore it is unwise to place too much confidence in that area.

The Canucks are, by these estimates, slated to score 213 goals, or about 2.6 goals a game. That rate of offense would place them 21st overall in the league last season.

On the other hand, the Canucks are estimated to surrender 2.8 goals per game, at number which would, place them about 20th overall in the league last season.

Overall, I worked out an estimated winning % for the Canucks of 0.463, which would work out to about 76 points. Once we add in a number of overtime losses, I’d expect the Canucks to finish the season 7th in the division, 13th in the conference, perhaps as high as 24th overall with 80 points.