Last season, the San Jose Sharks were content to trade chances all year, despite poor showings by their goaltenders. This strategy, whether intentional or not, worked because the team’s offense was one of the league’s best for much of the year.
Consider a heatmap of the team’s 5-on-5 unblocked shots and how likely they were relative to league average to be goals:
According to Evolving Hockey, the Sharks generated the fourth-highest rate of 5-on-5 expected goals and the fifth-highest rate of unblocked shots on offense. MoneyPuck’s flurry-adjusted expected goals model found that even after adjusting for a sequence of shots in quick succession, the Sharks still generated the league’s fifth-highest rate of expected goals on offense. No matter which model you use to measure offense, the results show the same fact: The Sharks were proficient chance generators at 5-on-5 last year.
On the power play, a similar story unfolded.
The Sharks capitalized on Joe Pavelski’s tipping abilities and to a lesser extent Tomas Hertl and Timo Meier’s capacity for turning passes near the goal line into opportunities. The team’s power play also used pre-shot movement, often looking for the likes of Logan Couture or Brent Burns at the top of the left faceoff circle.
According to Evolving Hockey, the Sharks’ 5-on-4 power play generated the second-highest rate of expected goals, shots on goal, and unblocked shots last year. MoneyPuck’s tabulations weren’t as kind, showing a Sharks 5-on-4 unit that took the league’s fourth-highest rate of unblocked shots and generated the eighth-highest rate of flurry-adjusted expected goals.
During the 2018-19 season, the team from San Jose delivered a top-10 offense last season at worst and one of the league’s premier scoring threats at best. We begin with last year’s team because projecting this year’s offense requires a bit of subtraction.
A few months ago, they had Joe Pavelski, Gustav Nyquist and Joonas Donskoi ...
To try to estimate the impact the loss of Joe Pavelski, Gustav Nyquist and Joonas Donskoi may have on the team’s offense, we’ll look at the last three season’s worth of forward impacts with Evolving Hockey.
Evolving Hockey’s regression model attempts to isolate individual player impact on six components of the game: expected goals for and against, shots for and against, and goals for and against. Between the beginning of the 2016-17 season and the end of the 2018-19 season, there were about 1650 individual forward seasons.
Last year, Joe Pavelski’s impact on offensive expected goals was in the 98th percentile among all of those forward seasons. That was a bit of an aberration for him, however. His work during the two preceding years were in the 69th and 78th percentile. Of the 607 forwards who played at least 100 even-strength minutes during the last three seasons, Pavelski’s combined impact on expected goals ranked 26th. So, while we could have expected some regression from the ex-captain, it’s pretty clear the team is missing the offensive impact of a top-line forward.
Nyquist’s 2018-19 impact on expected goals (using both his Detroit and San Jose minutes) was a 94th percentile performance. During the past three seasons, he contributed the 25th-best isolated expected goals generation. Donskoi was a 63rd-percentile contributor to expected goals last year. Between 2016 and 2019, his impact was 143rd of 607 forwards. That’s the equivalent of a borderline first/second-line forward.
Here is a look at some very rough math to estimate what the Sharks might be losing by waiving goodbye to these three players.
This chart uses each player’s three-season combined isolated impact on even-strength expected goals to estimate how many total expected goals he was responsible for during the 2018-19 season. Now, figuring out how the Sharks’ offense might suffer the losses of these three players isn’t as simple as doing the subtraction here. Instead, we can consider these figures to represent how many expected goals he might help the team generate relative to a league-average forward. If the guys expected to step into their skates, like Danil Yurtaykin, Jonny Brodzinski and Dylan Gambrell are less impactful than a league-average forward, these estimated figures become magnified.
Because hockey is very much a team sport (albeit one in which superstars are the strong link), we can look at line combinations to get a better sense than this oversimplified arithmetic of what the team may be losing. Luckily, the coaching staff has started to clarify the lineup, and we can examine what we may need out of the team’s new forward lines this season to recreate last year’s success.
The Sharks’ offense faltered down the stretch, and the 2018-19 numbers you see at the top encapsulate their regular and postseason. The forward lines also leverage data from both the regular and postseason. Because of limitations gathering just forward partner information, the core pair data is from just the regular season.
Now they have Yurtaykin, Brodzinski and Bergmann
The good news here is that the core pairings of Barclay Goodrow and Melker Karlsson, Joe Thornton and Marcus Sorensen, Hertl and Evander Kane, and Timo Meier and Logan Couture have a team-average offensive production in their range of outcomes. The bad news is that this isn’t a super accurate snapshot of the impact of just those pairs of players because it doesn’t account for who their third linemate was when it wasn’t the player in this table.
As of this writing, the forward lines look like:
Meier – Couture – Labanc
Kane – Hertl – Brodzinski
Sorensen – Thornton – Yurtaykin
Melker – Goodrow – Bergmann
During the past three seasons, Labanc’s impact on even-strength expected goals has been a 42nd-percentile effort, a far cry from Pavelski and Nyquist’s top line abilities. While Meier and Couture are a formidable pair, and Labanc has had a strong impact on shot generation, the Sharks top line may struggle to create the type of offense it did last season.
Brodzinski also chipped in with a 40th-percentile expected goals impact over his brief NHL career. He’ll struggle to replace the individual production of any of Nyquist, Pavelski or Donskoi, all of whom spent time on the young center’s other wing last season. As we can see when we add Pavelski to the mix, however, sometimes line chemistry is more than the sum of each player’s individual impact on the game.
Yurtaykin’s creativity has been a bit of a revelation this preseason. While he doesn’t possess much of a shot, his ability to create through traffic and find teammates in the offensive zone should give the Sharks middle-six forwards an element they lacked last year. It’s unclear who on this current third line will be responsible for finishing the chances the team generates, but Yurtaykin should at least help the trio avoid cycling and sending the puck back to the points for offense.
It’s a shame the Sharks coaching staff prefers defense over offense in its fourth line. Antti Suomela lent the team an expected goals impact commensurate with that of a second-line forward, but he appears destined for AHL or press box time to start the season. As impressive as Bergmann has been in September, he’ll struggle to create any offense with a duo whose individual expected goals impacts haven’t exceeded those of a third liner.
At even-strength and 5-on-5 at least, the Sharks will struggle to replicate the offense that was one of the league’s most fearsome at the beginning of the 2018-19 season. Unless some of these new forward lines can create an impact that’s greater than the sum of its parts, the team’s offense is essentially replacing the impact of three, top-six forwards with that of a bunch of mystery boxes.
We should also note that a team’s offense — especially that of a team built like the Sharks — begins with its blueline. A healthy Erik Karlsson in transition and Brent Burns on the ice for most of the game should help negate some of the impact of those three forwards leaving. And, while neither Dalton Prout or Tim Heed offers much in the way of offense individually, their impacts should dwarf those of Justin Braun, who was a veritable black hole once the puck entered the offensive zone.
The power play should be much of the same
The team’s impressive power play returns many of the same players. With what appears to be close to the full NHL roster, the Sharks lined up like this against Calgary:
It should be noted that, although commenters league-wide loved to harp on Joe Pavelski’s puck-tipping ability, the team should be able to replace his impact on the power play’s shot generation. Last season, according to Evolving Hockey, Pavelski’s impact on the team’s power play expected goals was seventh on the team and sixth among Sharks forwards, both of which placed behind the impacts of Evander Kane.
It’s not as simple as that equation, but if the team’s preseason game against Calgary was suggestive of anything, it’s that the position Pavelski used to play and that Kane will assume will be more responsible for passing the puck back out to the wings and points than anything else. The coaching staff isn’t asking anyone else to try to tip Burns’ shots like their captain did all of last year.
As for the other two losses, Donskoi barely played on the power play last season, and when Nyquist did play, his impact was poor.
Even strength offense may be a liability
The loss of Pavelski, Donskoi and Nyquist, coupled with Burns’ declining offensive impact, is sure to make a dent in the team’s fortunes this season. It’s difficult to quantify the losses of those three players and estimate how the newcomers might make up for those shortcomings. We know the team is losing the offensive impact of three top-six forwards and replacing it with Kevin Labanc and a bunch of mystery boxes. That in itself should suggest to even the most casual onlooker that the team’s chance generation should fall back toward the middle of the league table.
To help make up some of that lost ground, players like Meier, Labanc and even Sorensen should see their ice time increase. Yurtaykin, Gambrell, Brodzinski and Bergmann will be called upon at first to bridge the rest of the gap, but don’t be surprised to see those players cycled in and out of the lineup depending on the team’s performance. Any sort of offensive impact near that of a league-average forward from that group should be considered a victory for the organization. For now, we’ll have to move forward under the assumption that the bar for offense this season will be set a bit lower than we’re used to seeing.