A couple of weeks ago, I was asked to take a look at the San Jose Sharks and see if there was something obvious that could point to why they stink this year. I mean really, really stink. The Sharks used to be a really competitive team; and not all that long ago, either. What makes their demise all the more confusing is that they have become bottom-dwellers in spite of some pretty significant talent on the roster. All teams go through ups and downs but this seems like a hole they simply can’t get out of. Instead of looking at traditional stats, I decided to take a look at the roster composition between 2015 and 2023.
Why Roster Composition Matters
If you’ve listened to my podcast, you’ve probably heard me make comparisons between hockey teams and the everyday workplace. These players’ version of the office is the ice rink and office politics and dynamics exist in every sphere of work from the local McDonald’s to the SAP Center at San Jose. Teams that have been good for long stretches can often attribute it to the culture established by major leaders early on in their careers. Think of the Penguin’s core of Crosby, Malkin, and Letang, or Chicago’s Toews and Kane. When these players establish a particular culture, it can have a long-lasting impact on the team, even when the team roster changes over time. The Boston Bruins will put the strength of the culture established by Zdeno Chara to the test the day Brad Marchand hangs up his skates for good.
A Culture Change?
The Sharks haven’t made the playoffs since 2019. Between 2015 and 2019, the team was averaging a 0.607 points percentage per season and averaging a 0.552 points percentage through the first 17 games of the season. They easily made the playoffs in those years, entering as divisional leaders.
I fully expected the decline to be slow, steady, and hard to pinpoint. Instead, I was surprised by what I found: not a death by a thousand cuts, but an immediate nose-dive into awfulness in the fall of 2019.
Between the fall of 2019 through the start of the 2023 season, the Sharks have averaged an embarrassing 0.386 points percentage and averaged a 0.382 points percentage at the 17th game mark. The difference in points percentage at game 17 in 2018 (0.559) and the same game in 2019 (0.382) is astonishing.
There was a brief improvement in 2020-2022, but I suspect the changes to the way we played games due to the pandemic might have helped the team.
What Happened to the Players?
One of the most obvious trends I noticed in my evaluation of the roster was the fact that the team became a revolving door in 2019. While I did not dig deeper into the circumstances (financial or otherwise) around 2019, what I started to see was the lack of tenure by most players. With a constantly changing roster, chemistry and stability can’t form in earnest. Essentially, the Sharks look to be operating as a transfer station rather than a destination for most players.
Between the fall of 2016 and spring of 2019, the team brought in an average of 6 new players per year. Since 2019, they have averaged 15 new players per season. This is an insane amount of change. As a comparison, I chose three teams at random to look at the roster sizes from 2015-present: the New York Rangers, the Los Angeles Kings, and the Columbus Blue Jackets.
NYR: Average of 29 players on the roster per year. After their longest non-playoff stretch of two seasons and the season they made the playoffs again (2017-2020), they averaged 32 players per season. During this time they averaged 8 new players per season and the roster comprised an average of 27% new players.
LAK: Average of 30 players on the roster per year. After their longest non-playoff stretch of three seasons and the season they made the playoffs again (2018-2022), they averaged 33 players per season. During this time they averaged 9 new players per season and the roster comprised an average of 26% new players.
CBJ: Average of 32 players on the roster per year. After their longest non-playoff stretch of four seasons (2020-present), they averaged 33 players per season. During this time, they have also averaged a large number of new players per year at 13. This is due to the fact that the start of the 2021-2022 season saw a roster comprised of 47% new players.
SJS: Average of 31 players on the roster per year. Since 2019, they have averaged 35 players on the roster per season. Even though San Jose is cycling through the average number of players per season over the last nine years, they are cycling through a much higher number of new players since their skid.
Remember lesson number one of analytics: correlation does not equal causation. But it is very interesting that high turnover seems to be a contributing factor to the lack of success.
Player Cohort Tenures
In order to dive a little deeper into this player turnover trend, I grouped players into cohorts, based on the year they joined the team. The exception to this rule is the first cohort since there were players in this group who joined prior to 2015, but they are my baseline. I did this in order to see how long different cohorts stayed with the organization and if there was a difference between the players who joined the team between 2015 and 2019 compared to those who joined after 2019.
The results were very interesting. I think this is one way to measure the qualitative idea of “workplace culture”. If the workplace culture is great, employees will stay around for a while. If it’s not, they jump ship or get fired pretty quickly.
Take a look at the chart below:
This shows the number of players from each seasonal cohort and how long the cohort has had representation on the team since they started. I’m surprised (but probably shouldn’t be) to see players who were on the team nine years ago are still on the team today, while there are some cohorts (2017 and 2018) who no longer have players on the team at all.
The 2021 cohort has the highest rate of decline in players from year one to year two: 72% of players who joined the team in 2021 were gone by 2022. Another 60% of the remaining players left before the 2023 season. There were 18 new players to the team in 2021 and by the start of this season, only two remain.
Another way to look at the longevity of cohorts, especially pre- or post-2019, is with this stacked bar chart. This shows the percentage cohort makeup of the roster each season.
Cleaning House and Re-Building
I think it can be easy to say “just clean house” or work on re-building. But if the team management doesn’t do it thoughtfully or with purpose, the team is going to continue to suck, plain and simple. You’ve seen the strategic hire of the solid veteran who probably doesn’t put up the same points, but can be a mentor to younger players. Chicago did this by bringing in Nick Foligno and Taylor Hall for Connor Bedard. Buffalo also did this with Kyle Okposo, to serve as that veteran mentor who can help to build a positive work culture. For the Sharks, their big-name mentors started to leave in 2019 and management didn’t bring in good enough replacements for the culture. Until the Sharks GM and President decide to commit to the team culture and invest in solid leadership, they will continue to live at the very bottom of the NHL.