Andrei Vasilevskiy is having a rough year. By now, you know that. Last week, we tried to figure out if the team’s defensive play might be part of the issue. We didn’t find any evidence to support that idea. If anything, we found evidence that aside from last season, Vasilevskiy hasn’t been great to this point in his career. He’s been well below average in goals saved above expected in three of the four seasons where he’s had a starter’s workload.
Today, we’re going to explore another possible explanation for his struggles. This is a much more nebulous concept to try to measure but we’re going to do our best to look into Lightning goalie coach Frantz Jean’s potential impact on Vasilevskiy’s development.
The approach we’re going to take largely replicates CAustin’s 2013 article in which she looked at Lightning goalie performance while coached by Jean and while not coached by Jean. At the time, she found that every goalie performed worse under Jean than under different coaches. That group included five different netminders.
The measure she used for comparison was primarily save percentage. We have the advantage of some new ways of measuring goalie performance that weren’t available in 2013. We’re going to primarily use varying forms of goals saved above expected (GSAx). One of the questions CAustin addressed in the article was the idea that the defense was partially to blame. We can make an effort to control for this explicitly by using GSAx.
To start, let’s replicate exactly what CAustin did but using goals saved above expected per 100 unblocked shots. All data for this article comes from Evolving Hockey and includes all situations. The following plot compares all the goalies who have played in Tampa under Jean and for at least one other team during the years where we have expected goal data.
Well, that’s not great. Nothing much has changed since CAustin’s original article. Of the 11 goalies in our data set, only Louis Domingue has improved while playing for the Lightning. Everyone else has gotten varying degrees of worse. Vasilevskiy isn’t included here because he’s only ever played for the Lightning organization.
Trying to assess coaching by looking at player performance is a dicey proposition but 10 of 11 players getting worse seems like a signal that something is happening. Initially, I was concerned that maybe something weird was happening in the way shots are recorded in Amalie Arena that was causing this pattern. But if we apply the same process to the goalies who played in Tampa before Jean in years where we have GSAx data, four of the six improved. So while that’s not definitive, it does suggest that we’re not simply measuring a data anomaly.
The following plot shows all eleven goalies again. But this time we’re looking at how many shots they faced under Jean and how much worse they were per 100 unblocked shots.
We do have a cluster of goalies who faced a relatively low number of shots that are susceptible to variance in a small number of observations but even among those who faced a high number of shots, they all got worse. Ben Bishop, Mike Smith, Dwayne Roloson, Mathieu Garon, and Anders Lindback all performed worse while being coached by Jean.
One more way we can try to visualize this is to look at how many more goals each goalie allowed in Tampa under Jean than they would have allowed based on their career paces away from the Lightning.
In total, the eleven goalies included in our data set allowed 228 goals more than they would have if they’d just played in Tampa the way they did over the rest of their careers. That’s quite something.
All of this begs the question of what might Andrei Vasilevskiy look like if he wasn’t playing in Tampa. The following chart applies an adjustment to his results based on how much worse goalies have been while on the Lightning over the last ten seasons. This is a very rough method but to summarize, Vasilevskiy goes from being well below average to one of the best goalies in the league.
As stated at the outset, measuring coaching impact by proxy through player results is extremely difficult. And for goalies, a position prone to variance, even more so. Nothing in this article is hard evidence of anything. A myriad of confounding factors could be playing into the results that we see.
But here’s what we know. Nearly every goalie who has played for Frantz Jean has performed worse for him than they have elsewhere. That difference doesn’t appear to be the result of any underlying data issue. That difference spans multiple head coaches, defensive systems, skater groups, and goalie styles.
The Lightning just invested $9.5 million per season for the next eight years into their 25-year-old goaltender. They absolutely cannot afford for him to be a below average player at that cap hit. It will sink the team’s hopes of sustaining their recent success and reduce them to hoping for a hot run during the playoffs in order to have a chance at a Stanley Cup.
And based on talent, there’s no reason to think Vasilevskiy should be a below average goalie. He was a first round draft pick and flashes every tool in the tool box. For a goalie coach, that should be a dream player to work with.
I think all of that is enough to spark a more thorough examination of Jean by the front office. This is as far as I can go with this idea using public data. And for me, it’s far enough to raise concerns.
Coaching decisions need to be made based on a full set of information. I have a very limited portion of that set. The Lightning front office has better data and they also have an understanding of the interpersonal part of Jean’s impact on Vasilevskiy, which is impossible to assess from the outside.
But from where I sit, which is very much on the outside of the front office, I can’t look at this information without thinking this should be a priority area for further internal research. Goalies have a greater impact on the outcome of a game than any other player on the ice. The Lightning paid Vasilevskiy like he’s going to be one of the best at this position in the NHL for most of the next decade. Any indication of something suppressing his performance needs to be investigated thoroughly. And I think we see enough here to warrant that investigation.
If Jean is truly having this type of impact on goalies, or even if the picture isn’t quite as dire with more information but he’s still a net negative, the Lightning have to make a change in that role. And that change should be to bring in whoever is considered the ideal fit. Whether that means bringing David Alexander who coached Vasilevskiy in Syracuse back from the Blues or someone else, the front office and ownership need to be willing to do whatever it takes to get Vasilevskiy to his ceiling.
They’ve already made a huge investment in the player and there’s no salary cap on coaches, so they have to do what it takes to put one of the team’s highest paid players in a position to succeed. For Jeff Vinik, that might mean breaking out the checkbook to make a big offer to poach a top goalie coach. He’s never been shy about spending to make the team better. This might be another time for him to show that.