All statistics shown are from hockeyreference.com. Articles referenced are hyperlinked on the original author’s name.
Over the past several seasons goaltending workload has become a hot topic among hockey fans. How many games should a starting goaltender play? Is there any correlation between a dip in performance and more significant playing time? How important is it to manage a goaltender’s workload?
Recently, Joe Smith of the Tampa Bay Times wrote a piece that stated pretty plainly that Andrei Vasilevskiy is “exhausted” from his first season of shouldering a starting goaltender’s schedule. The inspired me to take a look at Vasilevskiy’s games played over the course of the season. Last season, Saima took a similar look at Vasy’s workload. But what has changed since Vasy became a starter?
First, let’s take a look at Vasilevskiy’s performance month by month to see how he has played.
Vasilevskiy’s average save percentage for October was 0.923; he faced an average of 33 shots per game. Aside from three sub 0.900 games, Vasilevskiy was good in net for Tampa Bay. Nothing out of the ordinary here.
In November, Vasilevsky started to settle into his role and dominated with an average save percentage of 0.939, while still facing an average of 33 shots a game. Around this time, Vasilevskiy started to gain respect around the league for his stellar play, but this still wasn’t his best month.
December was the month Vasilevskiy (and Tampa Bay) shined. Vasilevskiy posted an average save percentage of 0.945 (!?) and faced an average of 25 shots per game. It was at this point he thrust himself towards the front of the Vezina conversation and into the collective mindset of many in the NHL. There was still speculation by some, however, that Vasilevskiy would come back down to earth from this level of play.
January was "coming back down to earth" for the Lightning and their netminder. Vasilevskiy posted an average save percentage of 0.909 and the average number of shots he faced per game went back to 33. It was during this month where the Lightning lost Victor Hedman for three weeks due to injury and started to become leaky on the defensive side.
Additionally, Vasilevskiy’s workload, defensive personnel, and the defensive system became talking points due to the inconsistency the Lightning displayed during January. Many thought Tampa Bay would bounce back in the following month, but there was some legitimate worry about Vasilevskiy’s performance moving forward.
Things stabilized a bit more in February, and it seemed as though Vasilevskiy had too. He posted a save percentage north of 0.910 seven times. However, he also went below 0.900 three times and hovered around it twice. This, in turn, pushed his average save percentage down to a pedestrian 0.910 while facing 34 shots a game. At this juncture, questions about the defensive system, personnel, and over-reliance on Vasilevskiy shutting teams down reached its zenith.
At the end of February (I’m not counting any of the March games just to keep this relatively static), Vasilevskiy had an average save percentage of 0.926 (it’s 0.9258, but I rounded up) while facing 31 shots a game. That’s still an excellent season for any goaltender in the NHL and is something that shouldn’t be dismissed.
Workload and Worry
Now that we know how Vasilevskiy has performed in the first five months of the season let’s get into the workload aspect of the season. Currently, Vasilevskiy has started 55 games and appeared in 56. Tampa Bay has played 67 games so far. He’s on pace to appear/start in 67 games this season. Is this something Lightning fans should be worried about?
Charged Up guest Catherine Silverman referred me to an article written by Murat Ates for the Athletic (dated February 11th, 2018) about goaltending workloads. Here are some excerpts from the piece (but I highly encourage you all to read it if you have a subscription).
Catherine Silverman on goaltending workload:
“Based on what we’ve seen over the last few years, overall shot volume going up has resulted in game load volume needing to go down a bit,” she told me via e-mail, “But the style of play of goaltenders has also meant that they’re harder on their bodies than the goalies of old.”
Ates follows up with the following:
Citing her research into the workload of Chicago’s Corey Crawford, Silverman added, “I think that 55-65 games per regular season is the sweet spot in order to prevent breakdown and fatigue in the postseason. There are so many extra factors, though, that it really is hard to tell.”
Ates also goes into detail about the goaltender he wrote the article about, Connor Hellebuyck, and uses his conversations with Hellebuyck and coach Paul Maurice about how they all approach his workload. The critical thing through all of it is how they communicate with each other. They’re honest with Hellebuyck’s performance and how his body is feeling. This struck me as something that all teams probably do—it’s just a matter of how teams go about managing the workload (Winnipeg has handled it quite well).
Another interesting bit of information Ates brought up was a study conducted by Philip Myrland in 2008. I won’t go into too much detail about it, but the basic premise was to see how heavily used goaltenders compared month-to-month and into the playoffs. It’s an intriguing read and one that could help us gauge how Vasilevskiy could play in the coming months.
Excerpt of Myrland’s idea behind the study:
The justification for the importance placed on games played seems to be that it is more difficult to play more games. I remain unconvinced that playing additional games is significantly more difficult. I think that virtually every goalie in the NHL is capable of handling the workload. If they weren’t, then they would have failed on some lower level (junior, college, minor-league) when they were asked to play 70-75 games a season.
Myrland then goes to break down goaltenders’ performance month-by-month (remember, this is the 2007-2008 season) and comes to this conclusion:
What do the data show us? That the goalies tended to start slowly in October, and after that kept up a similar level of play the rest of the season. Their best month was December while their worst month was January, but there was only a .004 difference in save percentage and a 0.10 difference in GAA between the two. Even though these goalies played big minutes all throughout the season, they did not get worse as the season went on.
He adds that some goaltenders deviate from the norm, and he singles out Martin Brodeur as one of them, but his point remains throughout—games played, and goaltender performance does not correlate. Does this still stand true in today’s NHL?
Knowing this information can help us determine whether the Lightning is riding their goaltender a bit too much. Vasilevskiy is currently going to go over the 55-65 range that Silverman suggests in her interview with Ates, but if we are to believe what Myrland researched, then we shouldn’t have to worry about him. A slight caveat to this is that it is only over by two games. How impactful those two games are is up for debate, but the Lightning does appear to be hovering in the recommended range for starters.
We should look further than just this season, however, and the backup goaltending position is a unique one for the Lightning. Currently, the backup role has been split between Peter Budaj and Louis Domingue. Budaj has played in seven games with a GAA of 3.80 and a Save Percentage of 0.878. Those are horrid numbers for any goalie, and we’ve stated before in recaps and on our podcast that Budaj has struggled early in games.
Additionally, the team has not been stellar defensively in front of him, so, with that in mind we’d have to give Budaj some leniency on his numbers—it’s difficult to play every ten games or so.
Domingue’s played in seven games for the Lightning and has a GAA of 3.00 and a Save Percentage of 0.903. Domingue’s numbers a more respectable than Budaj’s, and he’s been sharper in the games he has played.
Budaj is under contract until the summer of 2018-2019 with a cap hit of $1.025M. Domingue is under contract until this summer (he is an RFA) with a cap hit of $1.050M. Before the season’s start, some of us at Raw Charge were a little skeptical of Budaj as the backup due to his age, play, and durability. Budaj resurrected his career while playing for the Los Angeles Kings (a team that inflates their goaltender's statistics due to their defensive structure) and was seen as a solid choice to back up Vasilevskiy for the reminder of last season.
Domingue was viewed as a riskier play for the Lightning due to his struggles with the Arizona Coyotes. While in Arizona, Domingue played seven games with a GAA of 4.00 and a Save Percentage of 0.856. Domingue’s improvement from his time in Arizona to Tampa is interesting. All of his numbers improved once he came to Tampa Bay and given his age and better performance than Budaj the Lightning should probably look to re-sign him during the offseason and move on from Budaj.
If Domingue could take 15-20 starts from Vasilevskiy in future seasons, then that puts him in the 62 game range—right in the recommended amount that Silverman states in Ates’s article. Protecting Vasilevskiy from the grind of a starter’s workload should be among the Lightning’s biggest priorities, 67 games might be a bit too much. However, if Tampa Bay can reduce them into the lower 60’s or even the high 50’s it should, in theory, allow Vasilevskiy to be fresher for the playoffs.
Additionally, we should take into account another goaltender that Jon Cooper rode heavily during his tenure with the organization—Ben Bishop. Bishop played three full seasons—and five seasons in total—with the Lightning. During those three full seasons, Bishop played 63, 62, and 61 games respectively. This is right in the wheelhouse of Silverman’s recommendation.
Conversely, we know the biggest issue with Bishop was health. He was injured in every playoff run the Lightning had, and there were some grumbles about his workload being the culprit. Bishop is an odd case, however, since he’s one of the tallest players in the NHL.
Tall players have always had injury problems (especially knee issues), and Bishop’s injuries weren’t what I would call “due to his workload.” He was given a starter's workload and posted a 0.924, 0.916, and 0.926 during the regular season—that’s good no matter which way you cut it. In the playoffs, he played 25 and 11 games in consecutive playoff appearances. His save percentage for those playoff runs? 0.921 and 0.939.
So, if the workload didn’t affect Bishop’s performance then the worry about Vasilevskiy’s workload affecting seems to be a little misplaced. Goaltenders are voodoo (I love this phrase) and there will always be ebbs and flows with their performance.
From what Silverman and Ates have researched we can safely assume that as long as Vasielvskiy does not break the 70 game threshold, then there should be little worry about his performance moving forward. However, given Vasilevskiy’s age, the Lightning would be wise to lock-in a strong backup to take 15-20 games (possibly more) to lighten the load on the young Russian. It never hurts to be too careful.
This is an inexact science, and Silverman even states there are many factors that affect a goaltender’s play. However, the premise of Ates’s article still applies to Vasilevskiy, to ensure he doesn’t become burnt out and is fresh for the postseason. Domingue provides Tampa Bay with a more consistent backup option than Budaj heading into the future (although we have a small sample size of both) and likely should be retained during the offseason.