For all the work that NHL teams (and those who follow them, on a rhetorical basis) do to assemble the perfect 18-skater roster, it’s never been more obvious that nothing in this league matters as long as you don’t have a goaltender to stop the shots you concede. In 2017/18, playoff teams averaged a save percentage of over ten points more than those that missed. 11 of the top 12 starting goaltenders in save percentage this season made the playoffs, and the one who didn’t (Antti Raanta) is a great example of why tank-calibre teams don’t actually end up in last unless they find a way to dress someone bad, unprepared, or broken in between the pipes.

The Toronto Maple Leafs brought in Frederik Andersen in June 2016 to solve that problem, figuring that the Anaheim Ducks 1A was going to be a wall of consistency for the team over the next five or so seasons that followed. In a lot of ways, that’s turned out to be true: the Great Dane has posted above league-average numbers while posting monster minutes in each of his first two seasons with the Leafs.

But great numbers in mid-season don’t matter if you can’t hold them in the later part of the season, and that’s perhaps the concern with him. More specifically, the concern is how well you can expect him to hold up if his workload keeps looking like this:

The above are the ten goaltenders who have played in the most games over the past two years. Edmonton Oilers goaltender Cam Talbot has the minutes lead (by about 350) and games lead (by eight), but when Andersen’s playing, he’s more engaged, facing more shots than any NHL goaltender since the beginning of 2016/17, and almost two more per game than anyone else in this top ten.

That’s an extreme reliance on one guy, especially if you want him to compete deep into the spring. Today’s NHL has gotten close enough to parity that, even though the playoffs do give a little bit more of a strong-link advantage to top teams, you ultimately rely on a hot goaltender and a couple of hot sticks to push you through the rounds. This year’s finals feature two great goalies in Marc-Andre Fleury and Braden Holtby, but one can argue that they’ve been fortunate to keep their bodies relatively rested this year. Fleury had a mid-season injury that kept him to just 46 games, while Holtby’s up-and-down regular season and the emergence of Phillip Grubauer as a super-backup have limited him to 54.

Incredibly, if this final goes to 7 games, Holtby’s total of 79 appearances won’t be much of an anomaly from his previous seasons as Washington’s consensus starter. Between the regular season and playoffs, Holtby played 86 games in 2015, 78 in 2016, and 76 in 2017 – all without escaping the second round. Perhaps that’s one of the reasons he had the energy to stand on his head in Games 6 and 7 against Tampa Bay, and make a save-of-the-decade candidate to secure Washington’s win last night.

But back to Andersen, and how the Leafs rely on him in the regular season. Many feel that giving him 66 appearances last year was excessive, particularly with Curtis McElhinney having the year of his life as a backup. With that in mind, I took a look at a few different ways to gauge what might be too much for him and didn’t find a ton in the short term. As you can see above there doesn’t seem to be a rhyme or rhythm connecting how many days off Andersen gets in a stretch of 10 games, and how he performs. Sometimes he does great on no rest, sometimes he does terrible with lots of rest, sometimes, he’s fine in between, sometimes it flips. Or, put another way:

There really isn’t much of a connectable pattern here. Andersen’s best game-by-game results came when he had the most rest, but he was fine when he had next to none of it. His single-game results with multi-game workload considered came when he had an average or slightly above average amount of breathing room, but he struggled when he had a lot of it. Trying to find a patchwork solution to giving him the best regular season might not be what we should be aiming at here.

Besides, the end of the road is what everyone is most concerned about, and that’s where you see a pretty recognizable pattern in each year. After Andersen shakes off his usual sluggish October (0.847 in 2016/17, 0.896 in 2017/18), he tends to spend the midway point of the season flirting with Vezina discussion, before sliding off a cliff. This year, Andersen posted just a 0.896 save percentage between the beginning of March and the end of the playoffs; not ideal for your starter.

Perhaps more concerning was that his workload never really changed in that stretch. Andersen played in 12 of the final 16 games of the regular season, which paces out to roughly a 62 game year, still enough to be ranked 7th in the league in games played. Again, that seemed like a very curious call by Toronto’s staff given McElhinney’s performance and the fact that their playoff spot was extremely secure, but they ran with it.

A few weeks later, they end up with a playoff series where Andersen allowed three or more goals in five of seven games, including four games where his save percentage was below 0.900 – most notably his concession of three of five shots in Game 2, and allowing 6 of 35 in Game 7.

How much of that was luck of the draw, how much of that was a lack of defensive support, and how much of that was a 28-year-old goaltender that was simply out of gas? It’s hard to say, but it’s still not unfair to suggest that there should still be an effort to mitigate the latter. Andersen now has two consecutive seasons of heavy workload where he shows a distinct taper in the later months, and when you combine that with injury scares that he’s had in that stretch, it doesn’t seem wise to test him as much as they have already moving forward.

Toronto already employs some level of game management with Andersen, in the sense that they avoid playing him on the second half of back to backs. But one wonders if there’s more that can be done. For example, if you look at his regular season performances, you’ll notice that, especially after his October slump gets out of the way, Andersen’s worst games come after travel. 10 of Andersen’s 14 games below 0.885 from November on come either on the road or in his first game back home, as do 6 of 7 of his games below 0.850. You’re not likely to take too many at-home starts away from your starter, but maybe there’s something to the idea that of giving whoever backs up Andersen the last start of a multi-game road trip. Here’s how Toronto’s two goalies performed in that situation this year:Given that Andersen was following up these often-rough nights to end road trips with just as rough ones afterwards, it might be worth considering using this as the pocket to stretch out the backup’s workload. It also fits in line with Andersen’s career splits; he’s put up a save percentage over ten points higher at home than on the road in three of the last five seasons, likely due to having that extra bit of recovery time that travel tends to cut down, even if the days are still there.

Who is that backup? That’s a different debate altogether, and one we’ll get into in another article. But regardless of who it is, they should be taking on significantly more starts than the 16 afforded to Toronto’s relief squad of McElhinney (15) and Calvin Pickard (1) this year. There’s no excuse to not use a backup that is putting up positive results as a buffer for your starting goaltender, especially now that it appears that his late-season taper has the potential to be a trend, rather than just an anomaly.

---
Thanks for reading! Hopefully you enjoyed this post. If you did, don't hesitate to share it on Twitter or Facebook; having more readers will help the site grow. As well, consider a subscription if you're interested in reading additional work that isn't available to guests.