Goalies are voodoo.

The goaltending position is arguably the most important one on an NHL roster, but no matter the amount of talent, netminder performance is notoriously difficult to predict year over year. Bundling this unpredictability with the heightened risk of injury due to the physically demanding nature and heavy workload of the position creates a potentially perilous situation. Goaltending can save, sink or significantly sidetrack a season, and even with the right pieces in place, teams need a lot of luck to go their way.

Performance vs. Investment

If the numbers show one thing, it’s that money can’t buy you glove. Just one look at the ten highest-paid goaltenders over the last three seasons and their performance in two key goaltending metrics (EV SV% – even strength save percentage & GSAx/60 – goals saved above expectation per-60) shows that paying top dollar for your starting goaltender does not necessarily mean getting top performance.

Let’s flip around the view. Below is a listing of goaltenders who finished top 10 in either of the aforementioned goalie metrics. The data shows that high-end goaltending performance comes from all over the salary spectrum.

What if we looked at it from the perspective of how a team’s overall spend on their goaltending tandem relates to success in the regular season standings? I’ve assembled a listing of every team’s goaltending tandem over the last 6 seasons to see how much of the cap was devoted to those two goaltenders and how many points that team accumulated over the course of season:

(Interactive Tableau dashboard here.)

If spending more of your cap space on goaltending correlated strongly to a higher number of standings points, the plotted dots in this diagram would mostly follow a north-eastern diagonal trend from the bottom left (low spend, low points) to the top right (high spend, high points) of the graph. Instead, the dots are seemingly random. The trend line is nearly flat with a slight negative downslope, indicating that there’s virtually no correlation between standings points and how much cap space a team allocates to their goaltending.

And what about playoff fortunes? The story does not materially change: devoting a large portion of your cap space to goaltending is not a strong predictor of playoff success, or even participation. Below is a list of the 25 teams who’ve spent the highest proportion of the cap on goaltending since 2013-14: 60% of those teams missed the playoffs, in a league where 52% of teams make the playoffs automatically each year (unless a pandemic strikes and all logic goes out the window, but I digress).

Finding the perfect mix

So what is the recipe for success? Well, one way to tackle this is try to drill down into what optimal roster construction looks like for a contender. For the purposes of this analysis, a contender will be considered anyone who made it to at least the third round of the playoffs in the last six seasons. Below is a graph that looks at every team that fits this criteria:

Key Insights

  • A Final Four team has on average spent 8.4% of the cap ceiling on their goaltending tandem.
  • A Stanley Cup winner has on average spent 9.0% of the cap ceiling on their goaltending tandem.
  • The Toronto Maple Leafs spent 7.0% of the cap ceiling on Frederik Andersen and Michael Hutchinson (who got the nod over Jack Campbell due to games played and time spent on roster) this season.
  • Most expensive tandem among Stanley Cup winners: Washington’s 10.1% spent on Braden Holtby and Philipp Grubauer in 2017-18.
  • Least expensive tandem among Stanley Cup winners: St. Louis’s 6.3% spent on Jordan Binnington and Jake Allen in 2018-19.
  • Boston came within one win of smashing Washington’s high score in last year’s playoffs, making it to Game 7 of the Finals with 12.3% of the cap tied up in Tuukka Rask and Jaroslav Halak. (We’ll get to why Boston is an outlier in the next section.)

How does this impact the Leafs?

The Leafs have one of the most unique team compositions in the NHL, with 49.7% of this year’s cap tied up in the four-headed monster of Auston Matthews, John Tavares, Mitch Marner, and William Nylander. This by far leads the NHL this season, with the next-most expensive quartets being San Jose (Couture/Burns/Karlsson/Vlasic) at 42.3% and Chicago (Kane/Toews/Seabrook/Saad) at 41.6%. The difference between the Leafs and the next-closest team is 7.4%, representing just slightly more than $6 million in today’s cap dollars. Putting things into perspective, that’s a top-four defenceman’s salary between the Leafs’ Big Four and second place.

Being this top-heavy has not historically been a recipe for success in the post season. Let’s look again at those final four teams from the last six postseasons and the percentage of their cap spent on their four most expensive skaters:

Key Insights

  • A Final Four team has on average spent 35.7% of the cap ceiling on their four most expensive skaters.
  • A Stanley Cup winner has on average spent 39.8% of the cap ceiling on their four most expensive skaters.
  • The Toronto Maple Leafs spent 49.7% of the cap ceiling on their four most expensive skaters this season. This is a full 14(!!!)more than the Final Four average and 9.9% more than the average Stanley Cup winner.
  • Most expensive Big Four among Stanley Cup winners: Pittsburgh’s 45.2% spent on Crosby, Malkin, Letang and a discounted Phil Kessel (you’re welcome, Pittsburgh) in 2015-16.
  • Least expensive Big Four among Stanley Cup winners: Chicago’s 35.2% spent on Kane, Toews, Sharp and Seabrook in 2014-15. (Context: This was before Kane and Toews’ $10.5 million mega-deals kicked in, and the Blackhawks were majorly helped by now-illegal long-term back-diving contracts given to Duncan Keith and Marian Hossa.)
  • Boston went to the finals last year on the strength of a Big Four (Krejci, Bergeron, Pastrnak, Marchand) taking up just 33.9% of the cap ceiling.

If the Leafs are to go on an extended playoff run with the way this team is current constructed, they’re going to have to be one of the great outliers of the cap era. Complicating matters is that these deals were negotiated and signed under the impression of a continually rising cap ceiling, the growth of which is now put into doubt by the current COVID-19 crisis. Even if the Leafs wanted to match the highest point on this chart (Pittsburgh’s 45.2%) the cap ceiling would have to hit $90 million. This may not happen until the tail end of Matthews’ and Nylander’s contracts which expire in 4 seasons from now.

What we’ve learned

So let’s recap what we’ve learned:

  • Goaltender performance is difficult to repeat/predict year-over-year
  • Top-end goaltender performance comes from all over the salary spectrum
  • Devoting lots of cap space to your goaltending tandem has virtually no correlation to team success
  • The average Final Four team spends 8.4% of the cap on goaltending and 35.7% on their top four skaters
  • The average Cup Winner spends 9.0%  of the cap on goaltending and 39.8% on their top four skaters
  • The Leafs are spending 7.0% of the cap on goaltending and 49.7% on their top four skaters

Finally, let’s combine the last two bar charts into one ungodly view of everything:

Key Insights

  • A Final Four team has on average spent 44.1% of the cap ceiling on their goaltending tandem & Big Four.
  • A Stanley Cup winner has on average spent 48.8% of the cap ceiling on their goaltending tandem & Big Four.
  • The Toronto Maple Leafs spent 56.7% of the cap ceiling on their goaltending tandem & four most expensive skaters this season. This comes in 2.6% higher than the highest team on this graph (Pittsburgh, 2015-16).
  • Interestingly, four of the top five teams on the list ended up winning the Cup.
  • Boston went to the finals last year on the strength of a goaltending tandem & Big Four taking up 46.1% of the cap ceiling.

You may be wondering why Boston keeps coming up. The main reason is that Boston, along with Tampa, is a huge obstacle in the way of the Leafs being able to go on an extended playoff run. The NHL’s (broken) playoff format means the road to the Stanley Cup likely goes through TD Garden, calamitous officiating, and Zdeno Chara sucker punches, which poses a unique challenge when you compare these teams’ cap situations. For years, Boston has been getting incredible value from Patrice Bergeron, Brad Marchand and David Pastrnak, who all make less than $7 million per season. These bargain deals will afford Boston the flexibility to supplement this extremely productive core with high-end support pieces, which is why they were able to ride a very expensive goaltending tandem to the Stanley Cup Finals last year.

The challenge for the Leafs is how to compete with a team that matches them in talent but spends 10% less on their best and most critical players. This is where the Leafs’ investment in non-capped elements like scouting and development come into play. While Leafs management have done a fantastic job of building depth through bargain bin signings (Ennis, Mikheyev, Spezza) and draft steals (Sandin, Robertson) in recent years, continued success in these areas is going to be essential to rebalancing the scales, especially under a flattened cap. For the last few seasons the Leafs have benefitted greatly from some bargain deals (Andersen, Hyman, Rielly, Kadri) signed under Lou Lamoriello’s inconsistent reign. Kadri’s gone, but looming contract expirations for Andersen and Hyman (Summer 2021) and Morgan Rielly (Summer 2022) are going to force Dubas into some of the hardest decisions of his general managing career. Let’s talk Andersen.

The Frederik Andersen decision

The Leafs have been getting great value out of Frederik Andersen ever since he signed his 5-year/$25 million deal in the summer of 2016. Since the 2016-17 season, Andersen has the 9th-best all situations save percentage among goaltenders with at least 10,000 minutes played and shows pretty favourably in advanced metrics as well:

Andersen has accomplished all of this while playing the most minutes and facing the most rubber of any goaltender in the league during that timeframe. The conclusion? Freddy is going to get paid next summer, probably in excess of 8% of the cap judging by what those of a comparable age have signed for in recent memory.

But will it be the Leafs signing that cheque? Cap space is finite: with the Leafs spending so much on four skaters, there’s an additional pressure upon management to maximize value in other parts of the roster.

Things aren’t going to get any cheaper. First, let’s look at 2020-21: the final year of Andersen’s current contract and first year of Jack Campbell’s two-year extension given to him by Los Angeles. Assuming a flat $81.5 million cap next season, Andersen and Campbell will combine for $6.65 million in salary, which comes to 8.2% of the cap, a rise of 1.2% from 2019-20. This is going to push the Leafs to 57.9% allocated to their goaltending tandem & Big Four, which again will lead the league by a considerable margin. The uptick isn’t likely to be too much of an issue: with $6.75 million coming off the cap this off-season via the (assumed) departures of Ceci and Barrie, and only Dermott and Mikheyev to re-sign (likely to bridge deals), the Leafs can swallow the extra ~$1 million in spending on goaltending.

Where things get complicated is in 2021-22. Both Hyman and Andersen come up for renewal, and the Leafs need to earmark money for Morgan Rielly’s extension in 2022-23. Andersen, now 32 years old, will be looking for a significant raise and long term on his next deal. With what we know about aging curves for goaltenders, how the slim differences in performance margins do not justify the gigantic disparity in pay among goaltenders, and how contending teams have historically allocated money to goaltending and top-end forwards, the Leafs as they are currently configured likely cannot stomach spending an additional $2-3 million of their cap on Andersen’s new contract. It’s like Jenga: the more you take from the bottom of the tower and put on the top, the more wobbly it becomes.

A sad realization begins to dawn that we may only have roughly 100 games left of Frederik Andersen as the Leafs’ starting goaltender. Ideally the Leafs would start shifting focus to planning for an eventual successor to Andersen in 2021-22 who can closely replicate his performance levels while being cost-controlled. That means going younger and cheaper. Who the Leafs should target is a topic for another day, but do not be surprised if the Leafs take a flyer on a goaltender in the first three rounds of the 2020 NHL Entry Draft, or make a deal next summer for a young, talented netminder who is the victim of positional depth in another organization. With how the Leafs are currently built, acquiring goaltending that can provide excellent contract value is an absolute necessity for their contending hopes.

(Goaltending performance data courtesy of and
Salary details courtesy of and