When the Toronto Maple Leafs signed Frederik Andersen to a five-year contract with a $5 million cap hit, they made a pretty bold statement. It’s clear that they see Andersen as a goalie who will backstop the franchise to success over a long period of time, not as merely a stop-gap. A five-year contract for a goalie is a major gamble. If you sign a winger to a big deal and it doesn’t work out, you can put him on the 3rd line and shelter him from difficult competition in order to try to salvage some value from the contract. On the other hand, if a goalie is signed to a hefty contract that doesn’t work out, you’re trapped. Look at Detroit’s situation with Jimmy Howard or Dallas’s with Kari Lehtonen as examples.
It’s clear that the Leafs think Andersen will be able to keep up a high level of play over the long haul. The key question is: are they right?
GOALIE AGING CURVES
Many hockey fans believe that goalies peak at an older age than forwards. It’s certainly true that goalies are less likely to become regular NHL players at a young age. Over the past five seasons, 74 forwards have played at least 40 games during their age 21 season. By comparison, only one 21 year old goalie played 30 games in a season over that span, and only two others even played 20. Even accounting for the fact that forwards outnumber goalies 12:1, that’s a very low total for goalies, reflecting the fact that they usually enter the league at an older age.
But the available evidence suggests that goalies do not peak at a later age than forwards. Steve Burtch found that goalies peak around the same age as forwards – about 23-26. Eric Tulsky (currently employed by the Carolina Hurricanes) found an even more linear downward slope. Garik16’s analysis produced a result similar to Tulsky’s. It seems fair to say that right now we have a pretty good volume of writing that demonstrates that goalies don’t become good any later than skaters do. Just like with forwards and defencemen, goalie ability falls off through the late 20s, and then falls even more sharply past 30.
One thing that’s always worth keeping in mind about aging curves is that they’re averages over a large population of players. Just because goalies on average decline past their mid-20s doesn’t mean that any particular goalie will. Individual players have very different paths; aging curves just give us a baseline expectation.
With that in mind, I decided to try to drill down into the data in a bit more detail in order to answer a related question: how have goalies with similar profiles to Frederik Andersen aged in the salary cap era? Andersen is currently 26 years old, with 125 games played over three seasons, and an above average career SV% (.918 vs about .915). If we want to know how similar players have aged, we need a list of every goalie in the salary cap era who has played at least 100 games between 24-26 years old with an above-average SV%, and who has played through their 31-year-old season.
Those criteria resulted in a fairly small list, just six players: Cam Ward, Jimmy Howard, Kari Lehtonen, Henrik Lundqvist, Marc-Andre Fleury, and Ryan Miller. It’s not a huge sample size, but given the small number of starting goalies in the league at any given time, it’s not too surprising, and it’s what we’ve got to work with. First, let’s see how they performed from the standpoint of save percentage over the five years following their 26-year-old season.
What we see is a big jumble. This shouldn’t be too surprising; goalie performance is notoriously difficult to predict, with potentially huge variations year-to-year. Even though the goalies here all started around the same point in the chart, they diverge quite a bit in subsequent years. At age 29, just two of the goalies are clearly above average, with two below average, and two right around average. It’s a fairly similar story at age 30, while Lehtonen’s fall-off at age 31 gives us a split of three below-average, one average, and two above-average goalies.
One problem with the list above is that these goalies weren’t all the same age at the same time. Ryan Miller and Henrik Lundqvist are in their mid-30s, whereas Cam Ward and Marc-Andre Fleury are a few years younger, having recently turned 31. League average SV% has risen steadily over the salary cap era, largely because the number of powerplay opportunities has fallen. That means it’s not entirely fair to compare these goalies in the fashion I’ve done.
One thing we can do to account for that is to look only at 5v5 SV%. 5v5 SV% has not risen nearly as much as overall SV%, so it gives us a bit of a better basis for comparison between goalies who played in different recent seasons. Here’s what the aging curves for these goalies look like with only 5v5 SV% included (for comparison, league average is about .923 and Frederik Andersen is at .926):
Once again we see a pretty big jumble, and a similar pattern emerges. Cam Ward and Jimmy Howard have clearly not aged well, despite earlier success. Kari Lehtonen performed at a fairly average rate before suddenly falling off last season at age 31. Lundqvist and Miller both played at a fairly high level through the whole period looked at here. And Marc-Andre Fleury . . . who knows.
So what does this all mean for Andersen? Well, the good news is that all of these goalies remained starters throughout the entire period looked at here. Some of that is due to inertia and the investments that teams made in these goalies, but it is at least somewhat reflective of their skill as well. Even though Cam Ward and Jimmy Howard failed to live up to their contracts, they did at least continue to put up numbers consistent with a respectable back-up goalie; nobody here played so poorly that they ought to have been out of the league.
The bad news is that the rest of the results make it harder to be overly optimistic. I think we can all agree that no matter how good Andersen is, he clearly isn’t on Henrik Lundqvist’s level. Of the remaining five goalies, only Ryan Miller consistently put up a high 5v5 SV% right through the five years looked at here. Two more (Howard and Ward) had clearly fallen off by age 30, while Kari Lehtonen remained pretty average until suddenly falling off. We can say that about half of these goalies continued playing well into their 30s.
If Andersen has just a 50/50 shot at remaining league average or higher into his 30s, the Leafs have loaded a lot of risk onto the back-end of his contract. Since those are years when the Leafs likely hope to be Stanley Cup contenders, that’s a pretty big risk to take. You might argue that his contract will be easier to move by then if that happens, but I don’t think that’s so clear. Look at Dallas’s problem with Kari Lehtonen right now. No one wants to trade for your $5M back-up goalie.
The general goalie aging curve doesn’t paint an especially rosy picture either. If Andersen has the same curve as an average goalie, we should expect him to slowly decline over the next few years, with a bit of a sharper drop once he hits 30.
None of this is to say that I can predict with any confidence what will happen to Andersen. Look no further than Marc-Andre Fleury’s sudden resurgence over the past couple of years as an example of how unpredictable goalies can be. Frederik Andersen could certainly continue to provide the Leafs with a high level of goaltending four or five years from now. But history suggests the Leafs have made a pretty risky gamble.