Craig Anderson: A Study in Risk Aversion

Let's get something out of the way: re-signing Craig Anderson is the safe thing to do. He's demonstrated over the course of his career that he's at worst a league average starter with the potential to catch fire for stretches; every Sens fan remembers his magical play during the '12-'13 season. Robin Lehner's mildly disappointing campaign last year sowed the seed of doubt in management's mind, and, coupled with the lack of an obvious internal candidate to act as a backup, the Senators went out and bought an insurance policy. Really, on the surface it's not the end of the world. But if you begin to peel away the layers underpinning the decision-making, it gets a bit shakier. Let's start with the underlying assumption that Craig Anderson is a safer bet than Robin Lehner.

Goalie Aging:

There was a telling bit right at the beginning of Pierre Dorion's media availability, that you can listen to here or read about here, where he expands on how he sees Craig Anderson performing in the near-term. From the 6th Sens transcription:

He plays with us (until) he’s in his mid-30’s and we feel that goalies through the years, even right now, are still performing at that age.

The above quote fits the current orthodoxy that goalies hit their primes later than skaters and that, more importantly in this instance, they also don't fall off as quickly. Lots of people wouldn't argue with the notion that a goalie in his mid-30s was still a valuable commodity. This is the most crucial piece of the puzzle in evaluating whether you think this extension is worthwhile: how do you expect Craig Anderson to perform in the last two years of his extension when he's 35 and 36? So far he's fought back age-based regression, but time remains undefeated. He's going to get worse at some point, but when? A quick summary of his performance the last six years:

A couple of notes on this table:

1. I've used Even-Strength Save Percentage (EV SV%) because it tends to be a more stable, relatively repeatable number than a goalie's save percentage across all situations. Short-handed save percentage has proven to be highly volatile and can have a very big effect on overall save percentage, especially in small samples. If Andy was getting worse, we'd see it here.
2. Andy's NHL Rank was calculated using goalies who faced at least 750 shots in the full seasons he's played and 440 shots in the lockout shortened season (750 pro-rated to 48 games),

His monstrous '12-13 campaign aside, Anderson seems like a reasonable bet to finish in the top 10 in save percentage during a good year and in the low 20's in a bad year. That kind of bankable performance is worth something, and $4.2 M is not an unreasonable price. In fact, if Anderson was 29 this would be a no-brainer. The problem is that a slew of recent studies into goalie aging have shown that goalies do not, in fact, usually perform into their mid 30s.

So now management is stuck making a tough, somewhat unknowable choice: do they bank on Anderson bucking the trend and remaining productive even past what should be his expiry date, or do they trust the pattern exhibited by virtually every other goalie that's played in the NHL? Remember, this extension covers his ages 34, 35 and 36 seasons.

Robin Lehner's Performance:

The opposite side of trying to figure out if Craig Anderson is setting up to fall off a cliff is deciding what to make of Robin Lehner. I think if you caught Sens' management in a candid moment, they'd admit this past year greatly shook their confidence in the young prodigy. That's not to say last season was a complete disaster for Lehner -- he did outplay Anderson after all, but there were a lot of peaks and valleys en route to a respectable .913 save percentage. If you believe the conventional wisdom that goalies age like fine wine, that's another point against him because he might not have hit his theoretical prime. This is, again, where I think the Senators might be wise to go against the conventional wisdom. There's good evidence that goalies actually attain their peak performance in their ages 24-27 seasons, and some projections have Lehner fairly substantially out-performing Anderson for the next few years. If you trust that Lehner's developed like the vast majority of NHL goalies in history, the time to hand him the reigns is coming sooner than Anderson's new contract extension is set to expire.

Bonus Consideration: The Internal Budget

As with every player personnel decision, opportunity cost has to be a primary concern. Much has been made of the internal budget in the last eighteen months or so, and I don't believe any outside observer can reasonably infer whether this signing is a good or a bad omen for the retention of Methot or Ryan. Dorion would have us believe it's great news:

I think it’s a sign from our ownership just to show that we’re making a long-term commitment here for as long as we can. So those are all things that come into play for us.

I don't see it being quite that simple. As it stands, in 2015-16 the Senators have $48M in salary commited to 16 players that do *not* include: Bobby Ryan, Marc Methot, Mika Zibanejad, Alex Chiasson, Mike Hoffman, Mark Stone, Erik Condra. That's a big part of the team's supposed core of the future that would need re-signing and there's just no way they would all fit into the $55-57M range we've often heard cited as the team's internal budget. It's tough to judge how much of an impediment Anderson's contract will be to the goal of keeping the band together until we know more about the '15 budget, but we can speculate as to where the money could have been optimally spent. For instance, for a team that struggled so mightily on defense last year would be well advised to spend on keeping possibly their second best defenseman around in Methot. Implied in Dorion's above quote: this is a statement to our fans that we are willing to lock up our core long term. I would feel a lot better about this if he had said something like: "We will have budget to spend to keep our best players", because otherwise the whole thing feels somewhat symbolic. Call it a push.

Conclusion:

It's not like this deal jumps out at you screaming bloody murder -- we're not talking about Ilya Bryzgalov to the Flyers or anything, but I think it's problematic in more subtle ways. Primarily, the Senators played it safe by betting on a goalie with a proven track record continuing to perform for the foreseeable future. It's hard to criticize a decision that takes the road most traveled. But it's a mistake because you're counting on Andy performing differently than the vast majority of goalies in NHL history. That's not a gamble I'd be quite so comfortable making. Secondly, they're worried about Lehner's development but that seems somewhat problematic too since Lehner's about to enter his prime.

Which brings us back to the over-arching thesis: this is a safe, risk-aversion decision by a team that can't afford to make these types of compromise choices. If Ottawa is going to compete on a budget, they'll need to make some calculated bets. Taking the $4.2M that they've given to Anderson and spending it elsewhere to improve depth, especially on defense, while trusting that Lehner will run with the starter's job is the kind of measured risk necessary to contend. If management feels it needs some reassuring that Lehner will be able to handle the load, they have a full season ahead of them to figure it out. Extending Andy today, at this price, hedges their bets but seriously mitigates the upside. Anderson's not going to perform like he did in '12-13, in the best case scenario he's a league average starer. When everything's considered, for this team, in this situation, it just strikes me as an overly safe play.

A final note: a huge thank-you to Garik, also known as @garik16, the 6th Sens, @6thSens, and Eric Tulsky, @BSH_EricT for all of their work that I've linked to and quoted above. Thanks also to Pension Plan Puppets and NHL Numbers for all their work as well.