Evaluating the Martin Jones Contract

One of the biggest changes of the Sharks’ 2015 offseason was the departure of long-time starting goalie Antti Niemi. While Niemi signed up for co-starter/backup duty in Dallas, Doug Wilson sent a 2016 first-round pick to Boston to acquire RFA goalie Martin Jones, who signed a three-year, $9M contract to become San Jose’s new starter. Prior to his trade to the Bruins, Jones provided two seasons of excellent work as Jonathan Quick’s backup in Los Angeles; as a King, his even-strength Sv% was a sterling 0.934 (0.935 adjusted for shot difficulty). Still, these numbers came in just 34 games, and Jones’s undrafted status and unspectacular AHL play led some to wonder whether the move was a mistake. It’s looking like a wise bet thus far, with Jones sporting a 0.941 5v5 Sv% through 17 games in teal. But what should Sharks fans expect moving forward, and how should they feel about the contract Jones signed?

Of course, when judging an NHL contract, it’s worth keeping in mind that fans and pundits can evaluate player moves in a vacuum, but teams don’t have this luxury. It’s easy to point to Jones’s inexperience and argue that Wilson paid too steep a price, but this assumes that a better option was available. If the Sharks’ goal is to return to the playoffs and contend during the final two years of Joe Thornton’s and Patrick Marleau’s contracts, a better goaltending tandem than Alex Stalock and Troy Grosenick was absolutely necessary. But the only realistic options involved resigning Niemi or finding a new goalie. Of the other goalies available as UFAs last summer, Niemi was the only credible starter, and all of the younger goalies who changed teams via trade (e.g., Robin Lehner, Eddie Lack, Cam Talbot) shared Jones’s lack of track record. Other goalies rumored to be available, like Craig Anderson or Jonathan Bernier, may have been worth pursuing, but it’s safe to assume that players who don’t change teams weren’t available for a reasonable price. So, if we’re judging whether bringing in Jones was a good idea, we have to compare its cost and likely outcomes against San Jose’s two realistic alternatives: a) matching Dallas’s three-year, $13.5M offer to Niemi, or b) trusting Stalock with 55-60 starts, and trusting Grosenick (who has played 118 total minutes in the NHL) to handle 20+ starts as well.

The first step in answering this question involves forecasting the next three years of Jones’s performance, and comparing it against similar forecasts for Niemi and Stalock. To estimate 5v5 Sv%s over the next three seasons, I used garik16’s “Hockey Marcels” method (which itself draws on work by Eric Tulsky). Briefly, this method creates a three-year forecast by using the prior four seasons of a goalie’s work, with more recent seasons weighted more heavily. It applies an appropriate regression to the mean, and uses another adjustment to account for age-related decline over the forecast period. For the mean regression, I added shots saved at an average rate until the total sample was 4,000 SA. I also applied different “average” Sv%s depending on the goalie’s role. For Niemi, an established starter, I regressed the data to the average 5v5 Sv% of NHL starters (just over 0.924), while Stalock’s results were regressed to the backup average of 0.917. Jones is not clearly a backup or a starter, so I regressed his data to a league-average 0.922. With these results, I used Monte Carlo simulation to estimate the probabilities of each goalie performing at the level of at least an average starter, and at a below league-average level, in each season. All the goalie data come from WAR on Ice, and the simulations were performed in R.

First off, you’ll notice that the forecast doesn’t expect spectacular things from Jones. Given his limited work at the NHL level, this arises from regressing his performance to league average. If you believed that Jones is a legit starter and used the starters’ average for his forecast, the numbers would jump substantially. Still, even a league-average Jones can be expected to outperform both Niemi and Stalock over the next three years. While this might surprise some of Niemi’s more impassioned defenders, the key here is the age adjustment. At 25, Martin Jones should be in the prime of his career, and whatever level of performance you think he’s capable of should be what you get over the next three years. In contrast, Niemi is already 32; based on what we know about goaltender aging, performance generally declines badly as goalies reach their mid-30s, and falls off a cliff at 35. Nemo might well be a solid goalie who took more criticism from Sharks fans than he deserved, but it’s safe to assume that Jones will outplay him moving forward.

If we consider the cap implications, the Jones deal looks even better. $3M for multiple seasons sounds like a lot, until you remember that NHL starters have a median annual cap hit of over $5.7M. If you think of the comparison between Jones and Niemi in terms of cost per goal prevented, Niemi is likely to deliver poorer play for a cap hit that’s 50% higher. Guessing at the terms of Stalock’s next contract is not easy, but even if he comes in at a cap hit lower than Jones’s, he’s likely to decline to below-average-backup level over the same term, and the chances that a contending team will get the goaltending it needs from him will become increasingly remote. And, of course, both the Jones and Niemi scenarios feature the added value of having Stalock in the backup role rather than Grosenick.

So, while we’d all be happy if the Sharks had replaced Niemi with the second coming of Dominik Hasek, it’s only fair to judge Doug Wilson’s work here in the context of the available options. Considering the alternatives, I’m pretty happy with the move and the contract. Despite Jones’s relative inexperience, it’s not clear that a better goaltender was available, and any team with a limited window of contention can’t afford to wait for the perfect long-term option to come along. Whatever your opinion of Antti Niemi’s years in a Sharks uniform, it’s unlikely that he’ll continue to provide even league-average goaltending over the next three seasons; assuming that Jones is at least a league-average talent, his performance should outshine Niemi’s over that period. That San Jose will get that performance at a lesser cost – preserving cap space that can be used to fill holes elsewhere in the lineup – is a further bonus. And there is a decent chance that Jones will exceed the NHL starters’ average Sv% of 0.924 in each year of his deal, which would make him an incredible bargain. It might not be how many of us expected the team’s goaltending situation to sort itself out back in the summer, but thus far it’s looking like strong work from the front office.