The goaltender is generally the most important player on the team. Even the best players don't play more than half the game, but the goalie is between the pipes almost every minute, and of almost every game. While hockey is a team sport, ultimately the primary responsibility in keeping the puck out of the net, which is half the job of winning a game, falls on the goalie. In recent seasons we've seen how a great goalie can take a mediocre team all the way to the Stanley Cup, whereas a great team can really struggle with the wrong man in net.
Given its importance, how much should each team spend on goaltending in the new NHL era of the salary cap? On a roster of 20 players there are 2 goalies, 10% of your roster space. However, a goaltender is almost always one of the six players on the ice at any given time, so possibly it makes more sense to spend 16.7% of your cap space on goalies. Let's look at the five teams that spent the highest and lowest percentage of their salary on goaltending last season.
G%: Percentage of team salary spent on goaltending.
New York R 13.5%
Los Angeles 5.9%
In 2008-09, only one team spent more than 16.7% on goaltending, the Chicago Blackhawks. Despite the small sample size, we can see that of the five teams that spent the lowest percentage of their total salary on goaltending, only the mighty Detroit Red Wings advanced to the postseason. Of the five teams that spent the most on goaltending, only the Toronto Maple Leafs failed to qualify. In Toronto's case, it may have been a case of wasteful spending.
Fortunately, we do have a way to measure how effectively money is being spent: the new Goals-Versus-Salary (GVS) statistic. GVS uses GVT, which measures, in goals, all of a player's or goalie's contributions relative to a replacement-level player or goalie. By subtracting the cost of a replacement-level player (roughly 0.5 million) from a goalie's salary and multiplying by 3 (120 average team GVT divided by 40 million in remaining cap space), we know what we should expect from each goalie given his salary. GVS is simply the difference between a player's actual GVT and his expected GVT. Here is last season's top ten values.
GVT: The value the goaltender provided over replacement value, measured in goals.
Sal: The goaltender's 2008-09 salary, in millions.
GVS: The value the goaltender provided, measured in goals, relative to what you could get for the same money.
Goaltender Team GVT Sal GVS
Tim Thomas Boston 35.8 1.1 34.0
Niklas Backstrom Minnesota 27.7 3.1 19.9
Steve Mason Columbus 16.2 0.85 15.2
Pekka Rinne Nashville 15.0 0.58 14.8
Tomas Vokoun Florida 27.7 5.5 12.7
Craig Anderson Florida 12.9 0.58 12.7
Jonas Hiller Anaheim 14.9 1.3 12.5
Scott Clemmensen New Jersey 11.6 0.5 11.6
Cam Ward Carolina 16.5 2.5 10.5
Mike Smith Tampa Bay 11.8 0.95 10.5
It's no surprise to see Tim Thomas at the top of the list. Not only is he the league's best goalie, but the Bruins got him for the bargain price of $1.1 million. Even next season when he stands to make a whopping $6 million, he would still have a GVS of almost 20.
The Florida Panthers were fortunate last season to have a rare example of a highly-paid player like Tomas Vokoun still manage to provide exceptional value. Even their back-up goaltender Craig Anderson delivered exceptional value for the dollar. When they lost Anderson to Colorado in the offseason, they picked up Scott Clemmensen, who was the 8th best value in the league.
The Toronto Maple Leafs' starting goaltender tops the list of the worst values of 2008-09 season, which helps explain why they could invest so much on goaltending but still miss the playoffs.
Goaltender Team GVT Sal GVS
Vesa Toskala Toronto -18.0 4.0 -28.5
Miikka Kiprusoff Calgary -1.7 8.5 -25.7
Marty Turco Dallas -10.6 5.7 -24.9
Chris Osgood Detroit -16.3 1.7 -19.9
Jean-Sebastien Giguere Anaheim -4.0 5.5 -19.0
Pascal Leclaire Columbus -9.6 3.0 -17.1
Manny Legace St. Louis -9.6 3.0 -17.1
Jose Theodore Washington -3.3 4.5 -15.3
Johan Hedberg Atlanta -13.3 1.0 -14.8
Rick DiPietro New York I -1.5 4.5 -13.5
Toskala isn't the only highly-paid starting goalie in the ten worst values of 2008-09. Miikka Kiprusoff actually performed worse than a replacement-level goaltender despite his whopping $8.5 million dollar price tag.
Still, you can't fault NHL GMs for paying the big bucks to goalies with the impressive historical credentials of Toskala, Kiprusoff, Turoc, Giguere and Theodore. At the time their contracts were signed, they definitely appeared capable of providing great value for the dollar. Looking ahead to next season, let's examine the free agent signings to see which GMs may have found the best values, and which ones may have fallen into some money-wasting traps.
GVT: 2008-09 GVT
Sal: 2009-10 Salary
GVS: 2009-10 GVS if GVT remains the same
Goaltender Team GVT Sal GVS
Martin Biron New York I 14.3 1.4 11.6
Scott Clemmensen Florida 11.6 1.2 9.5
Craig Anderson Colorado 12.9 1.8 9.0
Dwayne Roloson New York I 14.2 2.5 8.2
Josh Harding Minnesota 8.3 1.1 6.5
Kari Lehtonen Atlanta 5.1 3.0 -2.4
Jason Labarbera Phoenix 1.4 1.0 -2.9
Mathieu Garon Columbus -4.0 1.2 -6.1
Andrew Raycroft Vancouver -8.4 0.5 -8.4
Peter Budaj Colorado -6.3 1.2 -8.4
Based on their 2008-09 GVT and their 2009-10 salaries, the New York Islanders have picked up two goaltenders in Biron and Roloson that could provide great value. Since they both can't be the #1 starter, one or both of them are going to see their GVT drop, but any way their ice-time is divided, the Islanders ought to be much improved in the net for next year.
Florida lost Craig Anderson to Colorado, but replaced him with Scott Clemmensen, who is almost as good and cost the Panthers $0.6 million less. Colorado made an excellent move in picking up Anderson, but wound up overpaying Peter Budaj, which essentially negates the value of the move. Budaj's former netminding partner Andrew Raycroft was picked up by Vancouver, but even at the league's minimum salary he'd still be the league's worst value among the offseason signings.
In Vancouver, Raycroft would presumably only be used in relief of Roberto Luongo, but last season Raycroft was the league's worst goaltender when being brought in as relief. It's too bad they didn't hang onto Curtis Sanford (4th best) or Jason LaBarbera (8th best) instead.
Goaltender Team SV% Signed With
Antero Niittymaki Philadelphia 1.000 Tampa Bay
Mathieu Garon Pittsburgh .967 Columbus
Tomas Vokoun Florida .963 Florida
Curtis Sanford Vancouver .954 Montreal
Josh Harding Minnesota .952 Minnesota
Kari Lehtonen Atlanta .936 Atlanta
Craig Anderson Florida .931 Colorado
Jason LaBarbera Vancouver .931 Phoenix
Andrew Raycroft Colorado .787 Vancouver
In Relief, minimum 100 minutes
It's interesting that six of the top eight relievers in the NHL last season were signed to new contracts. Entering a game in relief is so difficult for goaltenders that it's rarely done. No matter how badly the starting goaltender might be struggling, in need of a short rest or even in need of a word from the coach, coming into a game cold and playing at such a high level is an undervalued talent, so undervalued in fact that most of these goalies got signed at bargain rates.
In the end, today's era of salary cap NHL does not necessarily reward the teams that spend the most on goaltending, but rather the organizations that get the best value for their money. I believe the Florida Panthers have it right, by spending whatever it takes to keep a very high-value goaltender like Tomas Vokoun, and backing him up with one of the tremendous values you can easily find among back-ups, like Craig Anderson in 2008-09 or Scott Clemmensen in 2009-10.
Earlier in the season we introduced the Quality Start (QS) statistic to help us remove some of the bias inherent in the more common Goalie Win statistic. Since a QS is awarded if the goalie plays well enough for his team to win, regardless of how well his team played offensively, it allows us to examine how consistently a team's goaltender is contributing. Here are the league's five best and five worst in 2008-09, minimum 20 starts.
GS: Games Started
QS%: Percentage of GS that were Quality Starts
Goaltender Team GS QS%
Tim Thomas Boston 54 70.4%
Henrik Lundqvist New York R 70 67.1%
Niklas Backstrom Minnesota 71 64.8%
Martin Brodeur New Jersey 30 63.3%
Cristobal Huet Chicago 40 62.5%
Vesa Toskala Toronto 53 43.4%
Patrick Lalime Buffalo 21 42.9%
Jean-Sebastien Giguere Anaheim 44 40.9%
Chris Osgood Detroit 44 34.1%
Johan Hedberg Atlanta 27 33.3%
For the 2008-2009 season, a .913% or higher save percentage represented a Quality Start.
At 62.5%, Cristobal Huet's consistent performance as a starting goalie last season is no doubt one of the reasons why Chicago felt comfortable letting Nikolai Khabibulin go to the Oilers. If he can continue to provide quality goaltending at a rate almost as high as the league's elite (Brodeur, Backstrom, Lundqvist and Thomas), there's no question that their high-octane offense should be able to fight the Red Wings for the division lead, especially if Chris Osgood continues to play so erratically (34.1%).
Robert Vollman is an author of Hockey Prospectus.
You can contact Robert by clicking here or click here to see Robert's other articles.