You can't simply buy yourself a championship in the new NHL. Even if you somehow get your hands on the league's best players, you still have to find a way to pay them all and fit your payroll under the current season's salary cap. In some cases you have to let great players go and replace them with inferior players in order to free up enough space to pay your other stars. The key to building a successful team is to identify which players offer the best value for the dollar, and to manage your roster accordingly.
Determining the value of a player requires only two components: salary and a way of measuring a player's contributions. The ideal way to measure a player's total contributions is by using Goals-Versus-Threshold, or GVT, because it summarizes all different types of contributions into one measurement, regardless of whether they are offensive, defensive, in nets, or in the shoot-out. Therefore, let's introduce Goals-Versus-Salary (GVS) which, instead of comparing a player's contributions against a replacement player like GVT does, compares those contributions against what you would expect for the same amount of money.
Based on their 2008-09 GVT and salary, here are the top 10 best values among skaters. GVS is listed in the final column, and it measures in goals the value the team received from each player. For example, if the Detroit Red Wings did not have Johan Franzen, they could have used his $1.15 million dollar salary to obtain one or more players who would have contributed, on average, 13.3 fewer goals than Franzen to their bottom line.
Player Team Salary GVT GVS
Evgeni Malkin Pittsburgh 0.98 23.4 +22.0
Zach Parise New Jersey 2.50 24.2 +18.2
David Krejci Boston 0.83 18.3 +17.3
Nicklas Backstrom Washington 0.85 17.4 +16.4
Phil Kessel Boston 0.85 15.3 +14.3
Johan Franzen Detroit 1.15 15.2 +13.3
Devin Setoguchi San Jose 0.85 13.9 +12.9
Loui Eriksson Dallas 1.50 15.5 +12.5
Anton Babchuk Carolina 1.00 13.7 +12.2
David Booth Florida 0.68 12.3 +11.8
The first observation of GVS is that the top values are generally players with relatively limited NHL experience and whose salaries are therefore temporarily low. As a frame of reference, the worst player on this list, David Booth, could have been paid $4.5 million (over 6 times as much as he was actually paid) and still would have been a bargain. David Krejci and Phil Kessel together offered 31.6 goals to the Bruins above and beyond what they could have hoped to receive using the same amount of money. Looking at goal differential (team goals for minus team goals against), we can see their impact.
Team Goal Differential
Bruins* 46 (Bruins without Krejci and Kessel)
From 1st in goal differential to 4th, can you think of a better way to spend $1.68 million? I can definitely find some far worse ways to use your money. Here are the worst values of the 2008-09 season.
Player Team Salary GVT GVS
Mats Sundin Vancouver 8.6 1.6 -22.7
Daniel Briere Philadelphia 8.0 4.3 -18.2
Wade Redden New York R 8.0 4.4 -18.1
Joe Sakic Colorado 6.0 0.6 -15.9
Dany Heatley Ottawa 10.0 13.0 -15.5
Marian Gaborik Minnesota 7.5 5.6 -15.4
Brad Richards Dallas 7.8 6.6 -15.3
Sergei Zubov Dallas 5.4 -0.4 -15.0
Mike Fisher Ottawa 6.0 2.1 -14.4
Scott Gomez New York R 8.0 8.2 -14.3
To be fair, several of these players suffered injuries during the season and played only a portion of the 2008-2009 year, but in many cases that cap room was still wasted. For example, imagine what the Rangers could have done with an extra $16 million dollars if they didn't have Wade Redden and Scott Gomez taking up cap room, and to think, they almost won the Sundin sweepstakes too!
It may appear that GVS unfairly punishes the higher-paid players. GVS works by subtracting the cost of a replacement player from a player's salary, which is roughly $0.5 million, depending on the season. Then that remaining salary is multiplied by 3 to determine what a team should expect in overall GVT from that player. Why three? A team of 20 replacement players will cost roughly 10 million in salary, leaving about $40 million in cap space, depending on the season. By definition the league average GVT per team is 120, meaning each player should contribute 120 GVT / $40 million = 3 goals for every $1 million in salary.
Looking at all NHL players last season, the average player earned only 2 goals for every $1 million in salary, but using 2 instead of 3 would just wind up in a statistic that validates what NHL GMs are already doing. The statistic should be based on what makes sense, not on what's actually being done.
Lower-paid players, especially rookies, tend to be better values than their higher-paid counterparts, but if you populated your team exclusively with lower-paid players, you would be wasting at least $10 million in salary cap room, leaving you possibly 30 goals short of being competitive. If you want to win, you have to seek out some higher paid players and get the extra GVT, even if your value-per-dollar return decreases. In other words, you're better off using about half of your cap room signing some lower-valued but higher-producing players.
GVS can be useful not just for figuring out what each player is worth, but also for studying how well NHL GM's are managing their payrolls. Looking at the offseason free agent signings, here are the top 5 and bottom 5 based on their 2008-09 GVT and their 2009-10 salaries.
Player Team Salary GVT GVS
Jeff Woywitka Dallas 0.60 6.4 6.1
Rob Blake San Jose 3.50 14.9 5.9
Ruslan Fedotenko Pittsburgh 1.80 9.3 5.4
Travis Zajac New Jersey 2.75 11.8 5.1
Mark Recchi Boston 1.00 6.3 4.8
Steve Sullivan Nashville 3.75 5.3 -5.2
Scott Niedermayer Anaheim 6.00 11.0 -5.5
Travis Moen Montreal 1.50 -4.8 -7.8
Chris Neil Ottawa 2.00 -3.7 -8.2
Jay Bouwmeester Calgary 6.68 6.4 -12.1
Since this is based on GVTs from last season as well as next season's salaries, we can see how much a player like Chris Neil would have to improve to earn his new salary. For Steve Sullivan it should be easy, since he only played half the season and is off-target by exactly half. On the other hand, Ruslan Fedotenko is such a bargain that he could play without Sidney Crosby or Evgeni Malkin, and have his performance slide dramatically and still earn his salary.
GVS highlights the interesting situation in which the Calgary Flames have placed themselves. Between Bouwmeester's value, and Dion Phaneuf's 6.1 GVT and $6.5 million dollar salary, the Flames are paying their top defensive paring over $13 million for only 12.8 GVT. That's 2.1 goals less than Rob Blake alone, and he's making only $3.5 million next year. Phaneuf and Bouwmeester will have to play three times as well next season in order for the Flames to feel justified in their investment. That's a very tall order.
The Montreal Canadiens, having signed Travis Moen and Brian Gionta to contracts that can't be justified by their GVT, and also having trading for Scott Gomez (one of the league's worst examples of over-spending), might not be using its money as wisely as possible. Even if all three of them play very well in 2009-10, the fans will have to wonder if the Habs could have gotten more value for the same $14.5 million.
So far this offseason there have been many discussions on what certain players need to do to earn their salaries, and arguments about whether various teams are pursuing the right strategy in the NHL's new salary cap era. By using GVS we hope to remove some of these opinions and replace them with the numbers that give us true insight into what matters.
In an upcoming article Tom Awad will use GVS at a team sum level to take a closer look at Darryl Sutter of the Flames, Bob Gainey of the Canadiens and the league's other GM's to determine who is making the best use of cap money.
Just like GVT, GVS works equally well for goaltenders as it does for players, so in the coming weeks I'll use GVS to rate the league's goaltenders by value.
Robert Vollman is an author of Hockey Prospectus.
You can contact Robert by clicking here or click here to see Robert's other articles.