Sunday, February 8, 2009

Evaluating Goalies: no clear link between performance and pay


 

Goalies are paid to do one thing. Stop the puck from going in the net. To evaluate goalies, it should be easy. To what degree do they stop the puck from going in the net (save %). This post looks to evaluate both save %, shot count and pay. The save % of a goalie may seem a little difficult for a casual fan to care about. What's the difference in a single game between goalies with a 92% save rate versus one with 91%? Well seemingly nothing in just one game. But that single % difference could mean a swing of 25 goals over the course of a season. That's big and can easily mean the difference between making and missing the playoffs.

The table below breaks down the teams that by shots and save %. This table simply looks at the player performance and salary. The Florida Panthers goalies look like good value here while the GM in Detroit and Dallas better give his scouts something to work on for next year.


 

Save % 

Shot Count 

Under 90% 

90-91% 

Over 90% 

Significantly more shots than average

Sives getting shelled

-Atlanta Thrashers ($6.3m)

-NY Islanders ($1.3m)

Sweating

-Tampa Bay Lightning ($1.3m)

-Anaheim Ducks ($6.3)

-Phoenix Coyotes ($1.3)

Overworked and playing great

-Florida Panthers ($2.2m)

About average amount of shots

Team Killers

-Colorado Avalanche ($6.1)

-St Louis Blues ($1.3)

-Toronto Maple Leafs ($1.3)

Mr. Average

-Pittsburgh Penguins ($1.3)

-Edmonton Oilers ($4.7m)

-Philadelphia Flyers ($1.3m)

-New York Rangers ($1.3m)

-Carolina Hurricanes ($6.1m)

-Montreal Canadiens ($2.2m)

-Washington Capitals ($1.3m)

-Calgary Flames ($6.1m)

-Vancouver Canucks ($1.3m)

-New Jersey Devils ($1.3m)

-Columbus Blue Jackets ($6.1m)


 

Over-achievers

-Buffalo Sabres ($6.1m)

-Boston Bruins ($6.1m)

-Minnesota Wild ($2.2m)

-Nashville Predators ($1.7m)

-Chicago Blackhawks ($6.1m)

Significantly less shots than average

Hiding their incompetence

-Detroit Red Wings ($6.1m)

-Dallas Stars ($6.1m)


 


 

Hiding behind their stats

-Ottawa Senators ($1.3m)

-Los Angeles Kings ($2.2m)

Awesome and bored

-San Jose Sharks ($1.3m)


 

If you buy that a team's +/- is the best indicator of team performance, then we can quickly see how a goalie individually contributes to his team's success. The average save % for NHL goalies in 2008-09 is 90.5%. If we replace all the team's current save % with the average, we can quickly get a view for how the records would be different. Watch the team with around an even +/- as they are the ones most on the bubble for the playoffs. To read it, the goal adjustment column is the amount of goals the team us up/down by based on their goalies play compared to the average goalie.


 

Team 

Current +/- 

Goal adjustment

Boston Bruins 

68 

+41

San Jose Sharks 

52 

+10

Chicago Blackhawks

42 

+16

Detroit Red Wings 

40 

-14 

Nashville Predators 

29 

+13

Washington Capitals 

23 

-3 

Philadelphia Flyers 

15 

+7

Minnesota Wild 

13 

+32

Buffalo Sabres 

10 

+13

Vancouver Canucks 

7 

0 

Calgary Flames 

6 

-6 

Montreal Canadiens 

6 

+1

Florida Panthers

4 

+26

Pittsburgh Penguins 

-1 

-5 

Anaheim Ducks 

-1 

+9

Los Angeles Kings 

-5 

-5 

Dallas Stars 

-5 

-26 

Columbus Blue Jackets 

-9 

-8 

St Louis Blues 

-12 

-12 

Carolina Hurricanes 

-13 

-2 

New York Rangers 

-16 

+8

New Jersey Devils 

-16 

+1

Edmonton Oilers 

-20

-5 

Colorado Avalanche 

-23 

-16 

Ottawa Senators 

-24 

-5 

Tampa Bay Lightning 

-25 

+6

Phoenix Coyotes 

-27 

+0

Toronto Maple Leafs 

-33 

-46 

Atlanta Thrashers 

-39 

-23 

New York Islanders 

-46 

-10 


 

The Boston Bruins enjoy incredible success due to good goaltending. That said, they would still be a good team with an average goaltending as well. The team most affected by poor goaltending is the Toronto Maple Leafs who appears to be putting AHL players in nets. And judging by the salaries they pay their goalies, it seems as though that is indeed the strategy. If the Leafs had even an average save %, they would be a better than average team. This could be good news for Leafs fans as there is an easy answer in the off-season. The Minnesota Wild is a poor team with very good goaltending. The Ottawa Senators are a bad team whose poor performance can only moderately be put on the goaltenders. Even an average goalie can't help the plummeting Sens.

If we look at individual goalie save % and compensation for the 2007-08 season, be prepared for a shocker. If you look at all goalies that played last season and their compensation, there is no relationship between pay and save % (correlation coefficient of +0.09). Crazy? I think so, maybe my MS Excel is broken or something. Let's look at all goalies that've played in at least 30 games:

Full Name 

Team  

Salary 

Sv%  

Dan Ellis 

NSH 

$500,000 

92.4% 

Ty Conklin 

PIT 

$500,000 

92.3% 

Jean-Sebastien Giguere 

ANA 

$5,500,000 

92.2% 

Tim Thomas 

BOS 

$1,000,000 

92.1% 

Marc-Andre Fleury 

PIT 

$1,600,000 

92.1%

Martin Brodeur 

NJD 

$5,200,000 

92.0% 

Ilja Bryzgalov 

PHX 

$1,362,500 

92.0% 

Niklas Backstrom 

MIN 

$3,100,000 

92.0% 

Cristobal Huet 

WSH 

$2,750,000 

92.0% 

Carey Price 

MTL 

$850,000 

92.0% 

Tomas Vokoun 

FLA 

$5,300,000 

91.9% 

Pascal Leclaire 

CBJ 

$1,400,000 

91.9%

Martin Biron 

PHI 

$3,500,000 

91.8% 

Roberto Luongo 

VAN 

$6,500,000 

91.7% 

Kari Lehtonen 

ATL 

$2,200,000 

91.6% 

Chris Osgood 

DET 

$800,000 

91.4% 

Mathieu Garon 

EDM 

$1,200,000 

91.3% 

Henrik Lundqvist 

NYR 

$4,250,000 

91.2% 

Manny Legace 

STL 

$1,800,000 

91.1% 

Evgeni Nabokov

SJS 

$5,000,000 

91.0% 

Martin Gerber 

OTT 

$3,700,000 

91.0% 

Jose Theodore 

COL 

$6,000,000 

91.0% 

Jason LaBarbera 

LAK 

$800,000 

91.0% 

Marty Turco 

DAL 

$5,700,000 

90.9% 

Nikolai Khabibulin 

CHI 

$6,750,000 

90.9% 

Alex Auld 

BOS 

$600,000 

90.7% 

Miikka Kiprusoff

CGY 

$3,600,000 

90.6% 

Ryan Miller 

BUF 

$2,500,000 

90.6% 

Cam Ward 

CAR 

$2,000,000 

90.4% 

Vesa Toskala 

TOR 

$1,375,000 

90.4% 

Peter Budaj 

COL 

$700,000 

90.3% 

Rick Dipietro 

NYI 

$4,500,000 

90.2% 

Dominik Hasek 

DET 

$2,050,000 

90.2% 

Dwayne Roloson 

EDM 

$3,500,000

90.1% 

Mike Smith 

TBL 

$950,000 

90.1% 

Chris Mason 

NSH 

$1,250,000 

89.8% 

Patrick Lalime 

CHI 

$950,000 

89.7% 

Fredrik Norrena 

CBJ 

$800,000 

89.6% 

Olaf Kolzig 

WSH 

$5,450,000 

89.2% 

Johan Hedberg 

ATL 

$1,150,000 

89.2% 

Ray Emery 

OTT 

$2,750,000 

89.0% 

Johan Holmqvist

DAL 

$1,000,000 

88.9% 


 

If we look at the goalies that played at least 30 games from a team salary and save % for 2007-08, we can evaluate the General Managers. I'm stunned to write this, but the relationship between pay and save % is -0.18. It looks like the Washington ownership got rooked on their goalie investment. I think I can guess a line item in resume of Montreal's GM. LA's GM put the lowest amount of coin into his goalies and still got decent results. Chicago paid 10x what LA did and got a worse save % for it.

Team 

Sum of Goalie Salary 

Sv%  

ANA 

$5,500,000 

92.2% 

PIT 

$2,100,000 

92.2% 

MIN 

$3,100,000 

92.0% 

MTL 

$850,000 

92.0% 

NJD 

$5,200,000 

92.0% 

PHX 

$1,362,500 

92.0% 

FLA 

$5,300,000 

91.9% 

PHI 

$3,500,000 

91.8% 

VAN

$6,500,000 

91.7% 

BOS 

$1,600,000 

91.4% 

NYR 

$4,250,000 

91.2% 

NSH 

$1,750,000 

91.1% 

STL 

$1,800,000 

91.1% 

LAK 

$800,000 

91.0% 

SJS 

$5,000,000 

91.0% 

DET 

$2,850,000 

90.8% 

CBJ 

$2,200,000 

90.8% 

EDM 

$4,700,000 

90.7% 

COL 

$6,700,000 

90.7% 

BUF 

$2,500,000 

90.6%

CGY 

$3,600,000 

90.6% 

WSH 

$8,200,000 

90.6% 

ATL 

$3,350,000 

90.4% 

CAR 

$2,000,000 

90.4% 

TOR 

$1,375,000 

90.4% 

CHI 

$7,700,000 

90.3% 

NYI 

$4,500,000 

90.2% 

TBL 

$950,000 

90.1% 

OTT 

$6,450,000 

90.0% 

DAL 

$6,700,000 

89.9% 


 

I got to admit it, I'm at a total loss for the implications of this post. If goalies are not paid on save %, what are they paid on? Guess what it is sports fans: The winning % of the team when they play (correlation coefficient of +0.54 of goalies that play over 30 games). As we've covered, a goalie is going to win lots of games if he has a strong defence and a strong offence in front of him. Even if he can stand on his head, he'll lose lots of games if he's getting shelled. That compensation system can clearly be improved. Note to Toronto Maple Leaf GM: Go grab some cheap high save % goalie in the offseason and watch the wins rack up. Don't worry, no other GM will see it coming.


 

Caveats:

  • The 2008-09 individual salaries are not yet available from USA Today. The 2007-08 figures were used instead as a "best guess." Its annoying but its the best I could do with the data available.

Fighting and team performance in the NHL: No relationship on the scoreboard or the financial statement


 

There has been some long overdue evaluation about fighting in the NHL recently. I suspect part of this self-reflection has been spurred on by internal legal counsel. But there is some conversation about the role of fighting in hockey. Is there a relationship between fighting and regular season performance? Is fighting tolerated to boost fan interest which translates into more money for teams? Let's take a statistical valuation approach and stick with the team overall +/- as the best indicator of team performance:

The figures are taken from the 2008-09 season a little past the half-way point (the games played are between 46 and 49):

Team

Major Penalties

+/-

Boston Bruins

27

64

San Jose Sharks

24

50

Chicago Blackhawks

39

41

Detroit Red Wings

7

36

Nashville Predators

38

25

Philadelphia Flyers

52

16

Washington Capitals

16

15

Montreal Canadiens

25

15

Minnesota Wild

28

8

Calgary Flames

40

8

Vancouver Canucks

51

5

Florida Panthers

22

1

Pittsburgh Penguins

27

-2

Buffalo Sabres

22

-2

Anaheim Ducks

47

-2

Columbus Blue Jackets

34

-3

New York Rangers

34

-4

Edmonton Oilers

39

-6

Los Angeles Kings

27

-11

Phoenix Coyotes

32

-12

Colorado Avalanche

30

-15

Ottawa Senators

21

-18

Carolina Hurricanes

17

-19

Tampa Bay Lightning

24

-19

St Louis Blues

46

-19

Dallas Stars

31

-22

New Jersey Devils

28

-22

Toronto Maple Leafs

28

-29

Atlanta Thrashers

33

-30

New York Islanders

27

-49


 

There is no relationship between the performance of a team of a team and major penalties (the correlation coefficient between majors and +/- is -0.03). That might of surprise to some but maybe not. Consider which players seem to sit in the press box when the playoffs come.

So if fighting has nothing to do with the on-ice performance of a team, is it used deliberately to manufacture fan interest with financially struggling US teams? Is there a relationship between the financial performance of a team and the use of fighting? Let's take a look at both:

Country

Average count of fights per team

Canada

34

USA

30


 

So it isn't US teams driving up the fights.

If we use the Forbes financial evaluation of teams and compare that to fighting majors, we can explore the relationship between fighting and the financial condition of the team:

Team

Value (Mill)

Revenue (Mill)

2008-09 Major Penalties

Toronto Maple Leafs

$ 448

$ 160

28

New York Rangers

$ 411

$ 137

34

Montreal Canadiens

$ 334

$ 139

25

Detroit Red Wings

$ 303

$ 110

7

Philadelphia Flyers

$ 275

$ 102

52

Dallas Stars

$ 273

$ 105

31

Boston Bruins

$ 263

$ 97

27

Vancouver Canucks

$ 236

$ 107

51

Colorado Avalanche

$ 231

$ 91

30

New Jersey Devils

$ 222

$ 97

28

Minnesota Wild

$ 217

$ 94

28

Los Angeles Kings

$ 210

$ 91

27

Ottawa Senators

$ 207

$ 96

21

Chicago Blackhawks

$ 205

$ 79

39

Calgary Flames

$ 203

$ 97

40

Anaheim Ducks

$ 202

$ 90

47

Tampa Bay Lightning

$ 200

$ 84

24

Pittsburgh Penguins

$ 195

$ 87

27

San Jose Sharks

$ 179

$ 85

24

Edmonton Oilers

$ 175

$ 85

39

Buffalo Sabres

$ 169

$ 76

22

Carolina Hurricanes

$ 168

$ 75

17

Nashville Predators

$ 164

$ 70

38

Florida Panthers

$ 163

$ 74

22

St Louis Blues

$ 162

$ 73

46

Washington Capitals

$ 160

$ 73

16

Atlanta Thrashers

$ 158

$ 70

33

Columbus Blue Jackets

$ 157

$ 71

34

New York Islanders

$ 154

$ 64

27

Phoenix Coyotes

$ 142

$ 68

32


 

There is no relationship between the financial value of a team and fighting nor is there a relationship between revenue and fighting. So teams that are in financial distress are not resorting to fighting to boost local interest.

The table below compares the year-to-year correlation coefficient relationship for major penalties by team:

 

2005-06

2006-07

2007-08

2008-09

2005-06

  

0.43

0.39

0.48

2006-07

  

  

0.44

0.54

2007-08

  

  

  

0.72

2008-09

  

  

  

  


 

What this means is that there tends to be a strong relationship between 2007-08 major penalties and 2008-09 penalties. While not a given, teams that fight one year tend to fight the next.

OK, so if fighting doesn't help a team win on the ice or in the box office, what is going on here?

From the NHL perspective, it is really unclear as to why fighting is essentially encouraged. It could be eliminated over night with a few rule changes without any obvious impact at the team performance or local level. Is there a concern that in eliminating fighting that the NHL television ratings would be negatively impacted? But if it is true that the TV viewers are deciding whether to tune in or not based on the expectation of a fight, then this would be a legitimate financial consideration. But i don't have access to TV viewership and the % of highlight content devoted to fights. Does the ESPNs/TSN/SportsNet devote a disproportionate amount of their air-time highlights to fighting? I don't know, but that's the only scenario that makes sense to this Hockey Outsider.