Home Unfiltered Articles Stats Glossary
Baseball       
Hockey Prospectus home
Click here to log in Click here for forgotten password Click here to subscribe

2013 NHL Entry Draft - Top draft prospects list and analysis

<< Previous Article
NHL Playoffs, Second R... (05/14)
<< Previous Column
Howe and Why (05/08)
Next Column >>
Howe and Why (05/14)
Next Article >>
Howe and Why (05/14)

May 14, 2009
Howe and Why
Keeping It Out

by Robert Vollman

Printer-
friendly
Contact
Author

A comprehensive defensive statistic is the Holy Grail of hockey statistical analysts. Who among us hasn't wanted to balance the scale of endless offensive statistics with the occasional metric that measures the equally important defensive contributions? Defensive statistics are essential to comparing the value and contributions of great defensive players to the great offensive players.

Take, for instance, SDP, the Shut Down Player stat we recently introduced. It has the advantage of being simple: a player's even-strength goals against average, penalized by 0.2 goals for each minor penalty he took, combined in a 4-1 ratio with his short-handed goals against average. Ignoring the relatively small sample size of the playoffs, let's take a closer look using this past regular season instead. According to the SDP statistic, the best defensive player in the NHL in 2008-09 was Ryan Jones of Nashville, against whom only 11 goals were scored in 46 games.

Legend:

GP: Games Played

ESGAA: Even-Strength Goals Against Average

SHGAA: Short-handed Goals Against Average

MP: Minor Penalties

SDP: Shut-down play, read it like a goals-against average

Player             Team      GP ESGAA SHGAA MP  SDP

Ryan Jones         Nashville 46  0.92  0.00 11 0.88
Scott Walker       Carolina  41  1.42  0.41 17 1.39
Andrew Murray      Columbus  67  1.15  3.79 13 1.79
Milan Michalek     San Jose  77  1.46  3.59 21 2.02
John Mitchell      Toronto   76  1.75  2.40 27 2.05
Aaron Johnson      Chicago   38  1.29  4.88  6 2.13
Mathieu Dandenault Montreal  41  1.69  3.51  6 2.19
Mike Weaver        St. Louis 58  2.02  2.51  6 2.19
Mark Stuart        Boston    82  1.57  3.81 23 2.21
Brett Festerling   Anaheim   40  1.05  6.29  9 2.25

(Minimum 20 GP, 1 MIN SHTOI/GP, 10 MIN TOI/GP)

There is absolutely no question that SDP has isolated ten talented defensive players, especially the shining defensive light of the Toronto Maple Leafs, John Mitchell. The list also points out some flaws in the SDP statistic, namely that it is highly dependent not just on the defensive contributions of the player himself, but also on several other factors, including:

  • The defensive abilities of his linemates

  • The offensive capability of his opponents

  • The goaltender's ability to stop pucks

  • The role he is asked to play

Mark Stuart, Ryan Jones and Mathieu Dandenault, for example, are not entrusted to defend their teams against the best opposing players, but instead find themselves facing second or third lines for the most part. While Milan Michalek had an excellent season defensively, no doubt his SDP would have suffered had he been relegated to the third or fourth line himself.

With offensive statistics, we're always careful to compare players in similar roles, and that's exactly what we can do here. Gabriel Desjardins keeps track of the Quality of Competition a player faces, and the Quality of his Teammates, both of which are determined by the plus/minus statistics of their opponents and linemates respectively. Now we can compare apples to apples, and find the best defensive player within a certain subgroup. For instance, if we examine only players who faced above-average competition playing alongside below-average linemates, who would have the best SDP?

Player           Team        GP ESGAA SHGAA MP SDP

Andrew Murray    Columbus    67  1.15  3.79 13 1.79
Mike Weaver      St. Louis   58  2.02  2.51  6 2.19
Brett Festerling Anaheim     40  1.05  6.29  9 2.25
Jiri Novotny     Columbus    42  1.48  5.13  7 2.35
Wayne Simmonds   Los Angeles 82  2.00  4.39 19 2.66

Among these particular types of players playing this particular type of role, Columbus' Andrew Murray was most effective. Teammate Jiri Novotny also makes the list, as does Wayne Simmonds of the Los Angeles Kings. Using SDP in context, we can get more value and insight.

Still, the fact remains that goals weren't scored against certain players while they were on the ice, and may very well have played just as well in other situations. The entire basis of this study is on the presumption that measuring the number of goals against tells us a lot about a player's defensive abilities. Even if it is low largely because they were playing a defensive role, or a defensive team, with defensive linemates, well that still tells us something about the coaching staff's opinion of their defensive abilities, relative to the other available players. If that's true, then we can look at the combined even-strength goals against average of each NHL player for every game since the lock-out, and the best defensive players should be at the top. Let's take a look.

Player                 Team       GP ESGAA

George Parros          Anaheim   177  1.28
Ole-Kristian Tollefsen Columbus  140  1.43
Matt Ellis             Buffalo   115  1.44
Petteri Nokelainen     Anaheim   107  1.46
Andrew Murray          Columbus  106  1.51
Kamil Kreps            Florida   156  1.61
Chris Chelios          Detroit   168  1.65
Brett Lebda            Detroit   217  1.68
Derek Boogaard         Minnesota 133  1.69
Johan Franzen          Detroit   212  1.70

George Parros? Derek Boogard? It may seem like even-strength goals against average isn't useful to rate players defensively after all, or is it? Goals aren't scored while they're on the ice because generally they're put on the ice only briefly and rarely, to serve a single purpose, and only when it is safe to do so. Even then, they generally take penalties, which counts against them big time with SDP. The rest of this list is populated by sound defensive players, who aren't known for making very many mistakes.

By no means has SDP achieved the Holy Grail of defensive statistics. Perhaps that can only be achieved by directly applying the quality of competition and teammates to remove that factor, and studying the quality of the shots taken rather than goals against to remove the influence of the goaltender. Even after all that effort, the numbers won't reflect the fact that players asked to play defensively are going to wind up with better results than those required to contribute offensively, too.

In the end, SDP strikes a certain balance between simplicity of the calculation without sacrificing too much of the insight. Thanks to SDP, we can finally give players like Ryan Jones, Scott Walker and Andrew Murray their due. They may have only combined for 20 goals, but in hockey, at least half the trick isn't putting the puck in, but keeping it out.

Robert Vollman is an author of Hockey Prospectus. You can contact Robert by clicking here or click here to see Robert's other articles.

0 comments have been left for this article.

<< Previous Article
NHL Playoffs, Second R... (05/14)
<< Previous Column
Howe and Why (05/08)
Next Column >>
Howe and Why (05/14)
Next Article >>
Howe and Why (05/14)

RECENTLY AT HOCKEY PROSPECTUS
NHL Playoffs, Stanley Cup Finals: Chicago Bl...
Top 100 Draft Prospects 2013: 91-100
Premium Article Conn Smythe Watch: Rask Leads Race
Top 100 Draft Prospects 2013: 81-90
Top 100 Draft Prospects 2013: 71-80

MORE FROM MAY 14, 2009
NHL Playoffs, Second Round: Capitals-Penguin...
Numbers On Ice: Being Conn Smythe
Howe and Why: Sudden Death Overtime
NHL Playoffs, Second Round: Red Wings-Ducks,...

MORE BY ROBERT VOLLMAN
2009-05-20 - Howe and Why: Offense Generators
2009-05-15 - NHL Playoffs, Third Round: Detroit Red Wings...
2009-05-14 - Howe and Why: Sudden Death Overtime
2009-05-14 - Howe and Why: Keeping It Out
2009-05-08 - Howe and Why: Shutting Them Down
2009-05-07 - Howe and Why: Rating The Postseason Goaltend...
2009-04-30 - NHL Playoffs, Second Round: Vancouver Canuck...
More...

MORE HOWE AND WHY
2009-05-28 - Howe and Why: Offensive Contributors
2009-05-20 - Howe and Why: Offense Generators
2009-05-14 - Howe and Why: Sudden Death Overtime
2009-05-14 - Howe and Why: Keeping It Out
2009-05-08 - Howe and Why: Shutting Them Down
2009-05-07 - Howe and Why: Rating The Postseason Goaltend...
2009-04-29 - Howe and Why: Using Playoff Scoring To Predi...
More...