Shut Down Defensemen - Part 1

How often have we heard the idea that defensive blue liners are a "dime a dozen" in the NHL? Probably far more than we should frankly. As a skill set, the defensive defender's repertoire is difficult to master at a high level, and in some ways it is even more difficult to quantify effectively. Knowing where to be on the ice, when to be there, and actually being physically capable of playing such a role is extremely demanding. The best defenders take the lion's share of a team's tough minutes, and prevent the best of the opposition from doing much damage offensively. This conveniently allows other less capable defensemen on their team to be "sheltered" from the proverbial storm and makes everyone look better in the process.

Daniel Wagner of the Vancouver Sun has recently been applying this very concept - which he terms 'enabling' - in his breakdown of the Vancouver Canucks usage of 4th line centre Manny Malholtra. He is counted upon to do the dirty work on the ice in the defensive zone so the Canucks more offensively gifted players can spend more time on offense doing what they do best - scoring goals and producing points both on the ice and eventually in the standings. This is basically the same concept I am attempting to get at only for defenders.

So how do we describe the guys that face the toughest minutes? Well generally speaking they'll be counted on to play on the Penalty Kill, and they'll be expected to face the best lines the opposition has to throw on the ice at even strength - for the method I plan on using I'm going to focus on the latter. Using behindthenet.ca we can form a pretty detailed picture of what it takes to be considered a defender in the shut down mold, and even more importantly, who the best players are in that role.

Using the past 4 years worth of data plus what has gone so far this year will allow us to have a relatively robust selection to choose from as well, so we know where we're sitting in terms of recent comparables. For the sake of ease of analysis I made one particularly large assumption, which is that the defenders facing the highest Corsi REL QoC on a team over a large span of games will generally be involved in the top 4 D on their team's PK, and examining the names that come up on the list, this is pretty clearly the case.

As a methodology I structured the analysis in a manner quite similar to the Heavy Lifter Index described by Ryan Poplichak at Hockey Prospectus, which has also been discussed or mentioned by Derek Zona at CopperNBlue and Gabe Desgardins at Arctic Ice Hockey. As I was looking at defenders and not forwards though I decided to change the categories slightly so as to get at what I felt was important.

Firstly, I decided to NOT include +/- QoC as I personally am of the opinion that +/- is an anachronistic statistic that is of little actual value in assessing players due to the large quantity of shooting luck involved season to season. While there is some skill in shooting over the long term, I feel that a defender's key role is to limit goal scoring opportunities which closely correlates to Corsi, thus I stuck with Corsi measures largely.

Thus I did include Corsi REL QoC as in the HLI, but I also decided to factor in Corsi REL QoT. A defenders job is immeasurably easier if the players they share the ice with regularly are more capable of driving play up ice, thus it made sense to me to reward players plying their trade while at a disadvantage and penalize those who are playing with superior line mates. In an effort to manage this fact, I combined Corsi REL QoC and Corsi REL QoT into a differential statistic, comparing the Standard Deviation ratio of both statistics in the given year. Therefore a player who plays against very difficult Corsi REL QoC, while toiling with inferior linemates from a Corsi REL QoT perspective is given a higher index rating.

Secondarily, I included the Standard Deviation ratio of the defender's Corsi ON/60. Unlike in the refined version of Ryan Poplichak's HLI, I felt that D men with negative Corsi ratings could still be present in this index. Given the situations they are placed in and their skill set, I didn't see much point in eliminating them from the analysis.

Thirdly, I included the Standard Deviation ratio of the defender's Offensive Zone Start %, in this case a negative score would be considered a positive contribution to the index. The last category I did include was the Standard Deviation ratio of the D man's penalty differential, but I weighted this statistic lower as I felt defensive D men do tend to rack up penalty minutes at a higher rate than forwards.

So in summary SDI Sit (Shutdown Defender Index Situational) will be the sum of values for Corsi REL Differential STD Ratio and OZ% STD Ratio, while SDI Res will be the sum of values for Corsi ON 60 STD Ratio and the 20% weighted value of Penalty Differential STD Ratio.

Here are the top performers from the past few years in the NHL by season (oh and keep an eye out for Jan Hejda)

2007-08

2008-09

2009-10

2010-11

2011-12 (So Far)

So a couple of thoughts on the lists above: firstly Jan Hejda is easily the most under rated defensive player in the NHL post lockout; secondly while I may need to adjust some of this to prevent the index from penalizing players from good teams, this seems like an excellent tool for identifying potential defensive workhorses on less talented teams. In fact you may notice some UFA targets I have mentioned either on PPP or Twitter for the Leafs to aim for (Jason Garrison and Tim Gleason).

So I'm going to follow up this post tomorrow with a detailed look at how the Leafs D has shaped up over the past 5 years - it paints a picture that goes a long way to explaining their non-playoff results. Luckily though I find some reason for hope, but you'll need to tune in for more on that.

Please discuss your views in the comments and if you have any questions or suggestions I'll get to them when I can.