Do “Defensive Defensemen” Exist? … Part 3: Possession Data & Conclusion

Reaching back to Part 2 (viewable here; see this link for Part 1’s original hypothesis), we’ve established that defensive defensemen, when compared to the other defense groups I’ve categorized (offensive, balanced, replacement), exhibit above-average shot blocking. But even beyond that, are they able to meet their assignments and move the puck in the right direction?

Note: Once again, issues with the interactive chart; here’s the data:

Player OZ% Exp CF% CF% On CF% Off
Offensive D 0.536 0.512 0.514 0.486
Defensive D 0.456 0.477 0.481 0.510
Balanced D 0.492 0.493 0.513 0.501
Replacement D 0.518 0.504 0.499 0.505

Using a combination of expected Corsi-For% (CF%; Corsi-For divided by the sum of Corsi-For and Corsi-Against) given a particular percentage of offensive zone starts (OZ%) and on- versus off-ice CF%, we can get a pretty solid comparative of these players versus their counterparts. And wow, do these defensive defensemen get saddled with rough zone starts…yet they outperform their expected CF%. Keep in mind, also, that the shot-blocking ability demonstrated in Part 2 means that a greater proportion of their Corsi-Against are rendered harmless.

Unfortunately, I don’t really have a way to capture the OZ% for when the player’s off the ice, but 45.6% is very low, so you’d have to expect their teammates get at least 48% on a bad team, and up to 60% on a good team. Split that down the middle and you have 54% OZ% which has an expected CF% of 0.51 or so and, wouldn’t you know it? That happens to be the CF% when defensive defensemen are off the ice. By outperforming their expected CF%, and allowing their teammates better zone starts which return better possession, these defensive defensemen are definitely doing something important. Offensive defensemen, to their credit, are also outperforming their expected CF%, and keep in mind they are also shooting better than any other group on there. So they are also contributing quite well.

And then there are the balanced defensemen, who are understandably the best of the bunch, and greatly outperforming their expected CF%. From one year to the next, I wouldn’t quite get 20 of this type of defensemen, so it’s clearly some pretty special company they keep.

One final thing would be to break CF% into Corsi-For and Corsi-Against to see if there’s any reduction/boost in either metric:

Note: data:

Player Exp CF/60 On CF On/60 CF Off/60 Exp CA/60 On CA On/60 CA Off/60
Offensive D 52.6 54.5 50.2 50.1 51.5 53.2
Defensive D 48.9 49.5 52.8 53.8 53.5 50.8
Balanced D 50.8 53.5 51.6 52.2 50.8 51.4
Replacement D 51.9 50.6 52.3 51.0 50.8 51.3

As with before, we’re seeing the big three groups outperforming their expectations, suggesting that coaches and scouts are indeed typically picking the right people (with there always being exceptions, of course). Those who have been paying attention to the replacement group this whole time will be able to tell you that they are more often than not the underperforming group.

Some final observations:

  • An interesting table to consider for these groups:
    Player Age
    Offensive D 27.5
    Defensive D 30.0
    Balanced D 29.0
    Replacement D 26.1
  • Players moved between these groups, often in accordance to their abilities and age. Older players that were perhaps not playing well in all-minutes deployment would be channeled towards their offense/defense leanings. Sergei Gonchar, for instance, was moved towards an offensive role, while Ron Hainsey was moved towards a defensive role. Young player progression was almost always from offensive to balanced groups, or replacement to offensive. That can certainly explain the table I just shared.
  • Among the 100 offensive and 100 defensive defenseman seasons I observed, players were able to play more than one season in those groups 28 and 26 times, respectively. Given the returns we saw from both groups, it sure seems that, contrary to popular belief, NHL teams are doing quite well in identifying these abilities. Whether that has anything to do with the statistics developments of the last decade, I’m not sure, but it certainly could.

Finally, though I went into this research thinking I would find that defensive defensemen were hardly even an identifiable category, even less a valuable one (my original title for these posts was “There Are No Defensive Defensemen”), what I’ve found has suggested the complete opposite. Defensive defensemen are identifiable, even by team management, they can outperform expected possession given their deployment, and their shot-blocking ability further reduces the threat of Corsi-Against. In my mind, the age table is maybe the most instructive of the bunch: experienced players that cannot (or can no longer) contribute offensively can help in supportive, damage-control roles. I suspect zone-entry and zone-exit research can lend more nuance to this conclusion, but from this data defensive defensemen do exist, they provide tangible benefits in tough deployment, and a good number of teams have been able to identify these players.




Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s