Defense and the Power Play

As part of the discussion after Joe's X-Factor piece on Michael Del Zotto, regular commenter NTB posed the following question:

What I want to know is how much room for improvement on the PP there really is? What rate do the top PP D-men score at? That is the comparison that is most meaningful, in my opinion.

Strangely, this peaked my interest.  Sure, it's easy to look up the numbers and find Mike Green's 98 points lead all defensemen over the past three years, but is that only because he logs tons of ice time?

Let's start by looking at the league wide averages for three years (this data is for 5v4 powerplays only):

Some clarification here, a legend if you will. The first column designates 4 categories of defensemen.  League is obviously league wide.  20, 40 and 60% indicates only those defensemen whose ratio of TOI/60 ON versus Total TOI was greater than the respective numbers.  For instance, MDZ was on ice for 3.77 min per 60, and off for 1.92 per 60.  So his ratio was 3.77 / (3.77+1.92) = 66.3%.  IPP% is individual point percentage, measured by the number of points earned divided by the number of goals scored while on the ice.

If you read my post on even strength shooting, you saw that league-wide there is not a lot of variation in skill over the last couple of years.  Despite the smaller sample size, we see the same trend here.  There's a small uptick in the last couple of seasons, but mostly the talent has stayed level. This gives us a solid baseline to go on.  A team's defensive corps should be producing about 3.2 p/60.  The team's primary PP defensemen should be around 3.8 p/60.  Let's look at those "quarterbacks" on each team last year:

The table is sortable by clicking the headers of each column.

For the purpose of this chart, I used the leading power play point scorer on this team. Lubomir Visnovsky was used for Edmonton, despite his trade midseason His numbers here reflect his entire season with EDM and ANA. For Phoenix, Ed Jovanovski and Keith Yandle were tied with 14 points, but Yandle was picked because of his slightly better TOI%.

As evidenced here, Del Zotto was trusted as much as anyone in the league to "run" their teams PP, yet his rate stats are mostly middle of the road with this group, and only slightly above the average expected of a top unit player.  He still falls short of some secondary options as well, like Kimmo Timonen in Philly, or Alexander Edler in Vancouver.  Nonetheless, at age 19, this is a rather encouraging result, and suggests there is in fact still room for him to improve his output, especially with new addition Frolov added to the forwards.

Of course, it should also be noted that with just over 300 minutes of ice time, this is not a particularly large sample, and should be taken very lightly when being used to compare players.  A perfect example is the p/60 leader here, Marc-Andre Bergeron.  Fans might look at his 19 PP points last year and see a solid FA target.  GMs haven't been so easily fooled by a player that in two previous years, logged more ice time, and posted p/60 rates at half this rate.

Finally, lets see how the Rangers team defense did:

Though the individual player data doesn't give much insight, the team totals are a bit interesting.  The defense itself was below average in their point production, but those points constituted a slightly above average percentage of the team goals.  How that could be interpreted is left up to debate, but it would seem that the best way to improve our PP would be to get someone to help move the puck on the second unit.  At this point, one has to hope it comes from within, with Girardi getting more time, or Staal developing his offense further.

stats credit as always to the incomparable behindthenet.ca