Philadelphia Flyers zone entries, part 2: Individual puck-handling

In part one of our year-end review of zone entry data, we gave an overview of the kinds of analysis made possible by the zone entry tracking and promised more detail would follow.

In this article, we continue with a look at individual puck-handling contributions. There are two basic questions we hope to answer: who was most involved in bringing the puck into the offensive zone, and how successful were they at it?

Before we get into the data, we need to decide which metrics are worth looking at. We calculated dozens of statistics using our zone entry data, but it isn't necessarily all valuable. A key question for any statistic is whether it measures an actual talent that the player has; we don't want to over-interpret something that could just be random noise.

A simple way to test that is to compare how players did in odd-numbered games to how they did in even-numbered games. If I give you a player's score in some metric for odd-numbered games and you can't make a reasonable guess at how he did in the even-numbered ones, then that metric probably doesn't mean much about the player's specific skills. We measured how well the odd-numbered games predict the even-numbered results for every statistic we calculated, and for the most part we'll only talk about the ones that proved to be correlated beyond what you might expect from a random sample of this size.

As it turns out, that means we aren't going to talk much in this article about what happened after the player brought the puck out of the neutral zone. When we started this work, I guessed that how often the team got a shot after a given player carried the puck into the offensive zone might be an indication of his puck-handling skills, but as it turns out, the odd/even correlation for shots per carry-in is tiny.

After we filter out the assessments that fail this test, we have three metrics left to evaluate individual puck-handling contributions:

• Entries per 60 -- how often did that player send the puck into the offensive zone?
• Individual entry percentage -- of the team's entries with the player on the ice, how often was this player the one sending the puck in?
• Percent of entries with possession -- of the times that the player sent the puck in, how often did he retain possession (carry or pass) rather than dumping the puck in? (Remember, entries where the team retains possession generate twice as many shots as dump-ins, so this is a measure of how effective the player's zone entries were.)
  

Obviously the first two are related, but they aren't quite the same. Entries per 60 is basically the product of two separate questions: how often did the team move the puck into the zone with the player on the ice, and what percentage of those entries involved that individual player.

Since each of those skills can be measured separately, I will generally not use a player's entries per 60 here, which narrows things down to two metrics for individual puck-handling. So let's take a look at how the top-12 forwards (players with at least 300 minutes of 5v5 ice time) did according to these metrics:

A few assorted observations:

• The most effective puck-handlers are able to retain possession most often as they enter the offensive zone. Jaromir Jagr is the clear leader in this area, followed by Claude Giroux, Jakub Voracek, and Daniel Briere.
• Decisions about who should handle the puck are generally efficient, as the players who are most effective with the puck are also the ones who handle it the most.
• One exception to the above is Wayne Simmonds, who was by far the least effective of the top 9 forwards at retaining possession of the puck, yet was the second-most involved. It may not be fair to Simmonds -- perhaps the team asked him to do this or the passes he was fed led to it -- but I picture the guy in your pick-up basketball game who jacks it up from the outside every time he gets the ball. He was on a line with Briere for most of the year, and for whatever reason, Briere ended up with much fewer zone entries than comparable performers (Voracek, Giroux) and Simmonds ended up with a lot more than he should have.
• The fourth line doesn't generate much offense. Sean Couturier, Zac Rinaldo, and Maxime Talbot all dump the puck in much more often than the top 9 forwards do -- the three of them retained possession on 45% of their entries on average, whereas the average for the top nine was 61%.
  
• The fourth line also relies on the defense to handle the puck much more than the top lines do -- the three of them combined handled the puck on 68.2% of the zone entries, whereas the average for the other three lines was 84.8%.
• Couturier notably did not handle the puck like a playmaker. Some of this may be because he was asked to just dump the puck in when playing on the fourth line. However, that does not explain all of it; he retained possession on 41.6% of his entries when Talbot was on the ice with him and 49.1% when Talbot was not. He retained possession on 44.3% of his entries when Voracek was on the ice with him and on 50.0% when Matt Read was on the ice with him. I don't know whether this is something that will improve sharply with time or a reflection of inherent talent limitations, but he needs to get better at it if he is going to become a top offensive threat.

For Jody Shelley fans who are distraught at not having his sparkling numbers, I've attached plots with all of the players at the end of the article. For now, let's move on to the defense:

In terms of how involved in the puck-handling a player is, Kimmo Timonen is in a different league from the rest of the defense, sending the puck in much more often than anyone else.

In fact, the gap is even larger than the plot above suggests -- remember that the fourth line leaned heavily on the defense, so we might expect defensemen who played with Couturier a lot to have their involvement percentage boosted. Yet Timonen spent only 17.3% of his 5-on-5 ice time with Couturier and no other defenseman was with Couturier less than 19.8% of the time; the high was Braydon Coburn at 26.7%.

So despite playing with the offensive-minded forwards more than anyone else, Timonen still managed to get a much larger share of the zone entries and was clearly the most heavily involved in advancing the puck.

The rest of the defensemen all had basically the same entry involvement percentage, but there were real differences in how they brought the puck into the zone. Matt Carle, Andrej Meszaros, and Marc-Andre Bourdon all carried the puck in about as often as a fourth-line forward would, whereas Coburn, Nicklas Grossmann, Andreas Lilja, and Erik Gustafsson were scarcely involved at all. Gustafsson is the one that surprises me on that list; I think of him as being more of a mobile puck-mover than the porch-clearers he's grouped with.

Ever since last year I found that defensemen assist rate has a near-zero year-over-year correlation, I've been looking for a stat that can identify the puck-moving defenseman. These two measures from the zone entry work look like a promising attempt.

Appendix: puck-handling plots for the whole team