clock menu more-arrow no yes mobile

Filed under:

Philadelphia Flyers zone entries, part 4: Team-level results

Sean Couturier was 17 years old when Ilya Kovalchuk signed a 17-year deal.
Sean Couturier was 17 years old when Ilya Kovalchuk signed a 17-year deal.

We previously gave an overview of the zone entry project, and looked at what the data tells us about individual puck-handling and off-puck results.

In this article we'll look at things at the team level, assessing the team's performance as a whole.

The raw numbers

Let's start with some raw numbers; I'll cook them in a bit. Here's the number of shots generated from each type of zone entry at 5-on-5:

Carry-in Pass-in Dump-in Deflect-in Misc other
Flyers 0.558 0.554 0.244 0.251 0.549
Opponents 0.590 0.570 0.249 0.270 0.505

This tells us a few things:

  • The Flyers were generally not as efficient in the offensive end as their opponents -- with the exception of the "other" category, they gave up more shots than they got from each type of entry.
  • Possession is everything. If you retain possession as you cross the blue line (carry or pass), you get more than twice as many shots as if you play dump and chase.
  • It's been suggested that dump-in numbers may be dragged down by the instances where a player dumps the puck because the defense has played well and he has no good choices so he dumps it in. However, the deflect-in plays are only very slightly more effective, and those are generally dump-and-chase by design.

And here are the numbers on how many of each type of entry there were at 5-on-5:

Carry-in Pass-in Dump-in Deflect-in Misc other
Flyers 1978 648 1502 471 379
Opponents 1859 556 1495 422 370

Again, some observations:

  • The Flyers won the neutral zone more often than their opponents did, getting more of each type of entry.
  • The Flyers were able to retain possession on more of their entries than the opponents did (52.8% of the Flyers' entries were by carry or pass, compared to 51.4% for the opponents).
  • Geoff recorded an awful lot of events, and deserves an awful lot of appreciation.

Zone scores and score effects

The zone entry data allows us to separate out components of shot differential that come from offensive zone, neutral zone, or defensive zone play.


The first table of this article allows us to calculate offensive zone and defensive zone scores, and the second table allows us to calculate a neutral zone score. The score answers the question "how did the team's performance in that zone affect their shot differential".

For example, to work out the offensive zone score, you plug in the team's actual shots per entry with possession, their actual shots per entry without possession, and the average across all teams for all of the other numbers. This lets you work out what their shot differential would have been with average performance in the neutral and defensive zones, which is what I'm calling the offensive zone score.

The result is as follows, broken down by game state:

Big Lead -0.4% -1.7% +0.4%
Close -0.7% +2.3% -0.7%
Big Trail +1.0% +5.1% -1.3%

The OZ and DZ scores are relative rather than absolute (see appendix for more on what this means), but the following seem to be true:

  • The Flyers had a big net plus in the neutral zone, but gave back about half of that edge in the attack zones.
  • When score effects come into play, the trailing team does better than usual in the neutral zone and their offensive zone, but suffers in their defensive zone as a result of their aggression.
  • I like having three bullet points in each group.

Head to head

We can also look at the Flyers' head to head results against various opponents. They played 4-6 games against each conference rival, which isn't enough to draw firm conclusions, but it can be interesting to think about and the large sample size of the zone entry work (~188 events per game) means that even these short series may have some utility.

Here's how the Flyers performed against each team in the Eastern conference this year (score-close only to minimize score effects):

Opponent NZ% OZ% DZ%
Rangers +5.9%
Maple Leafs

The bullets:

  • The Flyers skaters had great results against the Rangers and terrible results against the Penguins, yet the scoreboard had a different story. This is because Marc-Andre Fleury's best performance in his five meaningful games against the Flyers was a .897 save percentage, while Henrik Lundqvist posted a .942 save percentage in six games against the Flyers. Over a limited sample size, goaltending variance can swamp shot differentials.
  • While we're on the topic of goaltending, allow me to call attention to the Flyers' -6.5% offensive zone score against the Devils. This is in part because they got 21% fewer shots per dump-in against the Devils than they did against the rest of the league, which may be a testament to Martin Brodeur's puck-handling skills.
  • Remember when the Flyers were protesting Tampa Bay's passive neutral zone trap? The Flyers pretty well owned the neutral zone against that defense they were complaining about.

Geoff also tracked odd-man rushes, initially so that we could answer concerns that the benefits of carrying the puck in might be inflated by odd-man rushes. As it turns out, there were few enough of them that their impact is minor -- only 296 of the 9680 even-strength entries were odd-man rushes.

However, it is interesting to note that the Flyers generally gave up a lot more odd-man rushes than they got, particularly against certain opponents. They had more odd-man rushes than only two of their conference rivals, Tampa Bay (8-7) and Buffalo (11-10). Yet several teams got far more odd-man rushes than the Flyers did, including New Jersey (5-12), Montreal (7-13), Carolina (2-8), the Islanders (7-16), and Pittsburgh (9-17). The Flyers' aggressive play does come at a cost.


Overall, the Flyers were very strong through the neutral zone this year, but got outplayed by a bit in the attack zones. Score effects show up in all three zones, but most prominently in the neutral zone. And finally, we took a look at how the Flyers did against each of their conference rivals.

Appendix: a limitation in the data

Suppose I told you that this year, at 5-on-5, the Flyers got 0.56 shots per time they carried the puck in, but their opponents got 0.59 shots per carry-in. This tells you that the Flyers' opponents were a bit more efficient than the Flyers were in the offensive zone, but it doesn't tell you why -- for that, we'll need to have more teams tracked.

If we knew that the league average was 0.50, we could say that the offense did very well and the defense did terribly. But we don't know that, since only Flyers games have been tracked so far; maybe the league average is 0.60 and the blame should be placed on the offense.

We can still calculate solid numbers for the team's neutral zone performance, since that's an internal comparison of whether they got more entries than their opponent rather than whether they got more than the league average. But for the attack zones, we've focused here on relative comparisons: even if we can't say whether they were good in the offensive end, we can still look at whether they did better in some games than others. So there may be a future adjustment moving all of the offensive zone scores up (or down), but the relative differences should hold constant.