Analyzing the Toronto Maple Leafs Through the Neutral Zone

If nothing else, Mike Babcock brings structure with him to a franchise devoid of it for over a decade.

That’s meant changing the way this team approaches their offence, defence and transition. The Leafs don’t swarm the puck carrier in the defensive zone, and their breakout is much more sophisticated than glass-and-out. Offensively, they’re much less rush oriented and tend to cycle the puck more than in season’s past.

Observing minor tweaks in the neutral zone is nowhere near as easy. The pace is quicker, there’s an awful lot more noise and making sense of it all can prove to be an exhausting exercise. Its importance shouldn’t be understated, though.

I’ve been trying to bridge the gap this season with a neutral zone project geared towards tracking entries. My goal is to get 20-games for each team so that my data might carry some level of predictive value. I’m not there yet – not with most teams – but I’ve got 13-games for the Leafs and results to share.

It’s often said that a “new” or untested statistic has to confirm what we already know, as much as it challenges previously conceived notions about the players involved. The Maple Leafs are an excellent example of that marriage between traditional and analytics-based approaches meeting halfway.

Daniel Winnik, a traditionally strong possession player, leads the charge with 84 entries total. Just as impressive given his neutral zone contributions as a whole, Winnik has just six failed entries. Winnik’s 6.7% failed entries rate indicates he does as good a puck protecting the puck as he does driving play into the offensive zone. The Washington Capitals got a damn good player when they acquired Winnik near the trade deadline.

James van Riemsdyk isn’t far behind Winnik. van Riemsdyk has a high entry total, one behind Winnik at 83, but controlled entries account for 34 of those, so it’s fair to suggest van Riemsdyk as marginally ahead.

Defencemen are rarely showcased by zone entries. You shouldn’t place a tonne of emphasis on zone entries when evaluating defencemen. If a defender shows well by this metric, hell, that’s a bonus.

That makes Morgan Rielly’s high entry totals the cherry on top. Rielly leads all Leafs defenders with 49 successful entries – Matt Hunwick is second to Rielly, 14 entries behind at 35.

Raw totals are the starting point. They might not paint the fullest or fairest picture, but they’re not nothing, either. They’re nowhere near as indicative as rate production, though.

Winnik (35.6 successful entries per sixty) and van Riemsdyk (33.18 successful entries per sixty) are leading the charge, again. Although van Riemsdyk’s controlled entries rate of 13.59 trumps Winnik’s 11.89, the latter is generating shots off his entry at much higher rate – 19.9 shots on entries to van Riemsdyk’s 16.3. 

Byron Froese makes an appearance near the top in most of the metrics included – fourth in entries per sixty (28) and sixth (10) in shots per sixty on entries. Have to imagine Froese’s shot data is being significantly impacted by Quality of Teammate, too. 

For clarity’s sake, when I say “shot data” what I really mean is Fenwick data – unblocked shot attempts. The drawback to this statistic is that it can be subject to huge swings in either direction over the course of a small sample. Frankie Corrado, for example, is likely less dynamic than his shot data might indicate on a per entry basis.

If you’re noticing a recurring theme, it’s that Winnik is showing very well no matter the metric. Winnik leads the Leafs in shots per carry-in (1) and is second in shots per entry (0.56), among regular skaters. Jake Gardiner is first on the Leafs in shots per entry, with 0.64. 

Neutral zone burden percentage aims to take some of the guesswork out of the equation with regards to who’s doing the legwork on entries. The way it’s calculated is by finding out which percentage of on-ice entries are done directly by that player. So if a player has 15 entries, but is on the ice for 30, his neutral zone burden% will be 50%. 

Interestingly, Winnik’s accolades using raw and rate statistics are muted by this metric. So while it appears as though Winnik is a high-volume neutral zone event player, his impact on the Leafs ability to transition play might be getting overstated by these metrics. 

Joffrey Lupul is on the other end of the scale. While his raw and rate neutral zone statistics indicate he’s an average player between the blue lines, NZB% indicates that might be an effect of QoT. Lupul accounts for north of 30% of his team’s on-ice entries.

Well, if you’ve ever taken the speed of the game for granted, consider this for a second. The Maple Leafs have nearly as many successful entries as they have even strength ice-time played – 587 entries to 595 minutes played.

If you’re wondering how there can be so large a disparity between the per sixty and per entry statistics, with the controlled entry data especially, it’s simple. The difference in volume between controlled and uncontrolled entries is huge – 368 uncontrolled entries to 219 with control.

We care about neutral zone data because it impacts your ability to produce favourable shot and goal differential. Studies have found that controlled zone entries (where the puck is either skated in or passed into the offensive zone) produced twice the shot volume of uncontrolled entries (dump-and-chase plays, broken plays or errant passes that still enter the zone). Finding out which players are driving these results is of the utmost importance.

The Maple Leafs are a franchise in transition, but there are encouraging systemic results found in their neutral zone data. Although one would hope for a more evenly distributed ratio of controlled entries (which account for 37% of the grand total) it’s encouraging that they’re producing so high a volume of shots on controlled entries.

If you have any questions pertaining to the data I’ve displayed, please, feel free to reach out by email or in the comments section. I’ll do my best to get back to you. Also, you’ll find links to a better, larger picture in the tweets below. I realise the pictures posted aren’t the easiest to read.

Games tracked for this data: