Zone Shift

Last off-season I posted an analysis of the Leafs individual players based on Zone Shift.  The concept draws on the information from Gabe Desjardins' behindthenet.ca site, specifically Zone Start and Zone Finish locations, with a couple of modifications.

The way that Gabe tabulates Zone Starts is self explanatory - the zone in which a shift starting face-off is located.  This is obviously indicative of what the coaching staff thinks of the player in question.  If a player tends to start a shift in the defensive zone more than the offensive zone, odds are the coaches think he's defensively capable, and if a player tends to start in the offensive zone more often, than the reverse would be true.

Zone Finish is tabulated by Gabe using the location of faceoffs subsequent to any whistle.  Unfortunately all face-offs following a goal for or against are located in the neutral zone.  This introduces a level of discrepancy into the numbers, which I attempt to filter out by adding goals for to the Offensive Zone Finish and goals against to the Defensive Zone Finish numbers.  The sum of goals for and against is then subtracted from the Neutral Zone Finish numbers.  There are also situations regarding stoppages in play due to penalties or caused by the attacking team in the offensive zone where the face-off takes place outside the zone of attack, but for all intents and purposes I consider that a rare enough event that I'll ignore it for the sake of this analysis.

So without further ado, let's take a look at the Leafs numbers from this past season shall we?

And for the Leafs that left through trades during the season we have the following:

The stats I've listed for Qual DIF and CORSI Rel SUM are explained as follows.  The Qual DIF is the difference between the quality of a player's teammates and their opponents over the course of a season.  A negative Qual DIF would imply the team was consistently at a disadvantage with respect to the players he shared the ice with, while a positive would mean the ice was tilted in his favour.  The CORSI Rel SUM is the combined total of CORSI stats relative to teammates and opposition, so a higher combined value indicates better all around players.  The logic being that if you're putting up higher CORSI numbers than expected with respect to your teammates and with respect to the opposition, you're doing something right.

The way that I've calculated Zone Shift was to take the Zone Start numbers and subtract the defensive zone face-offs from the offensive zone face-offs, and use that for a net Zone Start value (positive means the majority were in the offensive zone, negative means the majority were defensive).  I performed the same operation on the adjusted Zone Finish numbers (including goals for and against as offensive and defensive zone finishes), in order to determine the net Zone Finish value.  I then subtracted the Zone Start values from the Zone Finish values in order to determine the Zone Shift. A positive value indicates that a player is generally moving the puck up the ice, as their combination of Zone Finish and Zone Start results generally indicates they're finishing in the offensive zone more than one would expect based on their Zone Start locations.  A negative value indicates the player is generally losing ground as they are finishing in the defensive zone more than one would expect based on their Zone Start locations.

As Gabe has discussed in his blog on the subject, generally speaking the flow of play is towards the neutral zone, as play for most players will end up finishing there far more often than it will start there based on their coaches decisions. 

There is also lost data in the set, as there is no easy way to account for a flow of play without whistles.  If a player joins the game on the fly, it will generally occur while play is moving into the offensive zone.  Similarly, if a player leaves the ice on the fly, it will again, generally occur while the play is moving up ice from the defensive to the offensive zone.  This has the dual effect of penalizing defensive players and ignoring the efforts of a player whose offensive skill allows them to play entire shifts in the offensive zone without a resulting whistle.  Defensive zone players will often begin their shift in their end as a result of their coaches decision, and then leave the ice as the play progresses to the attacking zone.  If the play is finished in the neutral zone or offensive zone, they get no credit for the Zone Shift in the net result we are recording.

Regardless, the stat is still a decent general indicator for all intents and purposes based on the sizable number of results for all players who play regular shifts. So what can we pull from this data set?

1. Matt Stajan was actually having a pretty solid season for the Leafs prior to his departure.  He may not be top line centre material on a winning club, but he was proving to do a decent job as a top 6 forward.

2. Despite the fact that Lee Stempniak was on the ice for 14 more goals for than against when he went to Phoenix, his Zone Shift number is still an atrocious -60.  His time with the Leafs was unfortunate.

3. Despite the fact that Ian White's Zone Shift looks really bad, if you look at his Qual DIF and CORSI Rel SUM numbers, you begin to realize he may have suffered from tougher competition and a lack of support from his team-mates.  In comparison to Tomas Kaberle's numbers from this season, he was actually doing a decent job this season.  Which brings me to point 4.

4. Tomas Kaberle and Mike Komisarek were horrible as a defensive pairing for the Leafs last season.  Both of them had positive Qual DIF values indicating they were playing with superior players on the Leafs.  Komisarek's CORSI Rel SUM value is respectable, but his lack of point contributions definitely hurts the Leafs.  In the end, their Zone Shift numbers indicate a greater problem than we may at first realize.  Kaberle's value may have been hurt at the end of the year, but in this light, it's hard to see how any team can expect an adequate level of defensive play out of him.

5. Phil Kessel, Tyler Bozak, Luke Schenn, and Nikolai Kulemin are all solid contributors for the Leafs.  Despite playing on a weaker Leafs side, often outclassed by the competition, all four performed admirably for players that are 24 or younger.

6. Luca Caputi may have more to his game than I first thought.  If he can put the puck in the net with more regularity I may be happy to see him play more top 6 minutes this coming year than I first thought.

7. Mikhail Grabovski had a bit of an off year this year.  His numbers weren't horrible, but his level of production should have been higher given his CORSI and QUAL numbers.

8. Comparing Francois Beauchemin and Carl Gunnarsson, you begin to realize just how stellar the young Swede was defensively for the Leafs.  He faced the stiffest competition level of any of the Leafs blue liners on a consistent basis, and he still managed to put up a lower Zone Shift value as an NHL rookie.  Comparing Gunnarsson to Komiasarek or Kaberle makes you wonder how hard he'll have to work to remain in the top 4 for Ron Wilson's team going into next season.

9. Garnet Exelby's sheltering was ridiculous.  Looking at his QUAL Dif and CORSI Rel SUM values is freaking scary.  The fact that he had a Zone Shift of only -8 is mildly reassuring, but considering who he was on the ice with and his negative impacts all around I can't say how glad I am he won't be returning next season.

Ok so that's it for now, feel free to post your own comments and let me know what you think of where the Leafs are heading.