The New Jersey Devils vaunted forechecking system made headlines with their 4-1 series destruction of the Philadelphia Flyers in the 2012 Eastern Conference Semifinals. It was a tenacious, swarming system that instantly made the Devils fun to watch. As someone who was really just getting into advanced hockey statistics around this time, the Devils made it fun to learn about puck possession and things like Corsi (all shooting attempts) and Fenwick (all unblocked shooting attempts).
However, with two straight seasons without playoffs, very strong possession numbers, and a scarcity of goals, I’ve often wondered if this system that Head Coach Peter Deboer has instilled in the team one that can have long term success. Is it effective? Efficient? Can teams more easily defend against it?
With my passing and shot generation data from the 2013 – 2014 season, I wanted to see what the numbers said about what the Devils do in the offensive zone and how it contributes or hinders their scoring and attempt at winning. Let’s get to it.
Summer Passing Series Links
Part One: Efficiency and Winning
Part Two: The Transition Game
How does the Offensive Zone Correlate to Winning?
As we can see in the above chart, the offensive zone is a bit of a perplexing area to analyze. On one hand, teams that are more efficient at generating shots off of their passes, which I’ve given the title SAGE (Shot Attempt Generation Efficiency), win 76.8% of their games, second only to overall SAGE, and about 20% better than transition efficiency (Shots generated from the defensive or neutral zones). Now, when we look at which teams are controlling the greater number of shot attempts in each area of the ice the trend is reversed: controlling the greater number of shot attempts in the offensive zone (OZ SAG) has a winning percentage of only 37.8%, nearly 20% lower than when controlling the greater number of transition shot attempts (D/NZ SAG). This is something I looked at in my previous article and will explain the other half of the debate here.
Why this discrepancy? If transition quantity better correlates with winning, shouldn’t the efficiency do the same? The obvious answer is yes, but here’s why I think it doesn’t. Generating a shot attempt in transition has a much greater chance of becoming a shot on goal, so quantity in the offensive zone isn't as inherent a boon as transition quantity.
This chart shows that both the Devils and their opposition generate shots more efficiently in transition than from offense generated within the offensive zone. So, whomever can be more efficient in the offensive zone can greatly influence the outcome of the game. Let’s look at which Devils contributed to this.
Which Devils are most Efficient in the Offensive Zone?
The way to read this is the volume of shot attempts generated is along the x-axis, the percentage of overall SAG production that occurs in the offensive zone is along the y-axis, and the size of each bubble is a player’s OZ SAGE (this percentage is displayed next to the name).
We’ll start with Zidlicky, who was a transition beast, and learn that he simply is not quite the same in the offensive zone. Sure, the volume is there as he nearly lapped the field in shot attempts generated, but there are two knocks against him here: 1) Over 80% of Zidlicky’s SAG contribution is generated within the offensive zone, which we’d like to see more balanced, and 2) His OZ SAGE is only better than Mark Fayne’s.
At the other end of the graph we see Adam Larsson who, apart from low volume, looks great. The more attempts a player can generate in the transition game, the less they’ll generate in the offensive zone, which, as I’ve shown above, are less efficient as a whole. 61.1% of the Larsson’s SAG resulted in a shot on goal, the best on the blue line. Of course, he generated the fewest shot attempts so the sample size isn’t conclusive. Similar to his transition numbers, these are encouraging signs by Larsson.
Also promising is the efficiency of Eric Gelinas and Jon Merrill, who are more balanced between transition and offensive zone production than most other defensemen, as well as more efficient in the offensive zone. Merrill holds the largest sample size of the three young defensemen, but Gelinas isn’t far behind. More promising signs.
Andy Greene was nearly as balanced as Merrill, with about 72% of his SAG production originating in the offensive zone, though he did generated nearly twice as many shot attempts. Greene’s efficiency (47.5%) is not great, but is not bad by offensive zone standards. His partner, Fayne, again is lowest in terms of efficiency, generating more of this empty possession that doesn’t matter. Put another way, if a Devils defensemen was generating a shot attempt from anywhere on the ice, and even more so the offensive zone, Fayne would be the last one you’d want to be passing the puck.
Bryce Salvador, Anton Volchenkov, and Peter Harrold are all bunched up near the left and top portion of the graph, which is where you don’t want to be. Volchenkov actually had a better OZ SAGE than both, which was surprising.
Moving to the forwards, we see Jaromir Jagr and Travis Zajac off on their own again. Similar to Zidlicky, the volume generated in the offensive zone dwarfed most other forwards. Adam Henrique and Patrik Elias were third and fourth in terms of volume, with Henrique the worst in terms of efficiency among the top forwards.
Dainius Zubrus and Steve Bernier were the next two in terms of volume, another strong showing from them after ranking 1-2 in volume of transition shot attempts generated. Michael Ryder was just behind them.
What you’ll notice among all forwards is that not a single one was at 50% OZ SAGE or better. As I showed above, both the Devils and their opponents averaged an OZ SAGE of less than 50%, so that’s not uncommon. In fact, the average OZ SAGE of a team that wins a game is only 49.2%. 50% is a boon for this metric. So, looking just at efficiency, Zubrus leads the way at 49%. Some of you may want to say, "Well, he played with Jagr and Zajac nearly the entire season, so they shoot better, which makes Zubrus look better," which is true to an extent. Neither Zajac nor Jagr managed to be as efficient as Zubrus, though their volume is considerably higher than Big Z. Perhaps the notion that Zubrus makes space for other skaters is evident here: if Zubrus is controlling board play and then makes a pass out into the slot or other open area on the ice, perhaps there is a 3 – 5% better chance of that pass resulting in a shot than one from Jagr or Zajac. Perhaps, perhaps not, but it can’t just be his line mates.
If it were just line mates, what would we say about Andrei Loktionov’s 48.8% OZ SAGE? He was the next best forward, 0.4% better than Patrik Elias. There weren’t many surprises among the rest of the forwards, though Bernier was 9.5% and 13% more efficient than Ryan Carter and Stephen Gionta respectively. Again, I don’t understand why Bernier is the target of such fan vitriol when he’s quite an effective bottom-six forward. Insert obligatory remark dreaming of a fourth line without Gionta here.
How do the Devils Skaters Compare to their Opposition?
We know the Devils opposition is more efficient, but is that due to the play of their forwards or defensemen?
So, while Zidlicky’s volume appears to dampen his efficiency, he’s right on par with the opposition defensemen. You’d like him to be better than average, of course, but considering his volume exceeds most defensemen, I would venture, this is not too bad. Greene looks even better when comparing to the opposition, as his production is just over half that of Zidlicky’s, but a few percentage points higher.
In fact, most Devils defensemen were above the opposition average in efficiency, with only Fayne below it. Another reason I think Devils fan won’t miss Fayne that much is on display here. As Merrill, Gelinas, and Larsson continue to improve, this group could be quite active and productive sooner rather than later.
So, if OZ SAGE is the second best correlation with winning that I have and the Devils defensemen are consistently beating their opposing defensemen…yea, it’s not pretty for the forwards.
Okay, so it’s not the complete tire fire I thought it would be, but it’s not pretty. Zubrus, Elias, Ryder, and Loktionov are the only forwards more efficient than the average opposing forward. Jagr, Bernier, Zajac, and Clowe are close, but not quite. So, there’s eight forwards either just over or just under the average efficiency of opposing forwards. Normally, this may not be that big of an issue, but you’ll see why this is a huge problem. At the beginning of the article I asked about Deboer’s forechecking system and whether it was a hindrance. We’re about to see that it is.
This chart tells us where shot attempts are generated from for both the Devils and their opponents. Now, I’ve already shown you why it’s important to generate shot attempts in transition, and above I’ve illustrated that efficiency within the offensive zone is important as well. The thing about the offensive zone is that it’s a double-edged sword: it is inherently more difficult to generate a shot off of a pass in the offensive zone than it is in transition. Why? The simplest answer is often the best and so I think it’s that fact that there are more bodies between the shooter and the goal.
So, in this chart, we see that the Devils generate 10% more of their offense from within the offensive zone, a zone where it is more difficult to generate a shot and where the Devils are worse off than their opposition. The other side of this is that the Devils opposition generates 10% more of their offense in transition, where efficiency is easier to come by.
The Devils have a higher portion of shot attempts coming from an area on the ice where they will be inherently less efficient, both by their own production and the fact of where that attempted production is taking place. Their system is set up so that it requires them to work harder than the opposition just to be on level footing with their opponents. If this is how Deboer wants them to play, the Devils are stepping onto the ice each night at a disadvantage.
So, What’d You Think?
This was the third article in a summer series looking at all of this data. There will be player analysis, team analysis, and sometimes general rambling by yours truly. Give me your questions, statements, feedback so I can better steer this towards where your interest lies. Sound off below!