Pavel Zacha & the Devils Forwards Were Amazing on the Penalty Kill in 2018-19

Not much went well for the New Jersey Devils last season. They finished 29th out of 31 teams. However, one thing about their performances was far from the bottom of the league: the penalty kill. They finished fourth in the NHL with a 84.3% (215 kills out of 255) success rate and allowed the fourth fewest power play goals (40). That the season went the way it did for the Devils means that a great penalty kill cannot make up for significant issues elsewhere. At the same time, this is something that they can try to sustain as the organization looks to re-tool, re-assess and re-think how they handled business in other situations. In order to know how to keep the relatively good times rolling when shorthanded, the team needs to recognize what made the penalty kill so great in 2018-19. A big part of it was due to the play of the forwards.

The New Jersey Devils’ penalty kill operates out of a base of a wedge-plus-one. Picture three skaters set up in a triangle around the slot with the fourth one, usually a forward, ahead of them. The job of the “plus-one” is to apply some pressure on a puck carrier or look to take away a passing lane high up in the zone. The three in the wedge move to protect the slot, pick up men by the crease, and react accordingly. Should someone in the wedge move away - such as a to engage a player on the side-wall - the plus-one moves back to re-form the wedge. As most NHL teams use a 1-3-1 formation for their power play, this approach to the penalty kill has also become popular. In theory, it keeps the puck to the perimeter of the power play, it defends the scoring chance area, and it allows for an occasional offensive opportunity provided the “plus-one” is fast and aggressive enough to pick off a pass or steal a puck away.

The Devils have utilized this formation as far back as 2014-15 (which I wrote about its effectiveness in length here) and likely even before then as the 1-3-1 formation became more popular for power plays. Even after changes in coaches and management, the Devils kept it in 2015-16 and still use it today. What makes it effective is that it requires contributions from all four players. The forwards have to be aware of not only the play but where their linemate is. The defensemen need to move with the formation and react accordingly. This is not the sort of formation you can just stick four defensively smart players in and hope it will work. It excels when you have an effective forechecker involved as well as forwards who are responsible in their own end. The Devils absolutely had those forwards killing penalties last season. Despite how the rest of the season went, they shined.

The Amazing Forwards & NHL Comparisons - Background

The they in question are as follows, in order of shorthanded ice time per Natural Stat Trick: Blake Coleman, Travis Zajac, Pavel Zacha, Brian Boyle, Kevin Rooney, Nico Hischier, Joey Anderson, and everyone else played fewer than 40 shorthanded minutes total. Coleman-Zajac was often the first choice on the PK; however, Zacha, Boyle, and others stepped in in case the player took a penalty or was not playing that night. This meant the second unit would vary but whoever was involved on a regular basis did not do well - they did really, really, really well. Especially Zacha.

I want to show why I used three reallys and highlighted Zacha. Using the data at Natural Stat Trick, I decided to compare all of the Devils forwards who averaged at least a minute of shorthanded ice time per game against all other forwards in the NHL. This meant I had to remove Hischier in this comparison; but I got to keep Boyle prior to his trade. In the NHL, 179 forwards averaged at least one minute of ice time per game on the PK last season. The thing about PK (and PP) on-ice rate stats is that they’re going to be high in one direction so comparisons to others helps identify what actually is and is not good.

Since this is about penalty killing, all of the on-ice stats are per-60 minute rates against the Devils: Corsi (attempts, CA/60), shots (SA/60), scoring chances (SCA/60), high danger scoring chances (HDCA/60), expected goals against (xGA/60), and goals against (GA/60). I also included individual shots and shooting attempts (iCF) to make an additional point. Ranks in green mean that the player was in the top-90 in the NHL in that category. Ranks in yellow mean that the player was in the top 10% in the NHL in that category, or the top-18. Ranks in bold mean they were in the top-10 straight up. For Boyle, he has an asterisk as he only played part of last season in New Jersey. I individually ranked his stats as a Devil compared with others instead of including his Nashville stats or splitting everyone’s season by team.

The Regular Devils Penalty Killing Forwards Compared with Other Regular Penalty Killing Forwards in 2018-19

Data via Natural Stat Trick

Your eyes are not deceiving you. Pavel Zacha was legitimately one of the best forwards on the penalty kill in the entire league last season. His on-ice rates for CA/60, SA/60, SCA/60, HDCA/60, and xGA/60 were all top-ten rates among 179 other forwards. His on-ice rate of GA/60 and his individual amount of shots were in the top 10% in the NHL too. Yes, even Zacha was creating a relatively high level of offense in shorthanded situations. Zacha was simply stupendous in shorthanded situations.

What’s more is that most of the players who ranked around him in those categories played fewer than two minutes per game on the PK. While Zacha was not usually on the team’s first PK unit, he did manage to play enough such that he would have had to have seen the other team’s primary and usually more dangerous unit. That the team’s on-ice rates were so relatively low with Zacha taking those shifts speaks to his actual abilities in how the Devils kill penalties. I understand that any praise of Zacha needs to be couched with disappointment about how he isn’t as good as Player X from the first round of the 2015 NHL Draft. Yet, but the haters and losers, of which there are many, need to recognize that Zacha was legit fantastic at this part of the game.

Similarly, the Devils received great performances from other secondary and tertiary unit forwards. When Boyle was on the Devils, he was performing on a level similar to Zacha on penalty kills. It was so good that it could not be sustained in Nashville; but he still finished well into the upper half of NHL forwards in all non-offensive categories. Boyle may be a bottom-six forward, but he was a top-class PKer on New Jersey last season.

The Devils also received great results from two (more) unlikely sources: Kevin Rooney and Joey Anderson. The Rooney supporters should be thrilled about this. Rooney did not play in every game but he proved he belonged on the penalty kill when he did. He put up on-ice rates for shots against, high danger scoring chances, and expected and actual goals against rates similar to Zacha. Rooney was in the top 10% in those four categories and in the top ten in the whole NHL in both. And he finished well in the top-half among NHL forwards for shorthanded attempts and scoring chances against. It remains to be seen if Rooney will be a regular Devil but if he is, he absolutely should be a part of the penalty kill. Anderson played just a little bit less in total than Hischier, but when he was in the lineup, he received longer PK shifts. And why not? Somehow, someway, the opposition never scored against Anderson. I did a double-take when I saw that GA/60 of zero. It’s legit. Anderson had the most shorthanded minutes of all NHL forwards who witnessed no goal against them during a PK all last season too. That’s something. Combined with good (but not great or amazing) ranking rates across the board, Anderson provided good value for his 1:15/per game for 34 games in shorthanded situations.

As far as the top two forwards on the Devils’ penalty kill last season, I come away pleased with Travis Zajac and Blake Coleman have done. While their rates were not as good as the others, it is important to keep in mind their situation. Typically, Zajac and Coleman would see the other team’s top power play unit on most kills. Those primary PP units tend to have the better attackers and longer shifts. This is not to discount the secondary or teritary units, but the first unit has it harder to limit the opposition. There are more opportunities to fail against better players. By the stats and the overall team performance, Coleman and Zajac were hardly failing. To that end, I think it is laudable that the on-ice against stats across the board for Zajac were all in the upper-half of the NHL. Ditto for Coleman, who just missed it in scoring chances against per 60. Let us also not ignore Coleman’s special ability to create offense. That it’s only 25 shots and 39 shooting attempts should tell you that shorthanded situational offense is really uncommon, but Coleman made it happen because he is ferocious as the “plus-one” in the Devils system.

All together, the Devils had six forwards who played over a minute per game on the PK and they all finished in the upper-half of the entire league in nearly every stat. (And they were 5-for-6 for SCA/60.) Half of those six - Zacha, Boyle, and Rooney - were in the top 10% of all 179 NHL forwards in this comparison in multiple categories. No matter who the Devils had available for forward, their common forwards were able to perform on a level of good to absolutely astounding. Most teams are happy to have a couple of forwards perform well on their penalty kill. Despite how 2018-19 went, the Devils had six of them. And had Hischier played more (he just had 0:36/GP), then it could be seven given how well he did in his spot-duty.

If nothing else, I hope the chart of stats also gives you an idea of what is considered “good” for an on-ice rate stat for shorthanded situations. For even strength situations, percentages are appropriate and easy. Anything above 50% is good, anything below 50% is not, and since it’s 5-on-5 (or the rarer 4-on-4 or 3-on-3), you can focus on whether the team out-performs their opposition. But for a penalty kill, the shorthanded team is forced to defend a lot. Therefore, it is more appropriate to look at against rates and compare them with others. In general, if you see that a player has, for example, an on-ice SA/60 below 50, a xGA/60 below 6.5 or a GA/60 below 6.8, then that player is better than their peers in the league in that stat. The lower it is, the even better and more of a stand-out they are in.

So Far, So Good, So What Now?

The Devils had six forwards who ranged from quite good to quite spectacular when it came to on-ice metrics on a penalty kill. What does this mean for the future? While Boyle was traded during last season, Zacha (pending a new contract), Rooney, Anderson, Zajac, and Coleman will all return in this coming season. That should mean it is not an issue, right? Well, not necessarily.

It is worth noting that it is not a guarantee that Anderson and Rooney will start the season in New Jersey or join the roster for a while. As good as Anderson was and as great as Rooney was on the PK last season, the majority of a player’s ice time in any game will be at even strength. After all, the rate of time used for this comparison was a minute per game in shorthanded situations and most forwards are in the range of just 1-3 per game. How they perform at even strength is going to really determine whether they should be on the roster. Anderson had some relatively decent numbers on the horror-show that was the Devils last season. Rooney was firmly in the lower half among the Devils forwards on a bad team that used a lot of forwards. Rooney’s salary is one-way this season per CapFriendly and Anderson’s time last season may give him a leg up over the other Binghamton Devils. I am of the opinion that if they are in New Jersey, they should be killing penalties. Of course, if they are not, then the Devils need some additional options.

The good news is that they may have them already. While he did not make the ice time per game cut, Hischier’s 44 minutes or so last season on the penalty kill were quite good. Despite the limited ice time per game, his rates ended up being in between Rooney and Anderson. The young pivot may not be great at taking draws but he is growing as a defender. I think it is worth giving Hischier more of a chance on the PK. Additionally, past data from Kyle Palmieri - yes, really - suggests he can take a shift here and there.

As for the Devils who are locks to be in New Jersey in this coming season, the coaching staff could consider using Zacha more. Keeping Zajac and Coleman as the top unit is perfectly fine. They did well. Zajac can win draws and his best value seemingly comes from defensive efforts. Coleman is a weapon on the PK and does well - when he is able to kill a penalty. (Aside: The Devils’ PK can improve simply by having Coleman take fewer than a team-leading 33 minor penalties.) At the same time, I think Zacha has shown he could take a larger role. His rates were too good to keep to a limited role, and he did not play a relatively small amount of PK time when he was active last season. The coaches could rotate the trio as needed with Rooney and/or Anderson taking a secondary spot and others (Hischier?) filling in as a tertiary player. Regardless of how you see it, it is a good problem to have. Much better than trying to figure out four players to fill in four spots (see: the defense, but that’s a different post for a different day).

Conclusions & Your Take

Maybe it is only me, but I find it fascinating how well the regular forwards performed on the PK last season for New Jersey. Zacha was legitimately elite at it. Boyle and Rooney were just about there. Zajac and Coleman had to do a lot of heavy lifting and were better than at least half of similar forwards last season. Anderson somehow did not see a goal against in under 43 minutes of shorthanded ice time. This crew were full of standouts compared to their peers on the PK, so much so that they were a big reason why the Devils had one of the best penalty kills in 2018-19. That most of them are returning and could be in New Jersey means the Devils coaches and staff can focus more on other areas. Even if there is a drop-off, I really, really, really doubt they’re going to go from being one of the best groups of penalty killing forwards to one of the worst in a season.

That’s how I see it. How do you see this? Were you also marveled at how high the Devils forwards ranked among other NHL forwards who played at least a minute per game on the penalty kill? Were you stunned to see Zacha, Boyle, and Rooney rate so highly? Ahead of training camp, which forwards would you put on the Devils’ penalty kill for next season and why? Please leave your answers and other thoughts about the regular forwards from New Jersey’s penalty kill in the comments. Yes, you can even discuss Pavel Zacha; please do not be a pain about it. Thank you for reading.