Improved Offense & Weaker Defense: The New Jersey Devils Under John Hynes So Far

During the many, many losses and poor performances by the 2018-19 New Jersey Devils, I have had many complaints like any other Devils fan who is paying attention. Some are undebatable: the goaltending, mainly. Others have been somewhat more contentious. I have been increasingly critical of the coaching staff in light of John Hynes receiving a contract extension back on January 3. This usually gets a response about how the roster is so bad so how can anyone fault the coaches.

I am sympathetic to the idea that the roster needs more talent. I agree. It sorely misses Superstar Taylor Hall, who has been out with a significant injury. However, every team in the NHL is always looking for talent. There are zero teams who do not believe they need more talent. Further, I think the team has more talent than they did back when Ray Shero took over as general manager and John Hynes was hired way back in 2015. I also think I can sort of prove it.

I have to say sort of because it has not been all improvements and good things since Shero gutted the roster to build what we have. That should surprise no one; the team has five road wins and sits in 30th in the NHL as the regular season resumes tonight.

Further, there are aspects of the performances this season where coaching bears at least a part of the responsibility. Specifically, the defense, My general thought is that the coaching bears at least a part of the responsibility. No, they are not the ones on the ice missing coverages or not getting the right position on opposing players. No, they are not failing to complete zone exits or remembering how to move after a faceoff loss. However, when a player makes a mistake - they make a mistake. If the same mistakes keep being made, then it is on the coach to figure out that they need to address it and how to address that.

What makes me feel more strongly about this is that the Devils’ penalty kill is one of the best in the NHL. In a primarily-defensive situation where defensive structure (usually a wedge plus one) is critical, the Devils have been great. An array of defensemen and forwards have been utilized and they have been effective at winning pucks, clearing or attacking with them, and keeping the other team from converting. On a PK, the Devils have been great. In 5-on-5 play, the Devils are prone to get lost in their own end and that is where mistakes have been made that have limited the Devils at best and goals against the Devils at worst.

Don’t take my word for it, take Andy Greene’s. Corey Masisak of The Athletic tweeted this quote after a 2-4 loss to Toronto where the team conceded 43 shots, 69 shooting attempts, and 17 high-danger chances to Toronto in all situations.

This would be one thing if the coaching staff were new. But they are not. John Hynes has been the head coach since 2015. Assistant coach Alain Nasreddine was also hired in 2015 as one of Hynes’ first assistants. As I understand it, Nasreddine is responsible for the defense and the penalty kill. Greene has worked with them for the last four seasons. If he’s saying the team is in chaos when they don’t have their structure, then that tells me the team does not have a good enough structure or a “Plan B” to fall back on when things get chaotic. I’m confident whatever they have that results in Ben Lovejoy being by one of the points, skaters being confused over who is in zone and who is covering players man-to-man, and opposing players not being tied up around the crease despite being in position is not it.

Circling back to my main point, while the team under Hynes has made some strides, the defense has been particularly problematic this season. I also think I can prove that alongside with showing that the team has improved in some regard. The proof can be seen with the team’s 5-on-5 stats from the last four seasons. We know the PK has been great and that is not really in question. But the majority of the game is in a 5-on-5 situation, and so that will show how the team has performed on the ice over the last four seasons.

Corsi or All Shot Attempts

Corsi is a count of all shooting attempts: shots on net, missed shots, and blocked shots. They are largely taken in the offensive zone. They are not everything but they are a good indicator of who is effectively controlling play in 5-on-5.

A good defense would want to have a low Corsi Against per 60 minute rate, as it represents how many attempts they have allowed. A Corsi For per 60 minute rate would represent how much the Devils have attempted shots. The Corsi For percentage is a proportion of both rates. Even if the Devils are allowing a lot of attempts, if they can generate even more then they will come out ahead. While the position of defense means the focus is on keeping those against rates low, it is often that offense is generated from the team playing well in their own end. Transition offenses are often formed from a defensive play. Winning the puck can help end an opposition’s attack and allow the team to start their own.

How have the Devils have done in the last four seasons in this category?

Natural Stat Trick

When Hynes and Nasreddine first joined the roster, the Devils were very good at limiting attempts. One of the best in the league. The problem was that the Devils were the worst at generating them. So night after night, they lost in this regard and it contributed to why those teams needed to be gutted.

However, while the Devils have made improvements to generate more attempts and improve their proportion of shooting attempts, the team has allowed a higher rate of attempts in each of the three previous seasons. If the current rate holds, it will be a fourth straight season. In four seasons, the Devils have went from being one of the stingiest to being one of the lesser teams in terms of preventing attempts. I can appreciate the offensive improvements, but this is a poor look for the defensive end.

This is also not the only stat where this happens.

Fenwick or All Unblocked Shot Attempts

There is a thought that blocked shots ultimately do not mean much. A block can absolutely be necessary and even helpful. They can also cause further problems like an inadvertant deflection or an injury to the shot-blocker. I agree with Kent Wilson about shot blocking; if you’re doing it a lot, there are bigger problems. But still, what if we take those blocks out of the situation and focus on the attempts that either hit the net or missed the net?

It’s a similar picture for the Devils:

Natural Stat Trick

Back at the start of Hynes and Nasreddine, the Devils were one of the best at limiting unblocked attempts - and the very worst at generating them. There has been a substantial gain in terms of generating unblocked attempts inn 2017-18 and the 2018-19 team has built on it. That is a good mark of improvement and an addition of talent. Unfortunately, the team has been allowing more and more unblocked attempts. The Devils are now a bottom-ten team in terms of FA/60 and it is a reason why the Devils percentage remains under 50%. It is not that being 50% in FF% will guarantee success, but it does represent an issue for the team. Unless things change in the next two months or so, the Devils will still be out-Fenwicked for the fourth-straight season - even with these improvements.

Shots on Net

Maybe you do not care so much for attempts or unblocked attempts. Maybe you believe that shots are the main thing for a defense to deal with. Goaltenders have to stop them and defensive work often has the goal of denying a shot. So let’s look at shots. It’s a familiar story:

Natural Stat Trick

Again, the first season of Hynes (and Nasreddine and others at the time) yielded a team that was dead last in creating shots on net and one of the best in the NHL in limiting them. Up until where we stand today, the team’s rate of shots have improved by leaps and bounds. At the same time, the team’s rate of shots against have become higher and higher. The good news is that in this category, the Devils are almost breaking even. The bad news is that since the Devils have had such terrible goaltending this season, not breaking even and giving up shots at a rate below 32 per hour in 5-on-5 play hurts the cause big time. Again, the more defensive component of this team stat has been getting worse under Hynes’ (and Nasreddine’s) reign. At least the offensive part has been good - which reflects some improvement in talent in that regard.

Scoring Chances or Shooting Attempts from the Crease to the Dots up to the Top of the Circles

Hynes and others in the NHL have increasingly mentioned scoring chance counts and measurements in their pre and post-game comments to the media. Over at The Athletic ($), Justin Bourne has a post that goes into how teams across multiple levels focus on chances - however they are defined. It is a good read about the background from that perspective. For this post, the perspective is to use the public stats definition is a “homeplate:” from the crease out to the dots and up to the top of the circles from the dots. These are the more dangerous attempts that a defense can focus on stopping. The idea is that it is not so bad to allow a lot as long as it is from the “outside” or apart from the homeplate. How have the Devils done in this category under Hynes? About the same as the other stats presented in this post:

Natural Stat Trick

Once again, the Devils started at terrible at creating chances for them but quite good at limiting them. Once again, the good news is that the Devils have created more and more scoring chances. Once again, the bad news is that the Devils have allowed more scoring chances too. This time, there is a twist. This season’s SCA/60 rate is a little better than last season’s; and that rate from the past seasons is not ranked in the bottom third of the NHL. The current season rates somewhat close to the league median. There is definite room for improvement, but it is not a stark shift as with attempts, unblocked attempts, and shots from the past four seasons.

High Danger Scoring Chances or Attempts from the Slot and At the Crease

One of the hallmarks of the Devils under Hynes has been their relative excellence with high danger scoring chances. These are the attempts in the two most common areas for goal scoring: the slot and at the crease. The thing about defending these areas is that while it may help to be big and strong and willing to clear it, being in position, sticking to your man, staying composed in the area, and putting the effort to disrupt the opposition is more critical. They can also be done by any player at any size too. The Devils have seemingly figured that out. However:

Natural Stat Trick

The Devils remain one of the top teams in terms of defending these chances. Their rate of allowing high danger chances has ranked in the top ten for all four seasons, assuming the current rate holds for 2018-19. What’s more heartening is that the Devils have upped their own rate of generating these chances in the last two seasons. This season has not been as good as last one in this regard (and in at least five others), but it is a respectable rate and it has yielded a good HDCF%,w hich is just outside of a top-ten rate. That all written, the rate of high danger chances allowed has increased in each of the last four seasons. It is not a positive trend, but at least the Devils are still better than the majority of the league in that regard. It is something that perhaps the team should consider recognizing.

Expected Goal & Actual Goal Rates

In a nutshell, the Expected Goals model looks at every shot location, shot type, and shot situation and assigns a value to each one to determine their likelihood to score. Basically, if a team takes 25 shots but most of them are in that scoring chance area, then their expected goal value would be higher than a team that took 25 shots but most of them were from the point or the sideboards. That does not mean that the first team actually scored more goals or not. The model is just that: a model. It is a good way to justify whether the work by the offense or defense has been good enough to be at least theoretically productive.

I like to compare expected goals with actual goals in general, so even though this is more about the skaters than goaltending and shooting percentage, I’ve included it for the sake of perspective. Also, unlike all of the other stats used so far, this comes from Corsica.

Corsica

Consistent with the other 5-on-5 stats presented, the 2015-16 Devils were quite good in terms of against-rate stats and utterly pitiful for for-rate stats. What’s more is that the Devils were even worse at creating actual goals in 5-on-5 hockey in 2015-16 and 2016-17. Still, we see the same general trend: the Devils have improved in the expected and actual goals for rates for the last four seasons. That points to an offense becoming more and more threatening. At the same time, they have a higher expected and actual goals against rate over the last four seasons - which points to the team’s defense being weaker and weaker.

What’s more is the divide between actual and expected goal rates. The model has the Devils with one of highest goals for rates in the NHL in 5-on-5 play. The Devils in reality have been somewhat unfortunate to not finish enough plays to meet those. As poor as the goaltending has been, more goals would yield more results. Or at least closer results. On the flipside, you can see the impact of the bad goaltending. The model expected the Devils to give up goals at one of the higher rates in the NHL in the past two seasons - and the actual rate for those seasons is even worse. While the Devils are on the right side of breaking even per the model; it has been worse among actual goal rates.

Conclusions or OK, So What?

In 2015-16, while the defensemen were not exactly a murderer’s row, the Devils had Corsi, Fenwick, shots on net, scoring chances, and high-danger scoring chance against rates among the best in the NHL. There may not be a true defensive stat yet, but this all points pretty strongly to the idea that the Devils were good to very good on defense in 2015-16. They were utterly bad at creating offense and scoring goals to take advantage of it. I would agree that the Devils absolutely needed a re-build and the roster needed a lot more offensive talent.

And so Ray Shero did that through a variety of ways and John Hynes made adjustments to have the team create more offense. The current team has only three players on the same contracts that they had back in 2015-16. It is a very different team.

Statistically, the changes did work in that the Devils have been attacking more and more. Over the last four seasons, the Devils have had increasing rates of creating shooting attempts, unblocked attempts, shots on net, scoring chances, and high-danger scoring chances. This has led to increases in both expected and actual goal rates. To that end, the Devils have add more talent and Hynes has utilized that talent to be more attacking. The stats show this. The Devils roster is not nearly as bad or bereft of talent as you may think.

The stats also show that the defensive side has been worsening under Hynes’ reign. There have been a lot of changes on defense in the last four seasons. They had a 33-year old Andy Greene leading the blueline in 2015-16. Only Damon Severson and Steve Santini (who played one game) remain from that season other than Greene. I’m not sure I would say they have not added talent given the acquisitions of Sami Vatanen and Will Butcher in the last two seasons. As well as the subtractions of Merrill, Gelinas, Quincey, and Schlemko. Nevertheless, the stats are all clear in this regard. The Devils began under Hynes and Shero as this really stingy 5-on-5 team; a squad that would not create much but try their best to prevent you from creating much in response. They now stand as one of the easier teams in the league to generate shot attempts, shots on net, and chances (but not high-danger chances) against. Yes, the goaltending has been horrible but since that the expected goal model has the Devils with a rate above 2.5 goals per hour, it cannot all be blamed on Schneider and Kinkaid being bad. The stats show that the team’s defense by way of against-rate stats has been worse over the last four seasons.

And I remind you all that the 2018-19 Devils have not made significant additions or subtractions from the 2017-18 team. Yet, the rates for defense have continued going in the wrong direction and the rates for offense have been a bit improved.

From my perspective, this further justifies my thinking that the coaches bear at least some responsibility for how the team has played, particularly in 5-on-5 situations. At the same time, the improved rates on offense point to the fact that the team has added talent. I agree: Shero needs to get this team more talent, but that’s a problem every organization has. Despite their current record, the 2018-19 Devils are not where the 2015-16 or 2016-17 Devils are from a performance perspective. I can agree that perhaps they deserve some credit for the offensive improvements. But I hope that the improvement is not at the cost of doing worse elsewhere - especially something that would be magnified by bad goaltending. That would make the decision to extend John Hynes to be even more baffling.

So that is why I think the way I do as the Devils continue to largely disappoint in their own end of the rink and suffer even worse away from the Rock in this season. Now I want to know your take. Now that you’ve seen the rate stats, what do you make of John Hynes and his staff so far? Do you see where I’m coming from? Do you agree; and if not, why not? Can the Devils improve their against rate stats without getting new defensemen? Please leave your answers and other thoughts about the team’s 5-on-5 play under Hynes in the comments. Thank you for reading.