Have Third Periods Been a Problem for the 2018-19 New Jersey Devils?

Every so often, I get asked the question online or in person as to why the New Jersey Devils are so bad this year? How can they be struggling when they have seemingly solid numbers in 5-on-5 play? Why do they keep dropping games and points? And one of the answers I keep going to is the third period. How the Devils lost leads. How they got tied up. How they struggled. How they finish games. I believed it was a problem. Is it the worst period for the Devils?

Technically not, that would be overtime where the Devils are 0-5-0 so far this season prior to Sunday’s game. The Devils’ being winless in overtime has already been examined in great detail in this excellent post by Evan Sporter at The Athletic ($). So please check that out for that.

Getting back to third periods, I have to check myself: Am I even right? Have third periods really been a problem for the 2018-19 New Jersey Devils? Let’s take a closer look and find out.

Third Period Scores

Let’s focus on goals first. After all, this is the part that matters. Goals decide games. How are the Devils entering third periods and what has been happening within them on the scoreboard? I went through the 27 game summaries at NHL.com and put together this breakdown of goals. I listed out total goals, 5-on-5 goals, and power play and penalty kill goals. Since I was concerned about giving up scores late, I included goals scored within the final five minutes, goals scored in extra skater situations (highlighting any non-empty net goals), and empty net goals.

I also highlighted the games where the Devils entered the third period trailing in the game and managed to get at least a point out of it in green. In red are the games where the Devils were tied or leading and managed to lose the game in some way. In light green are the games where the Devils made a comeback within the third period to get a result. In orange are the games where the Devils tied it up or took a lead in the third, only to lose it within the period.

Here is the total breakdown.

NHL.com

That’s a lot going on here. Let’s hit some highlights here:

• Out of 27 games, the Devils have had two games where they entered the third period trailing and earned something out of the game. The Devils have had two games where they were leading after two periods and ended up dropping points.
• Additionally, within the third period, they’ve had two games where they saw a lead go away but managed to get something out of the game. They have had four games where they tied it up or took a third-period lead and saw that go away by the end of the third. These four are what likely stuck out to me as the third period being a problem.
• Another way to sum up the previous two points is that the third period only swung the result of the game in four out of 27 third periods, with another six seeing changes within the period. A majority of third periods did not have a change. If the Devils were leading, they would win most of the time; and if they were trailing, they would lose most of the time.

This latter point points pretty hard to the third period not being a period of doom for the Devils. If games ended after 40 minutes, the Devils’ record would improve from 10-12-5 to 11-12-4 (assuming the Devils lose all the regulation-ties like they have been this season). That is a gain of exactly one point. So much for my thinking that this where the Devils have been getting owned this season.

Further, the Devils have been not stunned by the opposition scoring goals with the extra skater all that often. It only happened to them twice, both by Florida. The first was with 15 seconds left and it was a consolation goal that turned 3-1 into 3-2 on October 27. The second was more painful; an equalizer with a bit over a minute left on November 26, which ended in an overtime loss. Ignoring empty net goals, the Devils have actually scored more goals with the extra skater than allowed. Not that the Devils won the two overtimes they forced because of it (the one in Detroit was a no-impact goal).

But I think there is a problem here, and they are more subtle ones. The Devils have struggled to turn the tide in games where they are down after two periods. While several of these deficits have been large, the Devils have rarely broke through to grind a point out of any of them. Even in the smaller ones. In the few times the Devils tied up the game or took a one-goal lead, that has not been safe and they have fallen. This is an example of falling apart at the margins, as Corey Masisak wrote at The Athletic ($) last week.

Additionally, the Devils have been outscored over all third periods in this season so far. Here is a summary of the above.

NHL.com

Overall, the opposition has continued to out-score the Devils in third periods overall, in 5-on-5 play, within the final five minutes of the game, and in extra-skater situations. Granted, the extra-skater situations are mostly empty netters with the Devils giving up two ENGs twice on two separate nights. They should be a reminder to John Hynes that pulling the goalie again after giving up one is not a good idea.

By the way, the Devils pulled their goalie eleven times this season. Six of those times resulted with an ENG.

Anyway, the other curiosity here is that special teams have been a net negative what with the Devils power play converting just five times, conceding three times, and the PK being beaten four times. Only giving up 4 PPGAs is actually rather good over a range of 27 periods; but the PP issues arise here as well. But with so few goals even happening here, the larger concern is outside of special teams - so I will focus on that.

What’s more is that third period goals against have been somewhat frequent. In 18 out of 27 third periods, the Devils concede at least one goal. In just over half of all third periods, the Devils allow at least one goal in 5-on-5 play. Just under 30% of the time, the Devils are allowing goals late in the period and/or with someone having the extra skater. While the Devils do get on the board, it is not as often and not as much as the opposition overall. So while the Devils may not be losing a lot of their games outright in the third period, it is further evidence of a struggle to further secure leads or make them happen. In other words, it still doesn’t look like the team plays well in the third period. Of course, there are other stats that better speak to their performance.

Third Period Performances

Since how the Devils have played in third periods is of interest, I pulled their total 5-on-5 stats from each of the 27 games and their third-period stats from each game report at Natural Stat Trick. While 5-on-5 is not everything, it is the most common situation in games and in the third period. Knowing how the Devils and their opponents performed in attempts (Corsi), shots, scoring chances (SC), high-danger scoring chances (shots from the slot and crease, HDC) will provide a sense of how the run of play went. Including the entire game will provide a baseline to compare with the third period stats.

I also broke it up by games where the Devils were leading after two periods, trailing after two periods, and tied. There are plenty of posts going into more detail about it (example here from Second City Hockey back in 2013, but the concept goes as far back as 2009-2010), but in a nutshell, trailing teams tend to attack more, take more risks, and push the play in the hopes of finding a goal. Leading teams do not. Hence, stats like Corsi, shots, and scoring chances would be skewed in that way. It is not always a guarantee, but this is what tends to happen. Effects also tend to be smaller when the deficit or lead is by three or more goals since such a deficit/lead may demoralize a team to not necessarily attack more. Anyway, I broke it up to see what is happening between the situations.

Natural Stat Trick

Natural Stat Trick

The total versus third-period stats are fascinating. You can see the score effects more clearly with the percentage chart: the 12 third-periods where the Devils were trailing have them above 50% across the board in terms of attempts, shots, scoring chances, and high-danger scoring chances. The only area where they are not: goals. They still give up more goals than they score (7 to 9) when trailing going into the third. At least they have made attempts overall to make something go their way offensively in the final frame when trailing.

You can also see it in the leading rows. The Devils have been decidedly out-attempted and out-scoring chanced in third periods when leading after two periods. That should be no surprise; the Devils are leading and the opposition is not. What is a surprise is that the Devils have broke even in shots and they still generate more high-danger scoring chances than their opponents. Whether the opposition has had bad accuracy or the Devils defense positioned itself well or the opposition just faded away, it’s a good surprise. It likely contributed to the Devils maintaining most of those leads in those 11 third periods. Again, even with the ice tilting somewhat against the Devils (but not totally), the Devils have managed to outscore their competition, which is also a good thing.

There are two more problematic issues here. First, the tied-score situations. When it does happen, the Devils have been bodied in 5-on-5 play. Being below 50% in these stats means the opposition is generating more of those events. Being below 40% is just being dominated in attempts, shots, and chances. Oh, and the Devils have been out-scored there too. Those are four games that was literally up for grabs in the third period and the Devils have been second-rate in each one.

Second, the Devils entered the third period tied only four times this season. That is not good in of itself. It points to a larger “all or nothing” theme to this season that hasn’t been all-ways worth it (pun intended). You can see that in the total 5-on-5 stats in each chart. The Devils have played well overall in 5-on-5 when they end up ahead on the scoreboard after two periods. They have out-attempted, out-shot, and out-chanced their opposition while definitively outscoring them. For the 12 games where they have been trailing after two, the 5-on-5 percentages are treading around the breakeven mark except for GF%, which is expected since it points to the Devils being out-scored by a lot. 22 goals, to be precise. And in the four times they were tied, the 5-on-5 numbers for the whole game were really bad. So much so that I wonder how the Devils were tied at all. Over all 27 third-periods, the Devils have been OK when it comes to shot differential, good in high-danger chances, bad in terms of attempt and scoring chance differential, and really bad in terms of goals.

To put it another way, these are two more ways the Devils have failed in the margins this season.

Conclusions & Your Take

Third periods have not been good for the New Jersey Devils overall, but it is not a major reason why they went into Anaheim with a 10-12-5 record as the last-place team in the Eastern Conference. Again, if the games were 40 minutes long, then the Devils would be 11-11-4, which is not much of a gain and would still leave them in last in the East as of the morning of December 9.

It is true that the Devils have played plenty of poor third periods, especially when the Devils were tied going into the period. Over all 27 periods, the 5-on-5 numbers are not good and the team has been out-scored in third periods in general. However, through this post, I now see it as more of a symptom of a larger issue than a cause of their poor season. The Devils have been very much a team where they need to play well to get results. If they’re entering the third period with a lead, then they have a good chance of winning the whole thing. If not, then they struggle to get a result. Comebacks have been rare by the Devils. There have only been two going from the second period into the third so far this season, and even that yielded just one win (the other: an OT loss). This is all further evidence of Masisak’s thesis of what has been wrecking the 2018-19 Devils.

It’s not so much that I want the Devils to play bad hockey and somehow get wins like, say, the Islanders. But quality teams tend to turn those below-average, average, or slightly-above-average performances into wins. That is something the 2018-19 Devils have not really done. The third period performances and scoring speak to that. If they want to turn this season around, among other things (e.g. win some dang overtimes), the Devils need to be better in the first two periods to obtain leads. This way they are in a more favorable position to maintain leads going into the third than trying to make a comeback or manage one within the period - two events that have not successfully occurred for the Devils this season.

Now I want to know what you think. What do you think about the Devils’ performances in third periods this season? Did this post change what you thought about them as it did for me? Where do you see the Devils making some improvements for better third periods or better results going into the third period? Please leave your answers and other thoughts in the comments. Thank you for reading.