clock menu more-arrow no yes

Filed under:

Are the Devils Above, Below, or Exactly Average

New, comments

I look at the landscape of the conference and adjust for a few less reliable contributing factors to determine whether the Devils are in fact an average NHL team in this, the first year of their rebuild.

Ed Mulholland-USA TODAY Sports

Not too long ago I sent a tweet about the Devils being an average team.

Those numbers have shifted a bit since then, hence the investigation in this article.

Not long after that, I wrote an article specifying that we are an average team, uniquely susceptible to injury due to our top-heavy roster. Specifically I mentioned total dependence on top 5 guys offensively. I also wrote an article on our once-again elite goaltender. Then I wrote another on our young defenders in which I talk about our total defensive dependence on Larsson/Greene. In summary, we as a team are completely dependent on 8 guys (5 Forwards, 2 Defenders, 1 Goalie).

Due to this dependence, the Devils are likely to be a highly volatile team, riding the hot/cold streaks of these players. As Mike wrote earlier this week, the Devils are still in the hunt despite their detractors. In order to see if this is likely to continue, I have tried to make some adjustments on the typical standings points worth observing. All statistics for these tables were retrieved from Hockey-Reference. All stats were retrieved the night 12/30 and thus do not include the Ottawa game.

Typical Standings Points as a Percentage of Possible Points

Team GP W L OL PTS PTS%
Washington Capitals 35 27 6 2 56 0.8
Florida Panthers 37 21 12 4 46 0.622
Boston Bruins 36 20 12 4 44 0.611
New York Islanders 37 20 12 5 45 0.608
New York Rangers 37 20 13 4 44 0.595
Detroit Red Wings 37 18 12 7 43 0.581
Montreal Canadiens 39 21 15 3 45 0.577
Ottawa Senators 37 18 13 6 42 0.568
New Jersey Devils 37 18 14 5 41 0.554
Tampa Bay Lightning 37 18 15 4 40 0.541
Philadelphia Flyers 35 15 13 7 37 0.529
Pittsburgh Penguins 35 17 15 3 37 0.529
Carolina Hurricanes 37 15 17 5 35 0.473
Buffalo Sabres 36 15 17 4 34 0.472
Toronto Maple Leafs 35 13 15 7 33 0.471
Columbus Blue Jackets 39 14 22 3 31 0.397

Devils Conference Rank: 9th

Devils League Rank: 15th

We are pretty textbook "average" according to the shallowest of rating systems. We are a spot higher due to the win against Ottawa, but still clearly average,

Goal Differential

Team GP GF GA G+/-
Washington Capitals 35 108 72 36
Boston Bruins 36 113 96 17
New York Islanders 37 100 86 14
Florida Panthers 37 96 83 13
Montreal Canadiens 39 109 97 12
New York Rangers 37 106 96 10
Tampa Bay Lightning 37 94 85 9
Ottawa Senators 37 108 108 0
Detroit Red Wings 37 93 96 -3
New Jersey Devils 37 85 88 -3
Pittsburgh Penguins 35 78 84 -6
Buffalo Sabres 36 84 91 -7
Toronto Maple Leafs 35 89 96 -7
Philadelphia Flyers 35 75 93 -18
Carolina Hurricanes 37 87 105 -18
Columbus Blue Jackets 39 97 122 -25

Devils Conference Rank: 9th

Devils League Rank: 16th

One of the Ten Laws of Hockey Analytics is that goals are all that matter. However, they do not matter in fixing the Devils Standing position. After the Ottawa game, the Devils are now exactly even in goal differential. Predictably we are also in the middle of the NHL.

Rating with Adjustment for Goal Differential And SOS

Team SRS SOS
Washington Capitals 1.04 -0.05
Boston Bruins 0.54 0.04
Florida Panthers 0.49 0.03
Montreal Canadiens 0.4 0.07
New York Islanders 0.35 -0.06
Tampa Bay Lightning 0.26 0.07
New York Rangers 0.17 -0.07
Ottawa Senators 0.01 0.03
Detroit Red Wings -0.07 0.04
Toronto Maple Leafs -0.12 0.11
Pittsburgh Penguins -0.16 0.04
New Jersey Devils -0.2 -0.07
Buffalo Sabres -0.23 0.02
Carolina Hurricanes -0.47 0.05
Philadelphia Flyers -0.56 0.01
Columbus Blue Jackets -0.58 0.06

Devils Conference Rank: 12th

Devils League Rank: 21st

The SRS is the simple rating system which accounts for SOS (strength of schedule) and goal differential. It is a unique stat to Hockey-Reference. The Devils are in the bottom 5 in the NHL in strength of schedule. This seems odd with the observed strength of the Metropolitan Division, but a spade's a spade. The Devils have apparently had an easy road. And when adjusted, the Devils are a below average team. We would be 4 spots out of a playoff spot upon adjustment for these variables.

Goal Differential with Half-PDO Adjustment

Team G+/- PDO GC PDOadj G+/-
Montreal Canadiens 12 0.00 12.00
Washington Capitals 36 -27.49 8.51
Boston Bruins 17 -9.78 7.22
Carolina Hurricanes -18 23.69 5.69
New York Islanders 14 -12.97 1.03
Florida Panthers 13 -12.73 0.27
Tampa Bay Lightning 9 -9.46 -0.46
Buffalo Sabres -7 6.30 -0.70
Pittsburgh Penguins -6 4.39 -1.61
Toronto Maple Leafs -7 5.33 -1.67
Detroit Red Wings -3 -3.29 -6.29
New York Rangers 10 -17.40 -7.40
New Jersey Devils -3 -5.92 -8.92
Philadelphia Flyers -18 8.82 -9.18
Columbus Blue Jackets -25 13.86 -11.14
Ottawa Senators 0 -20.57 -20.57

Devils Conference Rank: 12th

Devils League Rank: 21st

PDO is commonly called the luck statistic because prior PDO is totally useless in predicting future PDO. However, goalies can have reliable save percentages, and players can vary from league average in shooting percentages in a very BABIP-esque way for any sabremetricians reading today. So I couldn't totally take out the effect of shooting percentages. I assumed that each team was halfway between there current location and 100. As mentioned in my "thesis," the Devils are riding a few players.

PLAYER GP G A PTS ATOI SOG SPCT
Mike Cammalleri, LW 38 14 21 35 19.5 98 14.3
Kyle Palmieri, RW 38 16 13 29 16.7 95 16.8
Lee Stempniak, RW 38 9 19 28 19.2 78 11.5
Adam Henrique, C 36 13 12 25 19.9 72 18.1
Travis Zajac, C 30 6 12 18 20.1 43 14
TOTAL 180 58 77 135 19.0 386 15.0

Our top 5 guys have a shooting percentage of 15. The league average is 9. For an individual to average above that is not so strange. The modern marvel of this as many who follow stats know is Alex Tanguay who has miraculously sustained an 18.7% over his anomalous career. But for 5 guys to average so significantly over the league average is simply unsustainable. Furthermore, Cory Schneider's SV% is 5th in the league at .927, his highest as a Devil (though not by much). These numbers are highly susceptible to regression due to a cold streak or injury.

Half-Special Team Adjustment

Team G+/- ST% ST% GC ST%adj G+/-
Washington Capitals 36 110 -10.40 25.60
Florida Panthers 13 98.14 2.27 15.27
Tampa Bay Lightning 9 96.64 4.03 13.03
New York Islanders 14 105.22 -5.45 8.55
New York Rangers 10 102.35 -2.63 7.37
Montreal Canadiens 12 103.61 -4.66 7.34
Ottawa Senators 0 95.68 5.21 5.21
Boston Bruins 17 111.3 -12.94 4.06
Detroit Red Wings -3 99.17 1.00 -2.00
Buffalo Sabres -7 98.23 1.76 -5.24
New Jersey Devils -3 101.87 -2.24 -5.24
Pittsburgh Penguins -6 99.57 0.51 -5.49
Toronto Maple Leafs -7 100.36 -0.39 -7.39
Philadelphia Flyers -18 95.49 5.01 -12.99
Carolina Hurricanes -18 95.45 4.55 -13.45
Columbus Blue Jackets -25 98.52 1.92 -23.08

Devils Conference Rank: 11th

Devils League Rank: 19th

Special Teams are a volatile portion of the game. This is not to say that they are necessarily going to regress to the league average as there is no evidence indicating special teams are random. However, the small sample sizes and shortened rosters make them significantly less predictable than 5v5 statistics. For example, take a look at the the season shot/goal rates for the Devils, and then the same rates when we were missing Zajac (and Henrique for a little).

Team Zajac? Gm SF% S+/- SF60 SA60 GF60 GA60
PK No Z 7 9.5 -34 6.5 61.8 0 9.8
PP No Z 7 89.7 23 40.1 4.6 6.2 0
PK Total 38 11.9 -154 7.3 54.3 0.9 5.5
PP Total 38 84.6 126 47 8.5 6.7 1.2

Significant difference. Now losing the top 2 PK forwards would hurt any teams PK%. Losing your two leading PP centers would hurt any teams PP%. Losing two of your top 5 overall forwards would hurt any teams point production. The problem for the Devils is that those were all the same two guys.

Since special teams are not random, yet highly volatile, I used the same logic s PDO and assumed a regression halfway to the league average. The Devils became slightly below average once again.

Synthesis and Conclusion

The whole file used for this article can be found here: Are Devils Average?

The Devils row of relevant data is below. Since its only one row i figured scrolling wouldnt be a problem here. If you want a more user-friendly representation, download the file.

Team Conf GP W L OL PTS PTS% ConfRk League Rk GF GA G+/- ConfRk League Rk PDO GC PDOadj G+/- ConfRk League Rk ST% ST% GC ST%adj G+/- ConfRk League Rk SOS SRS ConfRk League Rk
New Jersey Devils E 37 18 14 5 41 0.554 9 15 85 88 -3 9 16 -5.92 -8.92 13 23 101.87 -2.24 -5.24 11 19 -0.07 -0.2 12 21

My assessment of the State of the Devils would be that we are a slightly below average team. This is because we are an exactly average team when healthy, and a below average -- possibly bad-- team after 1 or 2 injuries. It is by no means impossible that the Devils snag a wild-card spot and make it into the playoffs, particularly if they stay off the IR. However, based on the data from this investigation (again, download the file for specifics) it appears as though it's only a matter of time before Tampa jumps us. In the long run I'd probably have my money on Ottawa to jump over us as well.

Your Thoughts?

Do you think we are average? Above average? Below Average? Where do you think we will be by year's end? What are the most intimidating teams in the WC race? Leave thoughts below -- and HAPPY NEW YEAR!!!!