Even as we continue moving forward towards the 2017-18 season, we still have time to take a look back on the one we’d like to forget. Today I’m going to look at what one model can tell us about the team’s performance last year.
What is GAR?
In case you haven’t seen it, Dawson Sprigings (@DTMAboutHeart) introduced a model called Goals Above Replacement (GAR). Its intent is to use six metrics to evaluate how many goals a player was worth over a replacement level player. If you want to read a detailed explanation, start here. (This is part 5 of 5 articles, and the links to the previous four are at the top of this one)
To keep it relatively simple, the six metrics he uses are:
- Even-Strength Offense (EVO)
- Even-Strength Defense (EVD)
- Power-Play Offense (PPO)
- Drawing Penalties (DRAW)
- Taking Penalties (TAKE)
- Faceoffs (FAC)
To see how he constructs each metric, please read the link above.
In this article, I will look at what this model tells us about Detroit’s performance last year and how it compares to other teams.
Why is GAR Important?
Part of my research for this article was trying to see if I could find a correlation between GAR and standings points to calculate how many wins a player was worth to his team. I’m still working on that, but to put it simply, the higher a team’s Total GAR, the more points in the standings that team is likely to have (with a few outliers). The following graph shows how each team’s overall GAR correlated to its points in the standings in the 2016-17 season. Out of the top 16 teams in Total GAR, three missed the playoffs (Dallas, Winnipeg, and Los Angeles).
UPDATE: I missed Florida originally. Four teams missed the playoffs.
The next chart shows that this trend exists beyond just last season.
Probabilities of making the playoffs based on total GAR pic.twitter.com/G4OUb4CdCN— Icy Pesos=Troll Bae (@yolo_pinyato) August 1, 2017
So, an increase in Detroit’s total GAR will lead to an increased chance of making the playoffs.
The 2016-17 Red Wings Through the Lens of GAR
Now let’s take a look at how individual Red Wings Players provided value to the team based on the GAR model. I will provide a link at the end of the article to all the charts used here.
I took the top 390 Forwards and 200 Defensemen from the 2016-17 season based on time on ice. For players who played for multiple teams, I multiplied their numbers by the percentage of games played for each team.
Red Wings Forwards
First let’s look at the forwards. The axes are based on the NHL maximum and minimum for TOI and GAR. Connor McDavid led the NHL last year with 24.6 GAR. You can see the NHL averages as well.
Not surprisingly, Zetterberg, Tatar, and Nyquist are well on the positive side of the league average. Some may be surprised, however, to see Frans Nielsen clustered with Nyquist and Tatar (The names overlap, but Nielsen is in between those two). Nielsen had a +/- of -19, but he still contributed on a level consistent with Tatar and Nyquist, showing the downside of looking at +/- as an evaluative metric.
Andreas Athanasiou is very close to the NHL average while playing 270 fewer minutes of playing time than the NHL average. This gives further support to the idea that he could greatly help the team if given more ice time. He nearly provided as much value to the team as Dylan Larkin, even though the TOI gap between he and Larkin is over 400 minutes. Also, while Larkin had a relatively poor season compared to last year, he still contributed nearly the league average (although not as much as you would expect given his ice time).
Red Wings Defensemen
It’s difficult to find too much positive when looking at Detroit’s defensemen’s GAR.
Not surprisingly, Danny Dekeyser, and to a lesser extent, Mike Green, did not provide the value to the team you want to see considering their ice time. Green was the only Red Wing blueliner over the NHL average in GAR, but with over 400 more minutes than NHL average.
(If you’re curious, Oscar Klefbom led NHL defensemen with a GAR of 17.)
How Did Detroit Compare to Other Teams?
So we’ve seen how individual players did, and just by looking at the NHL averages, it’s not hard to see that many players on the Red Wings need to improve from their performance last year.
But, let’s take a closer look at how the team’s forwards and defense compared to the rest of the league.
Detroit’s forward group compared favorably to the rest of the league, ranking 9th overall. Pittsburgh is head-and-shoulders above the rest of the league, followed by Washington and Minnesota. The Red Wings find themselves in a cluster that includes San Jose, Anaheim, and Los Angeles.
So, if the forwards are 9th, how bad does the defense have to be to drag the team total down so low?
29th in the league bad. Below Colorado bad.
So What’s the Takeaway Here?
The overall point is that while Detroit’s forwards did not have a great year, they weren’t that bad overall. Detroit’s defense, however, greatly hurt the team. Coach Jeff Blashill has indicated that he knows that Danny Dekeyser cannot handle the load of a #1 defenseman, so that’s a step in the right direction.
This analysis also shows just how much of a drag on the team Niklas Kronwall is. Even with reduced minutes, he provided negative 0.5 GAR. Jonathan Ericsson was even at 0 GAR. There were 16 NHL defensemen last season with a GAR of 0 or less. Detroit had 2 of them.
For GAR, replacement level is categorized as all players who are below the top 390 forwards and 210 defensemen in TOI, approximately 13F and 7D per team. That means a replacement level player is a healthy scratch or an AHL callup.
This shows that Ericsson provided as much production as a replacement level player, and Kronwall provided LESS than a replacement level player. For a total of $9M in cap space. A prospect would be much less expensive for (at worst) the same level of production). This is in line with what Prashanth wrote last year about maximizing the value of ELC contracts.
While Danny Dekeyser takes the brunt of fans’ anger over the poor Detroit blueline, Kronwall and Ericsson hurt the team more.
Detroit had an opportunity to take a step in the right direction by buying out Jonathan Ericsson, but Ken Holland did not pull that trigger.
Replacing both Kronwall and Ericsson would not fix everything, but it would start to move the defense corps in the right direction, and we see here just how much the team needs that.
All the charts from this article can be found (in interactive format) here.