Hand-Crafted Fancy Stats: Disciplined Aggression–Part 2

Earlier in the week, I wrote about a statistic called Disciplined Aggression Proxy, which gives us a better idea of which players play a very physical game, but do not take a lot of penalties in the process. If you didn’t read part 1, check it out now because there will be no other preamble today, I’m just going to jump into the results.

On Monday, I looked at just 2016-17, which I admitted was too small of a sample size. For today’s analysis, I used stats from the beginning of the 15-16 season, and used a cut-off of 50 games played. I suppose 50 games is still pretty small, ideally we’d want to look at 80-100 games or more. But, as we will see, a lot of players in the sample have about that many games anyway. In the tables, I will include GP so we have an idea of how long a player has been able to keep up a high DAP. I am sort of thinking of it like ERA in baseball–a closer who has a 2.0 ERA is great, but a starting pitcher who has the same ERA over a couple hundred innings pitched is a bit more impressive.

Because this stat hasn’t been extensively written about, we don’t have too much of an idea about the distribution of scores. To give the parameters on this sample, there were 464 players included, and the mean DAP was 9.23, with a standard deviation of 4.94. Here’s the distribution of scores.

So, this tells us that we would expect 95% of scores to be between basically 0-20…but technically that’s if it is a normal distribution, which it clearly is not. We could transform the data set to get a more normal distribution…but since I’m not running inferential stats, and I want it to be easily interpretable, I’m not going to transform it. This is just a long way of getting around to saying there are some players with extremely high DAP scores over the last season and change, which I think is telling. Without further ado, here are the top-twenty players in DAP:

Name Team GP Hits Takeaways Minors DAP
Alex Wennberg CBJ 94 58 51 3 36.33
Jack Skille VAN 94 130 20 5 30.00
William Karlsson CBJ 106 52 55 4 26.75
Jean-Babriel Pageau OTT 109 230 35 10 26.50
Paul Byron MTL 89 147 35 7 26.00
Aleksander Barkov FLA 94 63 66 5 25.80
Cal Clutterbuck NYI 102 377 35 16 25.75
Lance Bouma CGY 58 144 9 6 25.50
Jaccob Slavin CAR 90 52 72 5 24.80
Ryan Reaves STL 90 266 13 12 23.25
Nate Schmidt WSH 94 88 48 6 22.67
Rickard Rakell ANA 89 147 33 8 22.50
Andrew Copp WPG 97 115 20 6 22.50
Leo Komarov TOR 92 368 49 19 21.95
Derek MacKenzie FLA 92 226 38 12 21.75
Michael Del Zotto PHI 106 252 34 14 21.20
Chris Wagner ANA 64 141 5 7 20.86
Dennis Rasmussen CHI 70 64 19 4 20.75
Mark Letestu EDM 109 90 34 6 20.67
Brayden Schenn PHI 106 252 34 14 20.43

A couple of honorable mentions that were just outside the top-20: Matt Moulson (22nd, DAP = 20.25), Patrick Marleau (24th, DAP = 19.25), Michael Grabner (28th, DAP = 18.38), and Charlie Coyle (33rd, DAP = 17.42).

We still see that the stat is primarily driven by penalties, as Wennberg, Skille, and Karlsson have less than 5 minors in the last year and change. Which is still remarkable, they are consistently avoiding putting their team shorthanded, while still dishing out a bit of physicality. Also, it’s notable that two of the top three are Blue Jackets (both Swedes too,) and part of that may explain Columbus’ meteoric rise in the standings recently. Another name near the top is Cal Clutterbuck, who recently signed an five-year extension. My podcast co-host, Dan, told me right after we heard about the extension, “Yeah, he threw a ton of hits when he was with the Wild but they were all clean, he didn’t get penalized much.” His comment was partly what sparked me to dust off this stat…Clutterbuck is known for throwing big body checks, so Islanders fans might be pleased to see him among the DAP leaders, but let’s be honest, that extension still doesn’t make a lot of sense. I was also surprised to see Ryan Reaves up there, I thought he had a reputation of being a little more of a dirty player, but the numbers show otherwise (remember that his fighting majors are excluded here.)

Toward the bottom of the top-20 we start to see some players who are known for their offensive contributions–Schenn, Marleau, Grabner, and Coyle. For fans of players on those teams, it’s definitely good to see a two-way player who can play with some grit, score some goals, but not wind up in the sin bin.

Generally speaking, I think it’s more informative to see which players are on the high end of this stat as opposed to the low end, but because I’m sure people will be curious, here are the bottom-ten skaters in DAP dating back to the start of 2015:

Name Team GP Hits Takeaways Minors DAP
Keith Yandle Fla 110 53 35 29 3.03
Antoine Vermette ANA 104 81 34 38 3.03
Marcus Kruger CHI 70 38 22 20 3.00
Phil Kessel PIT 109 15 30 15 3.00
Daniel Sedin VAN 109 28 33 21 2.90
Henrik Sedin VAN 101 15 28 15 2.87
Jaromir Jagr FLA 107 39 49 32 2.75
Anthony Duclair ARI 106 36 37 27 2.70
Evgeni Malkin PIT 84 43 72 48 2.40
Tobias Enstrom WPG 99 34 16 27 1.85

Obviously, the players at the bottom of this list aren’t bad, by any stretch of the imagination. But, it’s clear that they do not pay physically at all (or if they do, they take a lot of penalties.) Kessel, Jagr, the Sedins, and Malkin–these are all basically one-dimensional players, and they’re good enough at generating offense that they are still superstars. I don’t think it’s necessary for a ‘good stat’ to confirm who we would already guess would be at the top and bottom of the ranks. It’s a little bit ironic that a pair of Swedes are the very top of the list, and another pair are at the very bottom, too. So yeah, I don’t know what we can learn from looking at the lowest scores, I just figured some people would be interested.

Overall, I wanted to look at a stat that doesn’t get a lot of attention, because if we fall into the trap of looking at a narrow group of metrics, we can forget that there are a lot of different ways to analyze a player’s performance. Any proponent of fancy stats worth his salt will tell you that there’s a lot that can’t be measured, so I think it’s useful to keep looking at things in new ways.

I will wrap this all up on Friday by looking specifically at the Minnesota Wild, and I’ve also uploaded my data file so you can check your favorite team or player and see what the DAP results look like. Here’s the link:


As always, thanks for reading, and don’t forget to follow me on Twitter, @BobaFenwick!