Watching Dallas, it certainly feels like there are times when the team lacks hustle, effort, grit etc., especially on defense. I wanted to try to quantify that, so I dove into the NBA.com “hustle” stats. Let’s get to it.
First of all, what are the hustle stats and where do they come from? Several years ago, every single NBA arena had tracking cameras installed. These track every player, during every second of game play. When people reference stats like “so and so shoots 40% on 3’s with a defender 4-6 feet away” …that’s the kind of data that comes from these cameras. They also can tell how far players run, how fast they accelerate etc. They also track…. hustle stats. I just rarely see anyone reference them. NBA.com tracks about 10 different official stats in the hustle category. I’ll link to it at the end so you can play around with the number for yourself. The main ones I focused on are deflections, loose balls recovered, charges and boxouts. The data is collected both at the player and team level.
Deflections
Dallas averages 11.6 deflections per game. You’re probably wondering “is that a lot”? In short, no. That’s the second least in the NBA, ahead of only the Pistons. On the other end of the spectrum, Orlando and the Clippers average 17-18 deflections per game, they’re both top 5 in team defensive rating. As I think about deflections, such a low number doesn’t necessarily mean that Dallas “doesn’t hustle”. This is also a measure of the general length and wingspan of a team’s defense. See Orlando and LAC, teams stacked with rangy wing and forward types, guys who are obviously a little more prone at getting their hands on passes. As a point of interest, there is a moderate correlation between deflections and a lower team defensive rating.
Looking at individual player deflections highlighted some interesting points. Keep in mind these are all per 36 minutes. I thought that would make the most sense or this would just largely reflect which guys play the most minutes. The graph is color coded to denote which quartile the player ranks in the entire league (I kept the “entire league” to rotation players…. very loosely defined. Basically, anyone who has played more than 75 minutes this season).
Curry has always had the reputation as a pesky defensive player and his deflections certainly back that up. Anecdotally I’ve felt like Luka has had active hands this year and again that shows up in the numbers. Good to see Kyrie fairly high up, as I do feel like he’s battled and given great effort on D this year. Green and Powell feel about right too, both extremely active and energetic.
Grant and Hardaway being so far below average feel significant to me, as they both account for a ton of important minutes. There are many ways to make an impact on defense, but one of the easiest is turnovers, deflections etc. and it just seems like Williams lacks the length and quickness to bother offenses in this regard. Kleber obviously jumps off the page in a bad way. Would be interesting to compare his numbers over the past couple of years….is he consistently this low or is this a reflection of his poor play and general physical deterioration this season?
Loose Balls Recovered
Let’s look at loose balls recovered. This is a stat that measures exactly what it says: “the number of times a team gains sole possession of a live ball that is not in the control of either team”. Dallas checks in with 6.4 per game…. which ranks 4th in the league. Like deflections, I think this stat measures a bit of a combination of true “hustle” and just the wingspan and length of a team. Looking through the entire league the numbers generally move in the direction that you would assume…. more loose balls collected = better defense. Until Houston. They have the #1 defensive rating in the league…and the least number of loose balls collected in the entire league by a wide margin. See the graph below. They’re the X in the upper left-hand corner all by themselves. Their number is so far out of the norm that I wonder if there is some issue with the data, or the cameras. If you include all 30 teams, there is 0 correlation between loose balls recovered and defensive rating. I eliminated Houston and ran the analysis for the hell of it and it did end up being very similar to deflections…moderate correlation.
See below for the Mavs individual numbers on loose balls collected. I’m a little bit unsure of how I feel about this stat….my gut says that this is more of a true “hustle” stat. I feel like it does a pretty good job of calling out general effort and activity on defense. Diving into a pile to grab a contested ball, getting your nose dirty etc. I’m curious what others think on this matter. I have a feeling that big men generally score low here, as being so tall makes you less liable to dive onto the floor. Just a theory, I’ll have to do some additional digging into that at some point.
Again, these are all per 36 minutes so playing time won’t impact the numbers. Holmes, Exum and Green would be near the top of my list if you had me subjectively guess who “hustles” the most on Dallas. I’m really surprised that Powell checked in at the bottom of the team. Grant is near the bottom here as well but watching him I feel like he’s giving maximal effort out there. Just something interesting to note, maybe more a reflection of his positioning within the defense or noise within the early goings of the season. Overall, this tells me that Dallas has quite a few guys who aren’t afraid to get after it. That clearly isn’t enough on defense, but it’s a start.
Charges Drawn
This one is also straightforward. How many charges does the team draw per game? Not to my surprise, Dallas ranks near the top of the league. Their figure of .94 charges drawn per game is good for #2 in the league behind only Thibs’ rough and tumble Knicks squad.
Statistically speaking, there is no correlation between charges drawn and defensive rating (at least with this data set, eventually I’d like to expand this analysis to prior seasons). To me, that tracks. Who is generally good at drawing charges? Players who aren’t good at blocking shots. Players who are shorter and stockier, less athletic. It makes sense to me, that drawing charges is kind of the yin to the yang of athleticism and flying around on defense. Players who excel in one of those areas generally don’t put in the time or effort to excel in the other area. Or they simply don’t have the physical capabilities. Let’s dive into the Dallas roster specifically.
This graph looks a little goofy because of the relative lack of charges by nature….as you can see 6 players have yet to draw one this season. Unsurprisingly, Timmy Hardaway is one of the league leaders here. Small sample size but encouraging to see Hardy here. He’s clearly going to be a mediocre defender at best so would be in his best interest to start to master this skill as early as possible. What would be more interesting would be to compare these charges drawn numbers to players considered to be top defenders throughout the league. Another project to add to the list.
Contested Shots
Contested 2- and 3-point shots is also tracked on NBA.com. I analyzed that data and just frankly didn’t come away with anything worthy of a longer writeup at this time. Basically, big men are all at the top of the scale for contesting 2 pointers, while the league leaders in 3 pointers contested were mostly on teams that simply allowed a bunch of 3’s. I’ll need to reevaluate and try to figure out some different ways to parse that info.
Boxouts
There’s quite a bit of data on NBA.com about boxouts. Let’s start with how they define boxouts; “The number of times a player made physical contact with an opponent who was actively pursuing a rebound, showed visible progress or strong effort in disadvantaging the opponent, and successfully prevented that opponent from securing the rebound”. My main focus for this analysis was on the defensive glass as essentially every scheme calls for strong boxing out on the defensive end. This should, in theory, be a decent measure of a team’s overall effort.
The Mavs, maybe to the surprise of no one, are left wanting. They are 22nd in the league in defensive box out % i.e… they give up a lot of unimpeded looks for other teams on the offensive glass. This certainly aligns with my eyes during games. Let’s look at the individual data because this is where things get super interesting.
The Dallas big men ALL jump off the page….in a very bad way. Powell and Grant are below average while Lively and Holmes are in the bottom quartile of the league. Of course, some context is needed. For a guy like Lively, I assume his boxout rate is going to suffer because he’s such a shot blocking threat. It’s simply not possible to do both at the same time. While Holmes is nowhere near Lively’s level deterring shots at the rim, he’s got a little something to him in that regard, so again, the numbers make sense. Powell and Grant are tough to reconcile, as neither has much ability to block or alter shots at all. Are both guys going for blocks when they have no chance and subsequently failing to box their man out? Are they simply losing track of their mark…or just not giving the required effort? If they’re not boxing out and not providing resistance at the rim…. where is their value add as bigs on defense? It’s funny because my eyes tell me Grant is playing well on defense and I don’t think these stats are necessarily saying he’s a part of the problem. Just some interesting data points that suggest that he may be better suited to be a piece of a good defense, rather than the catalyst that turns around a terrible overall defense.
As an aside, you can see there are several Mavs with 100% box out rates. That’s the cases on most teams. Have Luka and Kyrie really combined to miss 0 defensive box outs this entire season? I’m skeptical. The total number of box outs seems low to me as well. Kyrie is tagged as having 3 this season, while Luka has 14. These numbers track with other teams so it may just be that I need to get a more clear understanding of how they define a box out attempt.
Overview
I came into this exercise trying to ascertain if Dallas plays hard or not. Of course, I knew that would be tough to prove with just the numbers and frankly I don’t feel like I’ve drawn a strong conclusion in that regard. The deflections are an interesting data point that I hadn’t seen analyzed, but to me that says more about the lack of length in the Mavs lineup rather than their disdain for hustle. The box out percentages were really the key takeaway for me here. Again, I’m not sure that necessarily ties into Dallas’ effort levels on defense, but I do think it unequivocally highlights a major weakness area that needs to be addressed on the defensive end. If anything, looking at the hustle stats have opened more questions than answers for me. I’ve got a few threads I want to pull at, especially regarding taking a closer look at these stats and their correlations to strong team defensive ratings in previous seasons. Appreciate everyone who read all the way through, would love any feedback.
**All the stats above are current through 11/26/2023**
The biggest problem is Luka when he’s faced with an explosive or quick player.
No. Your worst defender being faced with his worst matchup is not your biggest problem. It’s a double unicorn and should happen rarely.