NBA Study: How do you build a good NBA defense?

Key Highlights:

• The three main ways to build a top-10 defense is through having good rim protection, defensive playmaking, or point-of-attack defense
• There is a fourth, less common, way to build a top-10 defensive unit
• If your offense isn’t elite, you need to be at least good (top-10 level) on defense

A couple of weeks ago, we discussed which NBA teams are looking like title contenders so far this season. In the article, we discussed what kind of teams historically go on to make deep playoff runs.

As a refresher, three types of teams tend to make it to the conference finals (making it to the conference finals is considered a “deep playoff run”): 1) teams with elite offenses, 2) teams with elite defenses, and 3) teams with balance on both sides of the ball.

Basically, if your offense isn’t elite, you need to be a good defense to make a deep run in the playoffs. But how do you build a good NBA defense?

Three Ways To Skin A Cat

Coming into this article, we had a hypothesis that, to build a good NBA defense, you need to be good at protecting the rim, creating turnovers, or defending at the point-of-attack (a cool way of saying on-ball defense). To verify this hypothesis, we looked at how every top 10 defense since 2003-04 (that is how far the Cleaning the Glass database goes) fared in each of these three categories.

For the sake of this study, we used NBA.com’s team defensive ratings. To measure rim protection, we looked at each team’s opponent rim frequency (available on Cleaning the Glass). We chose to look at rim frequency instead of rim accuracy because, as a general rule, it is better to not allow shots at the rim at all rather than contest those shots at the rim very well.

To measure defensive playmaking (a cool way of saying turnover creation), we looked at each team’s opponent turnover percentage (available on NBA.com).

And lastly, to measure point-of-attack defense at the team level, we looked at opponent assists per 100 possessions (available on NBA.com). This isn’t a perfect way to capture on-ball defense. But it is a good proxy, as good point-of-attack defenses don’t surrender a lot of assisted field goals. We are using per 100 possessions to adjust for pace.

What We Found

It looks like our hypothesis was correct. Of the 210 teams we looked at, 196 (93.3%) were in the top-10 in at least one of those three categories (more on the 14 teams that weren’t in just a moment).

For those wondering, of the 210 teams in our sample, 97 were in the top-10 in rim protection (46.2%), 94 were in the top-10 in defensive playmaking (44.8%), and 136 were in the top-10 in point-of-attack defense (64.8%). Based on those numbers, it seems the most common way to build a good defense is through strong point-of-attack resistance.

For those of you who are proficient in math, you probably noticed that those numbers add up to more than 210. That means that many of these teams finished top-10 in multiple categories. In fact, only 39.5% (83 of 210) of the top-10 defenses we looked at were top-10 in only one category. 44.8% (94 of 210) of the teams in our study finished top-10 in two of the three categories. And 9% (19 of 210) of our sample was in the top-10 in all three categories.

A Fourth Category

We mentioned that 14 of the teams that have finished top-10 in defensive rating since 2004 didn’t finish in the top-10 in any of the statistics we used to measure rim protection, defensive playmaking, and point-of-attack defense.

Well, 12 of those 14 teams share a common denominator that makes us think that there might be a fourth way to “build” a good defense. What is that pathway, you ask?

Those 12 teams all ranked in the top-10 in opponent 3-point shooting percentage. That means that they all (to varying degrees) had some opponent shooting luck on their side. We call it “luck” because, generally speaking, it is hard for defenses to have a serious impact on an opponent’s 3-point percentage.

So, that means that there are technically four ways to “build” a good NBA defense. You can do it through good rim protection, good defensive playmaking, good point-of-attack defense, or good opponent shooting luck.

Okay, what about those two other teams that don’t fit into any of the four categories we’ve discussed? How did they build their top-10 defenses?

The two teams we’re talking about are the 2013-14 Oklahoma City Thunder and the 2021-22 Cleveland Cavaliers. And as you will see in a second, both of these teams are victims of the (admittedly) arbitrary parameters we created for this exercise.

In 2013-14, the Thunder finished 11th in opponent assists per 100, surrendering 22.4 assists per 100. That is only 0.1 assists per 100 more than the two teams tied for ninth place that season: the Toronto Raptors and the Charlotte Hornets (then known as the Bobcats). So, the Thunder just narrowly missed the top-10. If they had averaged 22.2 opponent assists per 100, they would be in the top-10 in point-of-attack defense, and we wouldn’t even be talking about them (or typing, if you will). 

As for the Cavaliers, despite being 25th in the NBA in opponent rim frequency in 2021-22, they finished first in opponent rim accuracy. So, while Cleveland was surrendering a ton of field goals around the rim, they were still a good rim-protecting team, just not in the way our study was measuring it.

What Does this All Mean?

Okay, so we now have all this information, but what do we do with it?

At the end of the day, we follow/cover the NBA to figure out who is going to win the title at the end of the year. We use the regular season to help us predict which teams are in serious contention for that title. Now, we have another tool in our toolbox to aid us in this quest.

Like we said earlier, if you don’t have an elite offense, you need to be at least good (meaning top-10 level) on defense. And how do you build a good defense? You build it through your rim protection, defensive playmaking, or point-of-attack defense. If none of those things are going for you, you better hope that luck is on your side when it comes to opponent shooting.

And if the team you are analyzing doesn’t have any of that, and their offense isn’t elite, it is going to be very hard for them to make any noise in the playoffs.