Rebounding Thresholds, Offensive Investment, and Separability
An Indiana Pacers breakdown.
The past month has been spent studying for classes, interviewing for jobs, and reading Dean Oliver’s Basketball Beyond Paper - a remarkable book that has been steadily shaping some of my pre-existing ideas on how to better measure team performance with granularity. Dean’s work has served as an inspiration for my own work. I (along with most basketball analytics people) wouldn’t be anywhere without him.
I haven’t taken this much of an extended break since I started consistently writing in the spring, but not for a lack of ideas. For one, working with such a low sample size of data can prove troublesome - and second, some of my ideas haven’t crystallized until now. I now have a few articles that I’m proud enough of to publish.
Today, we’ll talk about the Indiana Pacers’ struggles to start the season.
Pacers’ Struggles:
The Pacers lived up to their name in 2023-24, ranking 2nd in the league in pace and 3rd in transition opportunities (20% frequency of transition possessions). This pace yielded good results, giving them an offensive rating of 2nd in the league. While their offense was considered brilliant, and reminiscent of some of the best offensive teams we’ve ever seen, their defense lagged behind, ranking 24th in the league last season in defensive rating. This defensive struggle seemed to be part of the strategic trade-off the team made, with maximal investment in their high-powered offensive engine and a higher focus on maintaining pace and tempo over defensive play.
We have seen this strategy in teams like the “Seven Seconds or Less” Phoenix Suns, who led the league in pace and offensive rating in 2005 and ranked 16th in defensive rating. In addition, they ranked 29th in defensive rebounding and 23rd in offensive rebounding percentage. This is the first clue we have towards the Pacers struggles. The Suns, despite having nearly the worst defensive rebounding percentage, were ever so slightly better on the offensive glass. The gap between their offensive and defensive rebounding theoretically has a lot to do with their wanting to push the pace. Cross-referencing this with the Pacers, we see that the Pacers were 26th in defensive rebounding percentage and 16th in offensive rebounding percentage. These baselines help us understand the team’s identity and willingness to gamble on their offense pushing them to the top. The following table displays the pace to start the season.
Looking at the Pacers’ play this season - while most of the discourse has been around Tyrese Haliburton’s struggles and the defense, the team has similarly struggled to keep up on the offensive end as well, dropping from 2nd in offensive rating last season to 14th this season. The pace has dropped from 2nd to 16th (now riding below league average). This drop is much more of a red flag than their defense - which we expected to be poor. Investigating this, I looked at their transition opportunities. They ranked 3rd in the league in transition frequency in 2023-24, at 20% - and now rank 13th, with 18.6% of their opportunities in transition. This decline suggests something faulty at play with their ability to generate quick offense.
Is it Haliburton? I checked transition frequency for all players on the Pacers, and while Halliburton had a decline, the bulk of the Pacers who have declined in transition frequency are their forwards/bigs (chart below).
McConnell and Andrew Nembhard, on the other hand, have had increases in transition frequency - suggesting that there is something keeping the Pacers’ forwards/bigs on the other side of the court. In the context of the offseason and now the beginning of the Pacers’ season, this makes sense.
Isaiah Jackson and Jalen Smith, both strong rebounders (defensive rebounding percentages of 18.7% and 21.9% respectively) are both not playing a role for the team this season. It’s important to note that Jalen Smith, who was signed by the Bulls, now plays more than 20% of his possessions in transition on a team that has jumped 27 spots to lead the league in pace.
The added pressure on Turner, Nesmith, Toppin, Jarace Walker, and Pascal Siakam to grab defensive rebounds has stagnated their ability to run to the other end of the court. The tradeoff between defensive rebounding and pace has now resulted in the Pacers giving a lot of their pace back to maintain a relatively mediocre rebounding standing. We can see this lack of rebounding in their offensive rebounding percentage, which has plummeted from 16th to 27th. In addition, Toppin, Turner, and Siakam have lower DREB% chances this season - suggesting that they are overextended on that end of the floor. Their rebound percentage when boxing out has also dropped, while Halliburton and Nembhard have stagnated.
This is a subtle insight - mainly because the defensive rebounding numbers haven’t changed much at all (only a drop of 0.6%) - but given their overextension, it looks like a strategic choice to “buy” some of those rebounds back at the cost of some pace. Instead, trading for players to “buy” those rebounds without trading any pace seems like a solid approach to the problem.
I would suggest players like Jabari Smith Jr. and Santi Aldama to get the Pacers above the threshold naturally, without sacrificing any pace.
Rebounding Thresholds:
The Pacers’ case study made me start thinking about the drivers of a good offense. We don’t consider defensive rebounding to be a significant part of offensive performance - in fact, the public opinion for years has been that defensive rebounding is, by definition, a defensive metric. But by analyzing the Pacers and other high pace teams - rebounding is certainly a part of pushing in transition. If your team isn’t good enough at rebounding, you do have to invest more effort into that side of the ball, which hampers transition opportunities.
I theorized the concept of a rebounding threshold. Above the threshold, a team can invest their rebounding into transition production, but below the threshold, a team must trade in some of their pace/offensive production into better rebounding. Below the threshold, a lack of rebounding production might be reason for trading in some pace in order to prevent the other team from beating you on second chance opportunities.
The weak linear correlation between defensive rebounding and pace has certainly played a role in our assessment of it as an important offensive driver. However, this weak correlation makes sense. A team that has a low defensive rebounding percentage doesn’t necessarily imply a high pace - they could just be a bad rebounding team. Similarly, a good defensive rebounding team (by percentage) can have high pace simply as a result of having super strong rebounders, so much so that their team naturally stays above the threshold.
Obviously, this isn’t fleshed out. Theorizing how to measure what this rebounding threshold is the most difficult part - and without being able to decide on “what could be” - it gets increasingly difficult to pull this all together into a quantifiable metric. The general idea remains though - above the threshold, you can trade in defensive rebounds for pace, but below the threshold you might have to trade in your pace for defensive rebounding. Depending on the style of the team, they might not choose to make that trade.
Instead, I am more interested in pointing out the holes in our current style of thinking towards rebounding and offense as a whole. If we analogize the Pacers’ transition offense to a racecar, we might see Tyrese Halliburton as the engine, and Toppin, Nesmith, and Mathurin as the wheels. But forgetting about something like valve oil is surely going to affect the performance of the car in a significant way. That’s the best way to look at something like rebounding. Without smooth transitions between defense and offense, we might get sputtering results.
I often emphasize granularity - but I want to reiterate that granularity doesn’t imply separability. If we view basketball too generally, we end up falling into, well, generalizations. But if we view basketball with too much separability, we run the risk of losing some of the more important factors when analyzing this very complex system. Instead, viewing basketball as a system of interchangeable part, rearranged in optimal or non-optimal ways, helps us better understand how to smooth out any potential issues in our thinking.
Wanted to add that some of these statistics were measured at prior dates. Will be updated with up-to-date statistics in upcoming weeks.