A Fresh look at the evolution of rebounding and the top Board Men

---

One of the most significant casualties in the modern NBA has been the value of the rebound. As recently as 10 years ago, players, especially big men, were summarized by their point-rebound line. In 2023, rebounds are scarcely mentioned beyond completing a triple-double or if a player grabs 20 boards. This is especially true in analytical circles, as previously obscure aspects of basketball like defensive versatility and spacing have been relentlessly quantified and analyzed, while rebounding remains relatively fallow ground¹Though of course, a ton of great work has been done, much of which will be referenced in this and future articles. The meta seems to be “Rebounds don’t matter.”

However, I’ve always been partial to elite rebounders. It’s a demanding, thankless job carried out by players who often don’t score or control the ball much. Also, while rebounding is not key²Team rebounding has generally correlated to ~10% of the variance in team wins using multiple regression of the Four Factors, as shown hereto success in the NBA, I do think it is under-analyzed and can provide value to teams through the possession aspect of basketball now that efficiency has improved so dramatically in the modern era. Consider this series of articles a comprehensive look at the history, strategies, stars, and future of rebounding in the NBA.

---

Who is the greatest rebounder of all-time? Looking at rebound per game (RPG) numbers, the answer seems to be either Bill Russell or Wilt Chamberlain, who both averaged over 24 RPG in their careers, ~10 more than 3rd place on the all-time list, Wes Unseld. However, like many stats in the NBA, rebounding numbers need context, especially when comparing across eras. Wilt and Russell played in an era with lower FG% and more possessions, leading to far more rebound opportunities. In a series of articles on his blog Skeptical Sports³These posts, written in 2010-2011, are excellent and form the basis of much of my own, reading them if you haven’t already is definitely worth the time, Benjamin Morris argued convincingly that Dennis Rodman was the greatest rebounder of all-time by such a large margin that Rodman was actually one of the most valuable players of his era, despite lacking much of an offensive game. A key statistic used to further this argument is total rebound percentage (TRB%), or what percentage of the available rebounds a player grabbed while on the floor. This measure is both pace and shooting percentage agnostic, key for comparing across eras⁴The average TRB rate is 10% as there are 10 players on the floor at once. D-Rod’s career mark of 23.4% TRB is far ahead of players like Moses Malone (19.8%) and good estimates of Wilt and Russell’s careers. However, a quick look at the current career leaderboard shows an interesting change…

So is Andre Drummond now the greatest rebounder of all time? He certainly thinks so. It’s hard to argue with him; TRB% is a very fair statistic. But I will. Dennis the Menace is still the GOAT. The reason why takes some explaining and a deeper look at the changes to rebounding over the past decades.

---

Sparked by rule changes limiting defensive contact as well as the growth of data in basketball, the NBA game has changed noticeably, particularly in the past 10 years. Teams take substantially more 3’s (which has led to better spacing) and play at a faster pace than in the 90s or early 2000s. These changes have correlated with a substantial decrease in league-wide offensive rebounding:

A decrease in offensive boards hasn’t been the only NBA rebounding trend. The percentage of contested boards has also decreased from ~35.5% in 2013-2014⁵The first year NBA tracking data is availableto 31% in recent seasons. These trends seem to be related:

The theory for this is simple. The default team to claim a rebound is the defense, as they typically have the inside position. More voracious offensive rebounding leads to more contested boards, as offensive players challenge defenders’ otherwise easy rebounds. If one assumes that OReb% and Contested Rebound% are linked linearly in this way, the contested rebound rate in Rodman’s time period was likely around 45%(!), substantially higher than the ~30% of today.

This argument by itself might not be convincing. Sure, D-Rod may have had to work harder for his rebounds. But a board is still a board. At the end of the day, someone needs to get it and Drummond has grabbed a higher percentage. So why do uncontested rebounds matter? It’s all about who grabs them. Consider this graph comparing the distribution of players’ Total Rebound% relative to league average in ’92 (Rodman’s prime) and now:

The distributions are vastly different! ’22-23 features only a single tall peak consisting of players who generally average a TRB% roughly 3 to 5% below league average (non-rebounders). On the other hand, in ’92-93 there was a smaller mass of players who were non-rebounders⁶Guards and smaller forwardswith another peak of above average rebounders. There are far more players not contributing on the glass in today’s NBA than in the ’90s. There’s another key difference that’s less obvious but is perhaps even more important. Notice how many more high volume rebounders⁷ >10% relative TRBthere were in ’22-23. It may seem like a tiny difference on the graph because the overall percentage of high volume rebounders is small, but there were nearly double the amount of volume rebounders in ’22-23 than in ’92-93! The differing distribution of rebounds has been a trend over the past few decades:

The Worm was a huge outlier in his day, but today not so much. This is immediately obvious looking back at the all-time TRB% career leaders again:

Look at how many current players there are! While raw rebounds are biased in favor of Wilt and Russell, rebound rate is biased towards top boarders today. Why? Because with less players competing for rebounds today (due to spacing and smaller wings), the top dogs get to eat.

---

So far we’ve seen that there are less offensive rebounds⁸ As well as contested rebounds presumablytoday than in the past, more players battled for boards in the ’90s-’00s, and that Dennis Rodman was more of an outlier in his day than many current top rebounders like Andre Drummond. However, there is a further reason why Rodman was such an elite rebounder, but to identify it we need to talk about why rebounding matters.

What is the purpose of a rebound? In simplest terms, it’s to gain⁹In the case of defensive boardsor maintain¹⁰Offensive reboundspossession of the ball. Logically this means that if a player’s team was going to get the ball anyways (by the ball going out of bounds or being rebounded by another player on the team), the rebound is not valuable. In other words, what teams really care about is when a player gets a marginal rebound¹¹Rebounds the team otherwise would not have had. This concept is central in other areas of basketball as well. This is why a player getting an offensive rebound is considered more valuable than a defensive board. The offensive team was unlikely to keep the ball if a player would not have grabbed the offensive rebound. But how can we tell which rebounds count and which don’t¹²Especially in Rodman’s day where there is no tracking data for contested rebounds or other information?

Luckily, there’s a method for accounting for this issue, though it is inadvisable in small samples. It’s the with-or-without-you (WOWY) method, also commonly referred to as On/Off splits and used in stats like +/-. Through this indirect measurement, it is possible to estimate how impactful a player’s rebounding skill was on their team’s ability to collect rebounds while they were on the court as compared to when they were on the bench. Stats like this can be skewed by who a player was on the court with, who their replacements were, as well as the identity of their opponents and their role on the team. However, most of the top rebounders in history have had long careers playing with a variety of teams and teammates, so their WOWY rebounding statistics should be reasonably indicative of their impact on the boards. This measurement of individual rebounding also correlates better with team success. Without further ado, here’s the On/Off impact¹³In their ~8-12 best seasonsof some prominent top board men¹⁴Courtesy of pbpstats.com and Skeptical Sports:

The first key takeaway has to be the greatness of Dennis Rodman. The only players approaching his impact on the boards are Reggie Evans, a part-time journeyman who was an offensive rebounding specialist and Enes Freedom¹⁵Formerly Kanterwho’s also rarely started. On the other hand, the Worm typically averaged ~35 MPG in his prime and doubled as a key defender for 5 NBA champions, yet still remains the most impactful and prolific rebounder in NBA history.

Another player that stands out to me is Nikola Jokic, the best player in the NBA today. While the Joker is known for his passing wizardry and shooting touch, his defensive rebounding is extremely impressive. In his 8 seasons Jokic has had by far the largest impact on his team’s defensive rebounding of any of these 10 elite rebounding bigs. This isn’t just a function of Denver’s lack of a true backup center in recent seasons either. Jokic’s On-Off DReb% splits are equally impressive in his early career, when he played with and was backed up by other very good rebounders, like Mason Plumlee and Jusuf Nurkic. Add another aspect of his game that’s all-time special to the tally.

The final player that I find intriguing is Andre Drummond. Not only does Drummond have the highest career TRB% of all time, he also has grabbed a particularly huge portion of defensive rebounds:

The difference between Drummond’s DRB% and second place all-time¹⁶Swen Nater played in the ABA and NBA in the late 70s and early 80s, is bigger than the gap between Nater and Cousins! Despite this, Drummond’s teams have consistently grabbed the same percentage of defensive rebounds with him on the court as off it. This has been true when he has played with the Pistons, Cavs, Nets, 76ers, and Bulls. Only with the Lakers did his On-Off DReb% splits eclipse 0.3% added and a +1.5% clip in a very small sample is not particularly convincing. To be honest, I don’t have a good explanation for this. Andre Drummond has played over 22,000 career minutes in the NBA with a variety of teams and teammates. Other top rebounding centers of the same era like Jokic, Howard, Gobert, and Jordan have increased their teams’ DReb% by a significant margin. But not supposedly the best defensive rebounder of all time?! Drummond’s advanced defensive rebounding metrics also support his greatness. He doesn’t grab a particularly low percentage of contested defensive boards. It’s a bit of a mystery (post your thoughts in the comments!). One thing is clear though: Dennis Rodman is still the rebound king.

---

One more thought: A key characteristic of Rodman and other great rebounders is their ability to collect both offensive and defensive rebounds. As Benjamin Morris points out, it his hard to do both at an elite level. I have a particularly intuitive potential explanation for this: elite offensive and defensive rebounders are typically a bit different. For instance, consider this graph:

Look at the absence of data points in the top right corner (high Usage% and high OReb%). Top offensive rebounders typically do not get a ton of usage in terms of shots and assists on the offensive end. Compare that with defensive rebounding:

The opposite phenomenon, though not quite to the same extent, is present. This makes sense. Offensive players that are more focused on creating their own shot or a shot for others cannot be as focused on getting good offensive rebounding position. As well, high usage players are often perimeter oriented in today’s NBA and would not get many offensive rebounds regardless. This phenomenon can result in offensive rebounding being slightly undervalued by some models, as noted by Nate Silver. This is not true on the defensive end and therefore there’s less of an effect, with perhaps more athletic players (who excel on both offense and defense) getting a slightly higher share of defensive rebounds.

Given this difference, it’s quite remarkable how players like Kevin Love and Nikola Jokic, who have been significant offensive contributors in their careers, have also impacted the offensive glass effectively. They are true rebounding unicorns.

---

There’s still much more to cover. Look for my upcoming article(s) in this series which will cover other rebounding topics like: 1) Quantifying rebounding skill (in terms of possessions added) using tracking and other data 2) Why teams should focus more on offensive rebounding 3) What individual and team strategies lead to more rebounds? Also, a similar series on forcing and preventing turnovers? Stay Tuned.

Subscribe to be Notified When I Post