Offensive Rebounds and Three Pointers

Jan 2, 2016; Dallas, TX, USA; Dallas Mavericks forward Dirk Nowitzki (41) shoots over New Orleans Pelicans forward Ryan Anderson (33) during the second half at the American Airlines Center. The Pelicans defeat the Mavericks 105-98. Mandatory Credit: Jerome Miron-USA TODAY Sports
Jan 2, 2016; Dallas, TX, USA; Dallas Mavericks forward Dirk Nowitzki (41) shoots over New Orleans Pelicans forward Ryan Anderson (33) during the second half at the American Airlines Center. The Pelicans defeat the Mavericks 105-98. Mandatory Credit: Jerome Miron-USA TODAY Sports /
facebooktwitterreddit
Jan 2, 2016; Dallas, TX, USA; Dallas Mavericks forward Dirk Nowitzki (41) shoots over New Orleans Pelicans forward Ryan Anderson (33) during the second half at the American Airlines Center. The Pelicans defeat the Mavericks 105-98. Mandatory Credit: Jerome Miron-USA TODAY Sports
Jan 2, 2016; Dallas, TX, USA; Dallas Mavericks forward Dirk Nowitzki (41) shoots over New Orleans Pelicans forward Ryan Anderson (33) during the second half at the American Airlines Center. The Pelicans defeat the Mavericks 105-98. Mandatory Credit: Jerome Miron-USA TODAY Sports /

It is a common coaching aphorism that defense isn’t over until you grab the rebound. This is doubly true in the NBA where the general stinginess of the league’s half-court defenses is compromised in scramble situations, such as those that occur when a team tracks down its own missed shot. In fact, offensive rebounds are second only to steals in terms of producing quality shots on the ensuing possession:

Dashboard 1 (43)
Dashboard 1 (43) /

However, as noted by many[1. including Zach Lowe at length yesterday.], there is a strong trend in the NBA towards eschewing the offensive glass in favor of defensive transition.  Offensive rebounding percentage is at an all time low of 23.8% through Tuesday’s games, punctuating a steady decline over the last few years. And if anything, this dropoff understates the real decline, as leaguewide free throw rates are also near all-time nadirs.  Missed free throws are rebounded by the offensive team far less frequently than are missed field goals – thus far in 2015/16 offensive rebounding rate is 10.4% while it is 24.7% on missed shots from the field. Thus, a greater share of rebounds being missed field goals rather than the relative freebies for the defense which occur from charity stripe misses should push offensive rebound rates upwards rather than downwards.

Still offensive rebounding has tailed off. To a degree, this is because teams have stopped trying. This season nearly 60% of field goal defensive rebounds have been secured by the defensive team with no offensive player “contesting[2. Defined by SportVU as a player being within 3.5 feet, just over arm’s length, of the rebound when it is secured.].” This figure has been around 57% for the first two years of full SportVU data in 2013/14 and 2014/15. Given that contested rebounds have proven to be about a 50/50 proposition, this roughly 2.5 percentage point decrease in “Chase Percentage” explains more or less the entirety of a 1.3 percentage point decline in offensive rebound rate.

The trade off teams are making seems fairly clear: going for offensive rebounds and failing can lead to more fast breaks – though the location of the missed shot itself, or more precisely the location and identity of the shooter, possibly has a much greater effect[3. More on this in a second.]. The correlation between chasing offensive rebounds and giving up fast breaks isn’t perfect:

Dashboard 2 (15)
Dashboard 2 (15) /

(note the inverted X axis, as the more transition attempts allowed the worse.) Other factors, such as the success of those rebounding efforts, the ability to avoid giving up steals and just general organization and defensive hustle all play important roles.

Is the trade off worth it? It’s a good question. Brian Skinner[4. Look for a Freelance Friday piece from him this week!] examined one side of the question to determine how much each individual team might expect to gain from the marginal offensive rebound and had this general observation from his results (the whole thing is slightly mathy, but worth a read):

"the takeaway is this: an NBA team generally improves on offense by about 0.62 points per 100 possessions for each percentage point increase in its offensive rebound rate.  This means that if NBA teams were to improve their offensive rebounding from 23% (where it is now) to 30% (where it was a few years ago), they would generally score about 4.3 points more per 100 possessions.So now the remaining question is this:  are teams saving more than 4.3 points per 100 possessions by virtue of their improved transition defense?"

That is the operative question, but not one I’m going to address here, other than to suggest there are some reasons aside from simple concession of rebounding opportunities for the overall decline. As I mentioned above, the chances of both fast breaks & offensive rebounds on a given shot are heavily influenced by the location and identify of a shooter. Andre Drummond missing at the rim is more likely to produce an offensive rebound than Dirk Nowitzki shooting a three from above the break both because Drummond is about the best individual offensive rebounder in the league and because “stretch bigs” reduce offensive rebounding by pulling what would normally be a prime rebounder away from the basket. That is merely a long-winded way of saying this – the drop in offensive rebounding might be as much about who is shooting where as it is about teams committing to defensive transition over all else.

By combining positional lineup analysis with shot log data, we can see a dramatic shift even over the three years of SportVU data:

Dashboard 1 (46)
Dashboard 1 (46) /

While wing players and even centers have seen a slight uptick in the proportion of their shots coming from beyond the arc[4. The vast majority of these additional shots are at the expense of attempts from the inefficient “donut of doom” between the three point arc and five feet from the rim, so RIP midrange game some more…], the increase has been far more dramatic from the power forward position. Two things are probably going on here – one is the increase in small ball lineups, where guys who would be commonly thought of as “threes” are sliding up to power forward. The other is the increasing willingness of teams to allow even only moderately efficient jump shooting bigs to move back beyond the arc, whether this is Serge Ibaka, Paul Millsap or Kris Humphries. Both probably produce similar impacts, either a big man is taken out of offensive rebounding position, or a smaller player who is naturally less inclined to crash the glass is on the floor in his spot. It should be noted that this removal of big men from good offensive rebounding position doesn’t just occur on the threes they actually take. Exact measurements will be hard to come by even with perfect data, but the number of threes taken is almost certainly only a fraction of the offensive possessions a “stretch” big spends 20-some feet from the basket[5. This is the semi-hidden part of the argument in terms of why players like LaMarcus Aldridge or Blake Griffin have not yet added significant numbers of threes to their arsenal, as there are quite possibly costs to having such versatile, mobile players stationed around the arc which aren’t recouped just in those few times they are actually able to launch a three. Such a change could quite easily improve their individual efficiency while being deleterious to that of their respective teams.].

While data on the effects of contesting for and losing an offensive rebound on the ensuing defensive possession are somewhat equivocal[5. So far this season there is about a 2 points/100 difference in offensive efficiency in favor of plays coming immediately after a “contested” defensive rebound as opposed to an “uncontested” one.], it does stand to reason that the hopeless pursuit of an offensive rebound a player was never likely to get near probably compromises the defense more than already being in position to battle for that same board. Unfortunately, public SportVU data doesn’t quite allow us to identify these “bad chases” yet, but those also tend to be the sorts of things which drive coaches batty and earn players seats on the bench, which in its own way disincentivizes aggression on the offensive glass.