All Andre Drummond Does Is Rebound.
We’ve all heard it before. Any conversation about Andre Drummond comes around to this point with such consistency that his name may as well be changed to Andre “all he does is rebound” Drummond.
And this is a fair assessment. The Pistons have been mediocre for a so long, and one constant has been Andre Drummond. Look through the past near decade of NBA basketball and you see the ever declining value of rebounding. The Heat didn’t become champions until they realized they didn’t need to rebound if they just put Chris Bosh at center and blew away opposing bigs on offense and smothered teams into turnovers on defense. The Warriors often employed a traditional big as a starter, but they reached their mythical status as a god-tier team when they slotted Draymond Green at center. Green may not be a shooter, but he’s a mediocre rebounder (as centers go) who makes it work by running opposing bigs off the floor and wrecking havoc on defense. Last season’s Raptors were closer to traditional, but Marc Gasol has actually spent his entire career as a mediocre at best rebounder, who instead brings value with whip-smart defense and being one of the more offensive skilled bigs of the decade.
The only team to win a championship in nearly a decade with a traditional center taking the bulk of the minutes there was the Spurs, and they had Tim-freaking-Duncan.
As teams around the NBA start to punt further and further on rebounding in favor of putting more speed and skill on the floor, it is reminiscent of the death of the enforcer in the NHL. And the Pistons should try to avoid being the old-man yelling at a cloud about how in his day rebounding is what won championships while the NBA champions start a team of all 6’7 wings who can shoot and handle the ball.
And to be clear, this experiment isn’t even looking at what the Pistons could’ve done if they swap Drummond out for a similarly paid player at different position and played someone else at center. In fact, given how far over the salary cap they were last season, it would’ve essentially been impossible to do so regardless.
On top of that, introducing say, Bradley Beal, to last season’s Pistons changes so many variables that it is pure speculation. But let’s dive into the numbers of last season to show the full degree to which Drummond fails to impact the game outside of his rebounding, and how much of his production could’ve been replicated by someone far cheaper.
So who’s our new center?
Shaqtin’ a fool champion JaVale McGee. The flint native was a free agent last summer (and this Summer) and has somewhat become a poster-boy for the benefits of a cheap rim-running center. McGee can replicate 80-85% of far more expensive centers at a fraction of the cost. He blocks shots, has a high work-rate, rolls to the rim, and catches lobs. He is the low-usage, hyper efficient, never needs the ball player so many people wish the Pistons had as a center. He does everything that Drummond does while being just a average(ish) rebounder for a center.
And before we get into this, as much as possible we will use the same numbers each player produced the past couple of seasons. So lets dive into just how low the impact of Drummond was for the Pistons last season. Also know that some of the comparisons will be over-simplified for the sake of, well, keeping this as simple and straightforward as possible.
This is where the difference is going to be the most stark. The Pistons have consistently been a wildly inefficient offensive team in Drummond’s tenure and Drummond is a inefficient player who takes a large chunk of their shots. It doesn’t take a genius to figure out what’s going on here. Gone will be the terrible jump-hooks and wild layups, replaced by McGee only dunking the ball when it is lobbed to him, and otherwise letting other, more capable, players take shots.
Drummond took 13.3 shots from the field per game last season and 5.2 free throws per game. This comes out to about 15.9 possessions with a shot attempt for Drummond, as opposed to JaVale taking just 8.5 shots from the field and 2 from the line for a total of 9.5. That’s a huge difference.
Does JaVale put back offensive rebounds nearly as often though?
Oh that’s right no he doesn’t. Of course JaVale, just in general, doesn’t get nearly the number of offensive rebounds and Drummond. Drummond averaged 5.4 offensive rebounds per game last season while McGee averaged 2.6. While Drummond doesn’t go back up with 100% of his offensive rebounds, it’s pretty close, and he also gets fouled on them a lot. Those 5.4 rebounds per game are good for 5.5 2nd chance points per game, best in the NBA by a wide margin, and that doesn’t even include the times he’s fouled on them. Then, even though McGee does occasionally like to back up himself, it isn’t as regular and we are just trying to prove a point here as much as anything so let’s just say none of JaVale’s offensive rebounds are put back up by him to maximize the number of shots we have to distribute.
Once you take away the 5.4 putbacks per game, Drummond is now taking 10.5 shots purely within the offense.
So by redistributing Drummond’s shots to other players the Pistons are maybe creating one extra shot attempt for other players per game?
Well it seems that way. Turns out that when you consider that a huge amount of Drummond’s shot attempts and usage percentage are created by himself out of thin-air he isn’t actually a terribly high volume player or use that many possessions. He just extends so many of his own abilities that he ends up finishing a lot more. But I digress.
And besides, 1 shot per game doesn’t seem like a lot, but it makes a real difference! The margins for error in the NBA are tiny and the reality is that not that much separates good teams from bad teams. If the Pistons could’ve improved their net rating last season by even a single point per 100 possessions they would’ve been on par with the Spurs. So let’s get this one shot per game out of Andre’s inefficient hands and into someone better.
So who’s it going to be?
First instinct would be Blake but the reality is that his usage was already higher than it really ought to be and led to him breaking down late in the season. If anything we would want to take a couple possessions away from Blake as long as we are shuffling things up in order to get him to the playoffs healthy but we won’t do that right now.
Yes, of course. Even if it is just that one shot per game, the Pistons sweet-shooting youngster is the one to take it. His TS% last season was 56% which is better than Drummond’s was (also the exact same mark he had his rookie season) and should make a big difference over the course of a season, after-all, we need shooters to be taking more shots.
What was Andre’s TS% again?
Luke’s is .5% better? As in half a percentage point?
Yeah, that’s actually a good point. Maybe I got carried away with Luke’s sweet stroke and forgot that it’s possible to be efficient scoring the ball other than just shooting threes. Let’s have another look at where that shot can go.
Wayne Ellington had a TS% of 58.4%, which is a much more meaningful improvement over Drummond. The only downside is that Ellington is getting his shots created by others, mostly Griffin, and we’ve made it clear that we don’t want to put more strain on Griffin.
So maybe we give that to Reggie Jackson? No he was actually less efficient a scorer than Andre was last season and that was a career best TS% from Jackson.
In fact, other than Blake and Ellington (or Bullock before he was traded, for which the same issue of it putting further strain on Blake) there was no other rotation player other than Luke who had a higher TS% and Luke’s isn’t a meaningful improvement.
Well that’s a bit of an issue.
Yeah. It turns out that any shots that JaVale isn’t using, which isn’t many regardless, are probably not going to be more efficient than what they would’ve been with Drummond. So maybe there won’t be that addition by subtraction thing or, as I like to call it: The “Anthony Tolliver is shooting instead of Josh Smith” effect.
But those other shots will still be much improved! the 9.5 shots per game that McGee is taking instead of Drummond will be much better.
Yeah about that.
A huge amount of Drummond’s inefficiencies actually come from those shots we are no longer getting. Last season McGee was assisted on a incredible 78% of all his field-goals on his way to that stellar TS%. Drummond was assisted on just 57% of his shots, and a huge reason why his is so much lower is that all those put-backs are unassisted. Just as important however, is that Andre Drummond didn’t spend last season with LeBron James and the two prior seasons on a Warriors team with 4 All-NBA players. So, when separated from one of the greatest pick and roll ball-handlers ever and paired with Reggie Jackson, McGee’s efficiency will almost certainly take a small hit.
And here’s the thing, when Drummond returned from concussion and Dwane Casey realized that it was a bad idea to have him post up and instead used him almost exclusively as a roll-man, the same way our hypothetical McGee will be used, Drummond was assisted on 60.6% of his buckets after having been assisted on just 54% prior. AFter the change, where Drummond was getting something closer to the same looks McGee will be getting, Drummond had a TS% of 61.1%, and that is with him still getting all the offensive rebounds and still being assisted on 18% fewer of his buckets than McGee was last season.
What does that mean?
It means that, in reality, The lob offense and such that McGee can get involved with (and likely would be less involved than he was in LA since Casey was so anti-pick and roll) is actually just as efficient with Drummond. In reality, McGee will probably see a slight drop in efficiency, but since he has no outlet to create shots for himself like Andre’s offensive rebounds, he will just see his shot attempts decrease due to the combination of not playing with LeBron and Lonzo ball as well as now being in Dwane Casey’s anti-lobs offense.
So in reality, this will create even more shot attempts for others, but as stated before, there really isn’t anywhere for those shots to go that is going to be more efficient in any meaningful way. Give them to Blake and he breaks down even faster than he ended up breaking down, give them to Bullock/Ellington and the only way you are getting the efficiency you need is by putting that same extra wear on Blake that you can’t afford, and no one else was more efficient than Drummond and most were, in fact, less efficient.
So what are you saying?
The Pistons offense is going to end up being largely the same inefficient mess that it was anyways. Drummond had a lower efficiency because he takes a lot of shots that McGee can’t/won’t take. The shots McGee is taking are ones that Drummond if finishing at the same clip, so really you are shunting extra shots to others but everyone else is as inefficient or worse than Drummond. So, essentially, your team efficiency is probably staying right about where it was.
Wasn’t rebounding the whole point of this? What’s the deal with all the offense stuff?
Obviously to prove that you can slot McGee in for Drummond, change nothing else, and not get any less efficient as an offense. Even if there isn’t a meaningful improvement, you are not likely to get worse.
That doesn’t seem like the point you were trying to make
It’s what I got out of this I don’t know what else you could come away with other than “McGee can replace Andre on offense without losing efficiency.
Ok so what about defense?
Let’s just assume the same thing on defense as offense. That other than the rebounding, there isn’t any meaningful change in either direction. McGee isn’t as active or switchable as Drummond but he does block more shots. Drummond’s overall impact with his quick hands is likely better, and McGee is one of the few guys that even Drummond’s biggest detractors can’t argue is a smarter defender.
So while Drummond probably makes your defense better than McGee, let’s just assume that other than rebounding, McGee is just as much of a positive impact on defense. Basically saying the same thing as on offense, but I don’t think that it requires as much explaining as to how we got to this point on the defensive end, after all defense is so much simpler to quantify than offense is. It’s as simple as blocks, steals, and HEART baby.
Good idea, the whole point was the rebounding so let’s say everything else stays the same to isolate rebounding as the only variable.
Exactly. So now that we’ve established that the Pistons offense will not lose any effectiveness other than Andre’s rebounds with McGee, and that the defense won’t lose any effectiveness other than some rebounding with McGee, let’s finally get to the greater point of the limited value that Andre’s rebounding has on the team.
Ok lets hear it
Andre Drummond is, of course, not just a good rebounder. He is one of the greatest to ever live. His career total rebounding percentage of 24.4% ranks first all time and owns the 2nd, 4th, 5th, and 6th best non-Rodman total rebounding percentages ever recorded in a single season.
Wait why are we using some wierd advance stat instead of rebounds per game?
Rebounding percentage is the percentage of rebounds a player gets while they are on the floor. Meaning that it evens out for pace and how many missed shots there were. It’s a much more accurate way of looking at how good a rebounder someone is.
Ok makes sense
So Andre has a chance to go down as the greatest rebounder to ever live, and McGee is a average to mediocre rebounder. With his career rebounding percentage clocking in at 16.3%. Last season, with the one of the greatest rebounders ever, the Pistons were the 8th best rebounding team in basketball. Good but not great. As such, I’d say it’s fair (and also easy on the math) to say that going from Andre to McGee will make the Pistons a league-average rebounding team. The drop-off from player to player is huge, but Blake Griffin is better than most power forwards and Dwane Casey is the sort of coach who is good at keeping his team on the fundamentals. So those combine to make the Pistons league average.
There’s about to be a lot of math here isn’t there.
Yeah there is, if you don’t care about how we come to our final numbers you can skip this next section.
Ok will do.
The league average rebounding team last season, 15th in the NBA, were the Brooklyn Nets. The Nets, home of legendary center Ed Davis, were exactly in the middle of the NBA’s 30 teams when it comes to rebounding (and just one spot behind McGee’s Lakers which furthers that this is about where the Pistons would likely be).
Once again, for the sake of simplicity, let’s just use the exact rebounding numbers of the Nets for our new-look Pistons. Once again, while there will certainly be some impact, it should definitively prove that Drummond’s rebounding has minimal value, after all the Nets were a better team than the Pistons.
The Nets had a rebounding percentage of 50.1% last season. in terms of on-court impact, it meant that the Nets took almost exactly the same number of shot attempts (plus free throws) as their opponents did every single night. Using our basic formula for shot equivalents, field-goal attempts + (free throws/2). With that, the Nets took 102.45 shot equivalents per game, add their turnovers and the Nets essentially had 117.45 scoring chances across 100.8 possessions per game. In the same number of possessions, their opponents had 117.4 scoring chances per game.
So essentially, the Nets, by fighting to a draw on the glass every night, gets the exact same number of scoring chances per night as their opponents. So lets do a wee bit more math and look at what the Pistons would look like with that.
Using the same formula as above, the Pistons had an average of 113.65 scoring chances over 97.4 possessions. Their opponents had 110.95. Giving the Pistons just under 2 extra chances to score per night than their opponents.
With those 113.65 scoring chances, the Pistons scored an average of 107 points per game. That is .941 points per scoring chance. Meanwhile, the Pistons opponents scored 107.3 points per game, good for .97 points per scoring chance. Those points per chance are the numbers we are keeping. Doing this technically we would factor in turnovers independently to this but the Pistons were only +.3 per game in turnovers so for the sake of simplicity we will call that close enough to be even.
Each time the Pistons had a chance to score, they scored .94 points, each time their opponents had a chance to score, they scored .97. The Pistons managed to be nearly even in total score-line difference (just -.3 per game) largely because they got more chances to score. This is, of course what rebounding does for you. So in our new hypothetical, average rebounding, Pistons team they now have the same number of chances as their opponents. So we split the difference between them (comes out to 1.35 fewer chances for the Pistons and 1.35 more for their opponents) and the Pistons and their opponents are both getting 112.3 scoring chances each night. Since we’ve already established that the Pistons offense and defense is the exact same other than the rebounding, we can plug our per-scoring-chance numbers straight in.
So what are our findings?
If the Pistons are an average rebounding team who has the same number of scoring chances as their opponents, our Andre Drummondless Pistons are scoring 105.56 points per game and giving up 108.931points per game for a per game net rating of -3.369 (nice). Let’s rephrase this into something most of you will be more familiar with and make it per 100 possessions.
The Pistons are now scoring 108.4 points per 100 possessions and allowing 111.8 points per 100 possessions. That’s good for a net rating of -3.4 points per 100 possessions. That would’ve ranked 25th on the season, well behind the now 24th placed Wizards -2.7, but comfortably ahead of the outright tankers, starting with the Hawks in 26th place with a rating of -5.5 points per 100 possessions. The Hawks won 29 games and the Wizards won 32. So it’s fair to say the Pistons would’ve won 30 or 31 games.
Yep. That’s right, literally nothing else changes. The Pistons just go from Andre Drummond rebounding to being average rebounding, and they go from a .playoff team that won 41 games, to a not quite tanking team that wins 30. Nothing else changes.
I’ve been bamboozled.
Yeah. Imagine being so dumb that you think you could replace Andre Drummond with JaVale McGee and see minimal drop-off. Like, this is even making the HUGE stretch that nothing other than rebounding would drop off and they still become outright terrible.
Without Andre the Pistons are last seasons Wizards, one great player dragging a team with only a couple of other competent players above the outright tankers but no further.
This doesn’t even take into account Andre’s far superior passing and finishing in traffic on offense. It doesn’t even account for the fact that without Andre the Pistons no longer are great at keeping opponents out of transition by default.
Opposing teams are so scared of Andre on the offensive glass that they need to commit multiple guys every possession to ensuring they clean the glass which largely keeps them out of transition. This alone would have a hugely negative impact on the Pistons defense since transition opportunities are highly efficient.
On top of that, the Pistons are already a team that doesn’t create a lot of turnovers and if you get rid of Andre you get rid of your main source of turnover creation and general havoc. And before you think “Well JaVale won’t turn the ball over on offense as much” JaVale actually had a higher turnover rate than Andre.
Any remarks on this whole process?
Don’t take the numbers as gospel. It’s overly simplified, I purposefully avoided using stuff from synergy or anything just to hammer home how lazy you have to be to think that Drummond’s rebounding has minimal impact. This really should be as simple as looking at basketball-reference, seeing that the Pistons took over 3 more shot attempts per game than their opponents (and almost identical free throws) and know how important the rebounding is.
So if someone wanted to use all the fancy stats, and is more mathematically inclined than me, they could certainly find an exact number to roll out here, that isn’t what this is for. This is just to make a very basic look at what happens when you isolate the rebounding.
While making the Pistons a league average rebounding team without Drummond is the most arbitrary numbers decision here, I actually think it’s fair beyond the fact that it’s convenient to make the numbers straightforward. That isn’t that huge a drop-off as a team to go from 8th to 15th, Blake is a good rebounder but the Pistons were tiny everywhere else on the floor. If anything the Pistons would’ve probably dropped further than league average.
Lastly, this is not a answer to folk who think the Pistons should take Andre’s money and give it to, say, a high-level wing. Depending on who that wing player is that’s a fair argument. If the Pistons had to start JaVale McGee but they also now had Bradley Beal that would make the Pistons better. If they had Buddy Hield though they would probably not.
So, next time you hear someone say that all Andre Drummond does is rebound as a degradation of him, know that they are not a serious person. Because even if that were true, that Drummond truly does only rebound (which he doesn’t), that is of huge value.