When you look at Andre Drummond, it’s hard not to think the 6-foot-11, 280-pound specimen was created in a lab by hoops-loving scientists. When evaluating Drummond’s physical prowess for a piece this past April, ESPN’s Tom Haberstroh wrote that Drummond “might be the most mechanically perfect body in the NBA.” Haberstroh based that hyperbolic-sounding sentiment off of the results from Drummond’s summer trip to sports science lab, P3 Peak Performance, where it was discovered Detroit’s big man actually moves with the dexterity of a much smaller player:

"He checks off all the same boxes that the perimeter players do," Dr. Marcus Elliott, the founder and lead scientist at P3, told ESPN. "That's super rare for a big man. You almost never have a guy that big who's comparable to a 200-pound guy. There's usually a give back."

You would think that with athleticism like that, Drummond—who still struggles on the offensive end, and at the free throw line (he’s shooting a career high 44.4 percent!)—would be a dominating defender. Except, through the first half of the 2016-17 season, the Pistons give up over 9 points more per 100 possessions when he’s on the court, according to NBA.com. The Pistons as a whole have a negative 6.4 net rating (the difference between defensive and offensive rating) when he’s playing, and a positive 4.7 rating when he’s on the bench. As mentioned, that alarming discrepancy stems from his defense. 

ESPN’s Defensive Real Plus-Minus (DRPM) has Drummond ranked No. 20 among centers, behind teammate Aron Baynes and in front of floundering New Orleans backup, Omer Asik. Yes, the single-season DRPM uses noisy box-score data early in a season, but ESPN’s regressive real plus-minus formula is not entirely without merit. But that’s not all that shows Drummond’s defensive shortcomings.

Drummond led the league in rebounds per game last year (14.8), and he’s third this season at 13.5 per contest, but the Pistons actually snatch a higher percentage of defensive boards when he sits: 82.2 percent when he’s off, compared to 79.2 when he’s on, per NBA.com. Not only that, but he’s averaging a career-low in blocks per game (1.2) even after swatting a career-high seven against the Kings on Tuesday last week. So what gives?

Why is one of the greatest physical specimens in the NBA today, someone who seemed to be ascending into superstar status earlier last season, dragging his team’s defense down so much this season? 

It’s not like he let himself go over the summer after signing a five-year, $130 million maximum extension in July.  He allegedly shed 20 pounds in the offseason. And it would be unfair to bring up his attitude because we don’t know him personally, and we don’t know his work ethic; we’re not in the locker room, or at practices, and we’re not in the habit of unsubstantiated armchair analysis. But there are some serious holes in Drummond’s defense, something anyone who watches the Pistons can see. More specifically, he’s failed to adapt to the high screen every team runs in the contemporary NBA.

These days big men shoot the three, and against Drummond they’re consistently popping for uncontested looks behind the arc, which is why they’re shooting over 41 percent from 3 against him, according to NBA.com’s player tracking data. Sure, there are some holes in that data, like everything, but lets use the aforementioned Kings game where he blocked a career-high number of shots as an example of what we’re talking about.

Sacramento’s superstar, DeMarcus Cousins, was 4-of-6 from downtown in Sacramento’s comeback win, and the only one of those four makes Drummond really contested was the last one (Aron Baynes was guarding Cousins on the 3-pointer not included in the video):

On both of the earlier 3’s, Drummond got distracted by a simple screen and didn’t get back in time to really contest. This is a recurring problem: he’ll get lost when there’s a lot of action in front of him, and when his man pops open, he’s caught flat-footed. Even when Drummond seems to play perfect defense, there are issues. Here’s an example on a possession against the Kings that actually ends with a tremendous Drummond block on a Garrett Temple dunk attempt: 

Unfortunately, Drummond’s earlier decision on the possession is what eventually led to him having to save it with an athletic block. 

On the original pin-down action to spring Rudy Gay, Drummond switches instead of letting Tobias Harris fight through to keep marking the curling Gay. As a result, the most dominant low-post player in the game, Cousins, got to go against a svelte Harris on the block, and Drummond was forced to leave Gay entirely to double. Cousins’ ensuing pass to a now-open Rudy Gay beyond the arc was deflected, but Gay still got an open three attempt, and Cousins got the offensive rebound over an out-of-position Drummond and a smaller Harris. Plus, there were a lot of other open Kings players as Drummond’s teammates tried to make up for his superfluous switch.

Drummond also struggles with Detroit’s defensive pick-and-roll scheme. In Stan Van Gundy’s system, Drummond is supposed to drop back on the screen, but not so far that he gives up easy mid-range floaters once the guard turns the corner and the big isn’t rolling to the rim. And if the guard turns the corner and puts on some speed, Drummond isn’t adequately protecting against the drive. It’s like he’s stuck in no-man’s land when he drops back: unable to contest a pull-up, or thwart a drive to the bucket. In these six clips from earlier in the season, his high screen warts are pretty overt:

First, he prematurely reaches on Victor Oladipo after the speedy former Magic guard turns the corner. The result: a layup. On the next sequence, he drops back so far on Anthony Morrow, he gives up a little runner in the lane, which can partially be attributed to Kentavious Caldwell-Pope’s inability to get over the screen to pick up Morrow, and the danger in leaving a rolling Enes Kanter open on the baseline. But this is Anthony Morrow we’re talking about, not exactly an elite level finisher off the dribble. The communication needs to come from Drummond because KCP can’t see Kanter’s screen.

Then there are the final four sequences of the video where, after a high screen, Drummond is supposed to be corralling Jamal Crawford and Dwyane Wade—two of the best mid-range jump shooters in the NBA today. Sure, the drop-down defense favored by Stan is supposed to entice these low efficiency 2-pointers, but Drummond doesn’t contest their shots at all, which makes them wide open high-efficiency attempts, especially when it’s Wade or Crawford taking them.

But Drummond’s issues go beyond his inability to keep track of 3-point shooting bigs, or blockading against a drive or an uncontested pull-up when the guard turns the corner in a pick-and-roll. Remember that Drummond has had a full two and a half seasons to figure out where he and his teammates need to be in Stan’s system, but his attention simply wanes too often, as in these next two examples.

Similar to what he did with Oladipo above, watch Drummond overplay on a reach for the ball, giving up an easy Robin Lopez spin for the layup: 

Or, look at him fail to contest as he stays under the rim even with a wide-open George Hill coming at him inside the paint for a floater:

Some of this is fixable, but it’s unclear how long it will take, or if Drummond’s lack of focus is just part of the package. Despite his age—just 23, seven months older than Joel Embiid—he’s been in the league for half a decade, and he’s spent half of that time with a real coach who knows a thing or two about dominating defensive big men (i.e. when Dwight Howard was winning Defensive Player of the Year awards and competing for the league MVP). Yet here we are, looking over a lot of evidence that Drummond might not ever become the defensive centerpiece the Pistons need him to be if they ever want to compete with LeBron for an Eastern Conference crown.

It’s probably premature to suggest Drummond might get dangled before the Feb. 23 trade deadline, but considering how he’s looked in the season’s first half and the fact he’s on a max deal already, it’s not very clear there’d even be that much of a market.