Even as Stephen Curry and LeBron James were dueling under the bright lights of the NBA Finals last season, the league’s collective presumption was that both would soon have to make room for Anthony Davis as the face of the NBA’s future. But by the time the postseason rolls around this year, almost certainly without Davis and his New Orleans Pelicans in it, that presumption may be due for a re-examination — and not just because Curry has only gotten more superlative since June.
Davis has stalled out this season. The Pelicans are significantly worse than they were a year ago, and Davis’s own numbers have endured a major setback. By Box Plus/Minus, his dip from +7.1 in 2014-15 (sixth in the league) to a mere +2.5 (33rd) this year is tied for the 11th-biggest single-season decline since the NBA introduced the 3-point shot in 1979-80.1
|PLAYER||YEAR||BPM||PREVIOUS SEASON BPM||CHANGE||LIKELIHOOD OF CHANGE|
It’s not all bad or all Davis’s fault
Some of these struggles can be explained, however, and some portion of Davis’s offensive drop-off — from a BPM of +4.2 at that end of the floor to +1.0 — may even be in the service of an expanding all-around game. For instance, during his first three NBA seasons, Davis scarcely ever wandered out to the 3-point line — fewer than 1 percent of his field-goal attempts came from beyond the arc those years — but this season 9.2 percent of Davis’s shots have been launched from downtown. He hasn’t been very accurate out there, converting about 32 percent of his 3-point tries (the league average is 35 percent), and it’s an obvious drain on his offensive efficiency when the alternative is making 51 percent of his 2-point attempts (the equivalent of making 34 percent on threes). But the eventual development of a reliable deep jumper would make Davis the most frightful scorer this side of Curry.
And many of Davis’s statistical markers are down because he’s being asked to do a little more with a lot less. Before the 2014-15 season, a simple projection of the Pelicans’ roster would have called for their offense to be 0.3 points better than average (per 100 possessions) even if Davis were swapped out for an average player;2 this season, the same process would have predicted an offense 1.3 points below average (again per 100 possessions) without Davis.3 Eric Gordon, Jrue Holiday and Ryan Anderson are a solid (offensive) trio to have on your side, but Davis has also shared the floor often with a steaming pile of Alonzo Gee, Dante Cunningham, Norris Cole and Omer Asik.
Downturns in the quality of a player’s teammates aren’t always associated with declines in personal offensive performance, but in Davis’s case, it does help explain some of why his usage rate has risen, his scoring efficiency is down and his turnovers are up. And in addition to Davis’s increased offensive responsibility, New Orleans has focused on getting him more low-post touches; according to Synergy Sports Technology, Davis’s number has been called on the low block about 1.3 more times per game than a year ago, with post-ups now making up over 18 percent of Davis’s total offensive plays. This helps account for why Davis isn’t notching as many helpers (per SportVU data from NBA.com, assists are about 20 percent less common on post touches) and may even explain why his rates of putbacks and other offensive boards are also down. (Anecdotally, Davis spent plenty of time following up on teammate misses last year; that has become less possible this season with the ball in his hands so often.)
But what about that defense?
With his combination of athleticism, mobility and length — including an albatross-like 7-foot-5.5-inch wingspan — Davis is the prototypical defensive big man for an era that favors speed over mass. It’s hardly unreasonable to expect a player with his off-the-charts attributes to at least anchor a solid defense, and during his first three seasons, Davis was on track to do just that. Over that span, he tied for the seventh-best defensive BPM of any player aged 21 or younger since 1973-74, and his defensive Real Plus-Minus4 was even better. He was, without question, following the same defensive path laid out by Tim Duncan, Dwight Howard and Kevin Garnett early in their careers.
This season, though, Davis hasn’t traced the same trajectory. At age 22, Duncan, Howard and Garnett were averaging a defensive BPM of +3.0,5 and their teams averaged a defensive rating 3.3 points better than the NBA average per 100 possessions. By contrast, Davis’s defensive BPM is only +1.5 — his RPM, at +1.9, isn’t much better — and his Pelicans are carrying a defensive rating 2.5 points worse than the NBA mean.
It should not come as a surprise to hear that the rest of the Pelicans roster stinks defensively. The guards are mostly sieves (here’s looking at you, Holiday, Gordon and Norris Cole!), Davis’s frontcourt partner Anderson is one of the league’s worst defensive bigs, and the plodding Asik, the only other Pelican capable of defending the paint, has been hobbled by injuries and is logging only 17 minutes a night in Alvin Gentry’s up-tempo scheme anyway (more on this later). But Davis’s teammates were equally putrid on D last year, and the team’s overall defensive rating has only gotten worse since then.
Besides, the whole point of a metric like RPM is to filter out the distorting effects of a player’s teammates using a complex, regression-based methodology to isolate individual performance. And those stats say Davis’s defense is trending in the wrong direction: Among big men with as many minutes as Davis this season, only OKC’s Serge Ibaka has seen a bigger drop in defensive RPM since last season.
A redefined defensive role
So clearly something is off with Davis and the Pelicans’ defense. But at the granular level of simply assigning plays to individual defenders and tracking the efficiency they allow, it’s tough to account for the decline. According to Synergy, the distribution of play types that Davis defends against hasn’t changed much from a year ago. More than two-thirds of his defensive tasks still involve either grappling with pick-and-rolls (he switches to the ballhandler around 70 percent of the time) or closing out on a shooter who has inched away from the paint for a spot-up jumper, and he’s improved against both on a per-play basis. And although his post defense has been slightly less efficient this year, Synergy still classifies it as “average.”
In other words, Davis’s individual defensive data points haven’t changed much from last season. But that’s also kind of the point — given the seismic shift that Davis’s defensive role has undergone, his numbers should have changed. In a roundabout way, that they haven’t helps explain why New Orleans’s defense has backslid.
Simply put, these aren’t the same Pelicans as last season. Led by Gentry, they’re a much smaller, faster team than they were under Monty Williams in Davis’s first three seasons, and that’s required Davis to log far more time at center than he did a year ago. Back then, Davis spent more than half his minutes sharing the floor with Asik, a highly traditional defensive center who ranked among the league’s better rim protectors. Davis himself was below average by SportVU’s rim protection metrics, but he and Asik clicked as New Orleans’s best defensive combo because each made up for the other’s shortcomings.
By contrast, less than a third of Davis’s court time this season has been spent alongside Asik, often leaving Davis as the Pelicans’ sole rim-protecting presence inside. It’s a role in which he’s improving — getting close to average this season, in fact, according to Nylon Calculus’s metrics — but still one to which he’s not fully suited. The situation is made worse, of course, by the fact that nearly half of Davis’s minutes have come manning the 5 alongside a grievous defensive liability like Anderson. As APBRmetricians, we still don’t have great ways to account for these kinds of changes to a player’s role, particularly on defense, where with-or-without-you style plus-minus measures are so central to our method of evaluation.
Earlier, I mentioned Duncan and Garnett as prototypes for Davis. But their best defensive teams — Duncan’s Spurs in 2003-04 and Garnett’s 2007-08 Celtics — had them playing power forward alongside lumbering behemoths like Rasho Nesterovic and Kendrick Perkins, much like Davis and Asik a year ago. Those combos worked in a league ruled by size, but they couldn’t be used for more than brief stints against, say, the 2015-16 Warriors. To compete in today’s NBA, Davis doesn’t get the luxury of playing power forward alongside a bulky pivot anymore.
Instead, perhaps Howard’s 2008-09 Magic offer a template for the type of team that could eventually thrive around Davis, with a single dominant defensive center surrounded by tons of floor-spacing shooters. (Coincidentally, Anderson bridges the gap between Howard’s Magic and Davis’s Pelicans, having played a sort of Rashard Lewis 2.0 role on Howard’s final Magic squad.) Howard at 22 was much more of a rim protector and one-on-one post defender than Davis is now, so Davis still has a bit of work to do in those areas. But if everything goes right, he could blossom into a more versatile offensive player than Howard with a similar level of defensive impact.
To do that, though, Davis first has to ride out the growing pains of this season. And if his game successfully adapts, Davis could find himself in the finals limelight like Curry and James sooner than we might think.