Andre Drummond’s lack of playing time masks a deserving All-Star
Remember Tom Sunnergren? Of course you do. He’s that nut who picked Andre Drummond for the All-Star game. He graciously offered to expand on his thoughts here.
-Dan
When Andre Drummond entered the league via the ninth pick in June’s draft, the only consensus on him was this: the Husky was a boom-or-bust proposition.
Turns out he’s boomed.
Just 42 games into his professional career, Drummond ranks third in the league among centers in offensive-rebounding rate, seventh in defensive-rebounding rate and third in overall rebounding rate. He’s 10th in true shooting percentage (eight places above Brook Lopez), and he has been even better by the reckoning of the catch-all advanced stats. According to PER he’s been the 11th best player in the league on a per-minute basis, and by way of Wins Produced – which, granted, isn’t for everyone* – of all the players in the Association who have logged more than 30 minutes, he’s been the most productive.
*Ed: It should tell you how much I respect Tom’s work that I take him seriously even though he’s a Wins Produced believer.
And then there’s his periodically excellent defense—which isn’t accounted for in the above metrics. (Wins Produced makes a team defensive adjustment, but doesn’t account for individual efforts beyond blocks, defensive rebounds, and foul avoidance.)
So I think he’s an All-Star.
The Internet doesn’t.
There are two arguments against Drummond’s (now-expired) All-Star candidacy, both of which hinge on his playing time.
1. He’s been great, but in 20 minutes per game, he just wasn’t on the floor often enough to justify a berth and
2. Not only is 20 minutes a night not enough time to make an All Star impact, but playing only 20 minutes a night is actually what’s driving his efficiency. He couldn’t do it over 30 or 35.
Both of these straw men are wrong.
As for No. 2: Although there’s some scuttlebutt that Drummond has conditioning issues (I couldn’t really speak to this, or the impact it would have on him if he played more—except to point out that in the two games this season in which he played more than 30 minutes, Dec. 26 against Atlanta and Dec. 5 against Golden State, he put up a 15/12 and a 16/12), as a general proposition there isn’t really any compelling evidence that reserves play worse as their roles expand.
Furthermore, the idea that he’s posting these numbers against backups doesn’t sway me much either: one, everyone plays against backups for extended periods, two, the backup/starter distinction is, most of the time, a function of offensive rather than defensive abilities (so he’s not necessarily posting his numbers against particularly weak defensive centers) and three, outside of Joakim Noah and Tyson Chandler, who are this tremendous defensive centers that starting big men have to labor against?
The other thing is this: Drummond’s been so good, that even if he lost considerable efficiency in playing an additional 10 minutes a night, he would still be one of the best big men in the sport.
Moving along to No 1.: let’s accept that Drummond is limited (though, again, I’m dubious) and is capable of playing only 20 productive minutes a night. Even if this is true, he does plenty in those 20 minutes to justify and All-Star berth.
Wages of Wins is more bullish on him than most, for the sake of simplicity, lets start off by assuming their valuation is accurate: Drummond produces, over the course of 48 minutes of playing time, .366 wins. (An average player produces .1 wins per 48 minutes.)
So, lets say we’re in a perfectly 50th percentile Drummond game and he plays for 20 minutes and produces, in the course of those 20 minutes, approximately .15 wins ([.363/48]x20).
In just those 20 minutes then, Drummond is giving the Pistons 150 percent of the production the average team gets at the center position over a full 48. (Teams win, on average, .5 games per game (clunky sounding, yes) and have five players on the floor at all times. Each position produces then, on average, 0.1 wins per 48 minute game. Drummond, in just 20 minutes, produces 0.15 wins.)
But what about the extra 28 minutes? If the Pistons get league average play at center for the 28 minutes Drummond isn’t on the floor, that gives them an additional .058 wins, and over 200 percent the production most teams get at the center position. (Even if Detroit plays a guy who’s well below league average during those additional minutes, which is likely, the Pistons will end up getting somewhere just under .2 wins a night from the center position. A team that gets that production at center and league average production at all other positions can be expected to win 60 percent of its games, or 49 in an 82 game season. This is the impact 20 minutes a night of Drummond would have on an otherwise completely average team.)
Okay. But Wages of Wins is probably off on him, right? They overvalue rebounding, and the system pumps out a lot of head-scratching figures for efficient big men. Forget about them. What a
bout PER?
Drummond is, as I type, 11th in the NBA with a 22.99 PER. To do a quick-and-dirty switch to numbers I’m a little more comfortable with (as in, very quick and very dirty) the 11th best WP48 mark in the NBA this season is .248. By this accounting, Drummond, in his usual 20 minutes per night, still gives the Pistons exactly 100 percent of the production most teams receive at the center position over the course of a full 48 minutes. ([.248/48]x20=.103). Anything they receive over and above this is gravy.
It’s difficult to look at season averages of 7.5 points, 7.4 rebounds, and 1.6 blocks and see this, but Andre Drummond is not just a promising up and comer. So far this season, he’s been one of the most productive players in the NBA.