NBA

2017 NBA nERD-Stars: Choosing All-Stars Through Advanced Analytics

A look at how this year's All-Star teams would shape up if they were selected by using advanced analytics.

The NBA All-Star reserves were named last night, and now we know the full rosters for the 2016-17 NBA All-Star Game taking place on February 19th in New Orleans.

This year's new voting process (which now includes input from the players and media in addition to the fans and coaches) seems to have worked fairly well (Zaza Pachulia didn't make it, after all), but there were still some pretty egregious snubs who have supplied plenty of fuel for debate among pundits and fans.

It's become a bit of an annual tradition here at numberFire to imagine a world in which the All-Star rosters are determined by advanced analytics rather than any kind of voting at all, with the idea that it might be a way to avoid such glaring omissions. To do so, we use our proprietary metric, nERD, to select our nERD-Stars squads.

If you're unfamiliar with nERD, it's a player rating that measures the total contribution of an individual throughout the course of a season, based on his efficiency. Comparable to win shares, this ranking gives an estimate of how many games above or below .500 a league-average team would win over an 82-game season with said player as one of its starters.

For example, Kevin Durant is posting a 13.9 rating this year as the top player in our NBA player power rankings. If he played on a team that started four league-average players alongside him, you would expect that team to finish roughly 14 games over .500 (48-34) based on his play so far this season.

For this exercise, we take the standard All-Star rosters for each conference (two starting guards, three starting frontcourt players, two bench guards, three bench frontcourt players, and two wild card selections each) and fill it out using the players with the highest nERD scores.

Without further ado, here are your 2016-17 NBA nERD-Stars (along with a few words on the players who made our team but weren't selected for the real deal, and vice versa).

Eastern Conference nERD-Stars

PositionPlayerTeamnERD
Starting GuardIsaiah ThomasBOS8.8
Starting GuardKyle LowryTOR8.5
Starting FrontcourtJimmy ButlerCHI11.4
Starting FrontcourtGiannis AntetokounmpoMIL8.7
Starting FrontcourtLeBron JamesCLE6.8
Bench GuardDeMar DeRozanTOR5.2
Bench GuardKemba WalkerCHA4.2
Bench Frontcourt*Otto PorterWAS5.7
Bench Frontcourt*Dwight HowardATL5.3
Bench FrontcourtKevin LoveCLE4.9
Bench Wild CardKyrie IrvingCLE4.2
Bench Wild Card*Myles TurnerIND3.9

* Denotes a player who qualifies as a "nERD-Star" but didn't make the actual NBA All-Star Team.

Western Conference nERD-Stars

PositionPlayerTeamnERD
Starting GuardJames HardenHOU11.2
Starting Guard*Chris PaulLAC8.5
Starting FrontcourtKevin DurantGS13.9
Starting FrontcourtKawhi LeonardSA11.8
Starting Frontcourt*Rudy GobertUTA9.6
Bench GuardStephen CurryGS8.2
Bench GuardRussell WestbrookOKC6.4
Bench FrontcourtDeAndre JordanLAC7.6
Bench FrontcourtGordon HaywardUTA7.0
Bench FrontcourtAnthony DavisNO6.8
Bench Wild Card*Nikola JokicDEN5.9
Bench Wild CardDraymond GreenGS5.0

* Denotes a player who qualifies as a "nERD-Star" but didn't make the actual NBA All-Star Team.

nERD-Stars Who Aren't All-Stars

Rudy Gobert, Utah Jazz
nERD: 9.6
Gobert is easily the biggest snub from this year's All-Star selections. He ranks fifth in our NBA player power rankings with a nERD of 9.6, trailing only Kevin Durant (13.9), Kawhi Leonard (11.8), Jimmy Butler (11.4), and James Harden (11.2). He averages 12.8 points, 12.6 rebounds, and 2.5 blocks per game, while shooting a whopping 66.4% from the field. He's easily a top Defensive Player of the Year candidate as the league-leader in defensive rating (97.8) and defensive win shares (3.6) but is no slouch on the offensive end either, as he ranks first in true shooting percentage (68.2%), second in effective field goal percentage (66.4%), and fifth in offensive rating (127.1) as well. His omission from this year's All-Star game is criminal.

Chris Paul, Los Angeles Clippers
nERD: 8.5
CP3 has been an All-Star for nine consecutive seasons, and his omission will eventually be a weird asterisk to explain away when listing his Hall-of-Fame credentials. He's played 36 of a possible 46 games this season, but the timing of his recent thumb injury got him left off the team entirely, when he should've at least gotten the nod with the understanding that there would be an injury replacement named for him later. Instead, he's somehow a player averaging 17.5 points, 5.3 rebounds, 9.7 assists, and 2.2 steals per contest, while shooting 47.1% from the field, 39.5% from deep, and 87.6% from the line, who isn't an All-Star. He's seventh in the league in player efficiency rating (27.4) and first in win shares per 48 minutes (.291), for crying out loud. Ludicrous.

Nikola Jokic, Denver Nuggets
nERD: 5.9
Second-year big man Nikola Jokic started the season off slowly coming off the bench for the Nuggets, but he has been an absolute beast since head coach Mike Malone took our advice (we assume) and inserted him into the starting lineup. In 26 games as a starter this year, Jokic is averaging 17.1 points, 8.7 rebounds, 4.1 assists, 0.9 steals, and 0.8 blocks in only 26.6 minutes per contest, while shooting 60.4% from the field, 34.0% from deep, and 79.2% from the line. His per-36 numbers on the year of 21.2 points, 11.6 rebounds, 5.4 assists, 1.0 steal, and 1.1 blocks suggest that he's got All-Star selections in his future, while the 65.4% true shooting percentage (fifth in the NBA), 61.9% effective field goal percentage (sixth), 128.2 offensive rating (first), and .234 win shares per 48 minutes (eighth) say he probably should've made it this year. He's a nERD-Star through and through, regardless.

Otto Porter, Washington Wizards
nERD: 5.7
You'd have a hard time finding anyone who even thought about thinking about mentioning Otto Porter as a potential All-Star this season, but he's having a monster year statistically, and that makes him worthy of a nERD-Star selection. His averages of 14.2 points, 6.5 rebounds, 1.6 assists, 1.5 steals, and 0.6 blocks per contest are kind of unassuming, but his shooting split of 53.3% from the field, a league-leading 45.6% from long range, and 78.5% from the charity stripe has him ranking sixth in the Association in true shooting percentage (64.1%), fourth in effective field goal percentage (62.6%), and second in offensive rating (128.2).

Dwight Howard, Atlanta Hawks
nERD: 5.3
Dwight Howard has eight All-Star selections to his name, but he's experienced a resurgence in Atlanta this year and probably deserved his first nod in three seasons this time around. He's played 41 of a possible 46 games for the Hawks this year, averaging 13.7 points, 12.8 rebounds, 1.3 assists, 0.9 steals, and 1.3 blocks per contest, while shooting a career-best 63.4% from the field (third in the NBA). He's also third in the NBA in total rebound percentage (24.2%), third in effective field goal percentage (63.4%), and fourth in defensive rating (100.4).

Myles Turner, Indiana Pacers
nERD: 3.9
Myles Turner has plenty of All-Star appearances in his future, but he's already secured nERD-Star status in just his second season. This year, he's averaging 15.7 points, 7.6 rebounds, 1.0 assist, 0.7 steals, and 2.3 blocks per contest, while shooting an ultra-efficient 53.1% from the field, 40.0% from deep, and 80.7% from the line. He's third in the NBA in blocks per game and first in block percentage (6.6%) and still hasn't even reached the legal drinking age in the US. His future is bright.

All-Stars Who Aren't nERD-Stars

Marc Gasol, Memphis Grizzlies
nERD: 4.8
Marc Gasol is a very deserving All-Star this season, rocking averages of 20.6 points, 6.1 rebounds, 4.2 assists, 1.1 steals, and 1.4 blocks per contest, and a shooting split of 46.4% from the field, 39.5% from long range, and 82.5% from the charity stripe. His 4.8 nERD caused him to miss the cut on our Western Conference nERD-Stars -- but only by a margin of 0.2. It's hard to criticize his inclusion legitimately, even if he's less efficient than Gobert or Jokic (who made it over Gasol on our squad).

DeMarcus Cousins, Sacramento Kings
nERD: 4.4
Like Gasol, it's hard to knock Boogie's inclusion in the actual All-Star game, but he missed the cut to be a nERD-Star by 0.6. His "meh" shooting split (44.9% from the field, 36.5% from deep, 77.3% from the line) and high turnovers (3.4 per contest) contribute to his low ranking by our standards, but it's hard to deny that his 28.0 points, 10.3 rebounds, 4.4 assists, 1.4 steals, and 1.4 blocks per game are All-Star-worthy numbers.

John Wall, Washington Wizards
nERD: 3.4
John Wall absolutely deserves his All-Star nod, based on his fat stat line of 23.1 points, 4.6 rebounds,10.1 assists, 2.2 steals, and 0.5 blocks per game, but he lacks the kind of efficiency it takes to be a nERD-Star. His shooting split of 46.4% from the field, 31.7% from three-point range, and 82.4% from the line leaves a bit to be desired, and his whopping 4.3 turnovers per contest represents the third highest average in the league after James Harden's 5.9 and Russell Westbrook's 5.6.

Paul Millsap, Atlanta Hawks
nERD: 2.4
Millsap is typically a nERD darling, but he's shooting a career-worst 44.4% from the field this season, and that's dragging him down. His averages of 18.0 points, 8.1 rebounds, 3.8 assists, 1.5 steals, and 1.0 block per game are All-Star-caliber, though, even with his slight dip in efficiency.

Klay Thompson, Golden State Warriors
nERD: 0.7
Klay isn't over-the-top efficient and doesn't really fill up the box score outside of scoring, but his 21.1 points per contest make him a household name and have earned him his third consecutive All-Star selection, even if he doesn't do enough overall to earn him any love by our nERD metric.

Paul George, Indiana Pacers
nERD: 0.6
PG13 has the lowest nERD of any All-Star this season, but we'll call that progress after Kobe Bryant made it with a -12.7 last year (the second worst mark in the entire NBA at the time). His averages of 22.2 points, 6.0 rebounds, 3.2 assists, 1.7 steals, and 0.3 blocks scream All-Star in the traditional sense, even if his declining defense, shooting split of 45.0% from the field and 38.8% from deep, and 2.8 turnovers per contest keep him from being worthy of a nERD-Star selection.