The writers at numberFire took a look at some advanced metrics at the halfway point of the NBA season and came up with a list of players deserving of a spot on the Western Conference All-Star team. The results may surprise you.—Ed.
by Russell Peddle / @rustypedalbike
Yesterday, I considered what it would look like if the All-Star teams were picked using an algorithm instead of the traditional method of fans voting for the starters, coaches picking the reserves and the Commish selecting injury replacements. The goal was to eliminate biases and to present a solution for the general public’s inability to decide on criteria that defines just what exactly constitutes an All-Star player in the NBA. To compile the rosters, I used our NBA Player Rankings and player nERD scores.
Part One dealt with the Eastern Conference nERD-Stars and the results were interesting, to say the least. Many of the familiar faces, such as LeBron James, Paul George and Carmelo Anthony, were there as expected. Where it got interesting, was in the inclusion of guys like George Hill, Andre Drummond and Anderson Varejao over players like Dwyane Wade and Kyrie Irving. Those types of selections would no doubt cause many people to oppose the idea of using metrics to select a team, but you know what they say; numbers don’t lie.
Make sure you check out the above link for a full explanation of the concept and the exact criteria I used in selecting the teams. Today, let’s take a look at how things shook out in the Western Conference.
Warning: There will be snubs.
Western Conference nERD-Stars
Starting Guard 1: Chris Paul (nERD 15.1)
Let’s just pretend that Paul isn’t out injured until after the All-Star break and that he’d be available for the game. The season he was having before going down would certainly warrant a selection. In 34 games before the shoulder separation, Paul was averaging a ridiculous 19.6 points, 4.6 rebounds, a league-leading 11.2 assists, 2.4 steals and 1.2 three-balls per contest.
The steals per game average puts him at third in the NBA, while he also places third in player efficiency rating (27.4), first in assist percentage (54.1), second in steal percentage (3.5), fourth in offensive rating (123.3), fourth in offensive win shares (5.1) and fifth in total win shares (6.9). He is one of the most sure-fire and deserving All-Stars that the NBA has to offer.
Starting Guard 2: Stephen Curry (nERD 10.9)
The redemption selection. Curry was last year’s most obvious snub, as no one put him on the All-Star team. The fans didn’t vote him in, the coaches didn’t pick him as a reserve, and even David Stern overlooked him as an injury replacement. This year, he’s even better and he’ll be impossible to leave off (the latest returns even have him overtaking Chris Paul for a starting spot).
He’s putting up video-game numbers (a term that’s quickly becoming a cliché, but is too apt to ignore), averaging 23.5 points, 4.6 rebounds, 9.2 assists, 1.9 steals, 0.3 blocks and 3.2 threes per game. He’s the league’s seventh leading scorer, second in assists per game and sixth in steals per game. He currently ranks second in assist percentage (40.9), 11th in offensive win shares (4.0) and eighth in total win shares (6.2). He takes shots from places most people have no business taking shots from and makes a large portion of them (career .434 shooter from downtown). He is exactly the type of young and exciting player that any All-Star game needs.
Starting Big 1: Kevin Durant (nERD 30.1)
Like LeBron, KD is an automatic selection no matter how you pick the teams. No one in their right mind would leave him off such a collection of the league’s brightest stars, especially not this season. He is having an MVP-caliber year and carrying the Thunder during the absence of his teammate, Russell Westbrook. Durant is currently averaging career highs in points (31.0), assists (5.1) and steals (1.5), while chipping in 7.7 rebounds, 0.8 blocks and 2.1 three-pointers per game. As if that wasn’t enough, he’s within range of his second consecutive 50/40/90 season, currently shooting .504 from the field, .413 from deep and .882 from the line.
He’s leading the league in scoring, player efficiency rating (30.7), offensive win shares (8.2) and total win shares (10.9). He’s also third in true shooting percentage (.640), 18th in effective field goal percentage (.558), third in offensive rating (123.7), 20th in defensive rating (100.3) and sixth in defensive win shares (2.8).
With Westbrook sidelined over the last 15 games, Durant has averaged a ridiculous 36.5 points per game, including performances of 48, 48, 54, and 46. He’s also currently on a streak of scoring 30 or more points in nine straight games. His nERD of 30.1 matches the highest of any player from the last 15 seasons (LeBron’s 30.1 from 2008-09 might even be passable before all is said and done). If he doesn’t win the MVP award this season, it might just go down as the best statistical season in the history of the NBA not to have an MVP trophy attached to it.
Starting Big 2: Kevin Love (nERD 18.6)
Kevin Love is having a bounce-back year after basically having to write-off his last season due to a recurring hand injury (you know, the one where he broke his hand doing knuckle pushups and then re-broke it). He’s averaging a gaudy 25.0 points, 13.0 rebounds, 4.1 assists, 0.9 steals, 0.4 blocks and 2.4 threes per game. After struggling mightily with his shooting last year, this year he’s got his split back up to .459 from the field, .386 from deep and .802 from the line.
He’s fourth in the league in points per game and second in rebounds. He’s also fourth in the league in player efficiency rating (27.2), third in defensive rebounding percentage (30.4), sixth in total rebound percentage (19.5), eighth in offensive rating (121.1), third in offensive win shares (5.8) and third in total win shares (7.9). He’s a bit of a liability on the defensive end, but his elite offensive numbers tip the scale in his favor every time. He’s a lock to make the All-Star team and will be for a long time.
Starting Big 3: Blake Griffin (nERD 11.0)
Known mostly for his high-flying dunks, Griffin is somewhat quietly putting together his best season as a pro and carrying the Chris Paul-less Clippers to a respectable record in the process (29-15 on the season, 6-3 without CP3). On the season, Griffin is averaging a career-high 22.6 points per game, while chipping in 10.0 rebounds, 3.5 assists, 1.2 steals and 0.6 blocks. He still shoots a solid percentage from the field (.524), but even more impressively is shooting .715 from the line (up from a career mark of .630). That’s nice, considering he has shot the third most freebies in the whole Association (351).
Free throws used to be one of the biggest knocks on him, but he’s clearly improving where needed and developing into one of the best frontcourt players in the whole league as a result. He’s currently ninth in the League in scoring, 12th in rebounds per game and 11th in player efficiency rating (22.4). Despite never being considered a lockdown defender, Griffin is currently tenth in the NBA in defensive win shares (2.4), leading to him being seventh in total win shares (6.3). He’s almost definitely getting voted in as a starter by the fans, which is one thing they are definitely getting right.
Reserve Guard 1: James Harden (nERD 9.7)
Since coming to the Rockets in a trade prior to the start of last season, Harden has had All-Star weaved into the fabric of his beard. All-Star games are offense-heavy, so a guy that’s fifth in the league in scoring (24.3 points per game), 14th in true shooting percentage (.598) and seventh in offensive win shares (4.3) is an All-Star in every sense of the word. The fact that he also chips in 4.9 rebounds, 5.4 assists, 1.4 steals, 0.6 blocks and 2.1 threes per game doesn’t hurt either.
He would probably be even higher in the rankings if it weren’t for his atrocious defense (defensive rating of 107.0), which has become a known issue. It’s especially troubling, considering he plays 38.6 minutes per game (second most in the NBA). Our metrics don’t look kindly on minus-defenders (especially those who play a lot of minutes), hence the slight drop. No matter, he is and will be an All-Star when the rosters are announced, even if his teammate Jeremy Lin is currently ahead of him in the fan voting. At least the current system allows for the coaches to right that wrong.
Reserve Guard 2: Damian Lillard (nERD 9.0)
Lillard, the reigning Rookie of the Year, is continuing to turn heads in his second year as a pro. He has been a key part of the Blazers surprising 31-11 season, averaging 21.2 points, 3.7 rebounds, 5.8 assists, 0.8 steals, 0.4 blocks and a blistering 3.1 three-pointers per game. His shooting splits have been impressive, as he’s gone .421 from the field, .430 from downtown (eighth in the NBA) and .891 from the line (fourth). The three-point percentage is particularly impressive, as he’s attempted the third most shots from deep in the whole league (305) and has made the second most (131).
He’s 13th in the Association in scoring average, 19th in assists per game, fifth in offensive win shares (5.0) and ninth in total win shares (5.7). There are far too many guards in the Western Conference that are deserving of an All-Star spot and who gets in and who doesn’t will be one of the most interesting developments when the final rosters are named next week. As far as we’re concerned, Lillard should be a lock.
Reserve Big 1: Anthony Davis (nERD 10.3)
While most second-year players are usually susceptible to a sophomore slump, Anthony Davis has squashed that concern by having an absolutely bonkers follow-up to his rookie campaign. His averages of 20.2 points, 10.3 rebounds, 1.3 assists, 1.5 steals and 3.0 blocks per game are out of this world. In fact, the only players that have ever put up numbers like that in their second year are Alonzo Mourning, David Robinson and Hakeem Olajuwon. If Davis keeps it up, he’ll be the youngest to ever do so. He’s leading the league in blocks per game and block percentage (7.1), while coming in fifth in player efficiency rating (26.5).
Davis could conceivably be snubbed from this year’s All-Star game, as the Western Conference is filled with elder statesmen who could garner lifetime achievement votes when coaches fill out the rosters. That would be an absolute travesty, as Davis is having one of the best statistical seasons of any player currently in the NBA. Our metrics have him placed clearly and deservedly as a member of the team. It’s only a matter of time before he becomes a regularity All-Star starter.
Reserve Big 2: DeAndre Jordan (nERD 9.9)
This is where the picks get more and more controversial. The Western Conference is absolutely loaded with deserving players and regardless of how we pick them, there will inevitably be some All-Star-caliber players sitting on the outside looking in. This is where using nERD to select the team is likely to cause some disagreement, because DJ is considered an All-Star above some unquestioned locks to get in (more on those guys in the snubs section). Honestly, the decisions coaches will have to make between some of these players will be darn near impossible, so this is one area where letting the numbers say what they may might not be the worst idea. In the case of Jordan, he is having an insanely good year, even if he’s not on most people’s short list of All-Stars.
He’s leading the league in rebounds per game (13.9), field goal percentage (.641), effective field goal percentage (.641) and total rebound percentage (21.6). On top of that, he’s third in the NBA in blocks per game (2.5), sixth in offensive rebound percentage (13.4), fifth in defensive rebound percentage (29.6), fifth in block percentage (5.4), sixth in defensive rating (97.2) and second in defensive win shares (3.2). I know that scoring is often the biggest factor in choosing All-Stars, but that résumé is almost impossible to ignore and should be considered before you scoff at this pick. There’s no chance he makes the real team, but that’s not to say his improved play and dominance in some categories has not earned him a spot in the conversation.
Reserve Big 3: Dirk Nowitzki (nERD 9.6)
Dirk has had a Hall of Fame career and will be a prime candidate for legacy votes to play in All-Star games until the day he retires. While some players in such situations continue to get All-Star nods, perhaps even past the point that they deserve to, Nowitzki has been enjoying a renaissance season that is more than worthy of the invite to New Orleans. For the season, Dirk is averaging 21.1 points, 5.9 rebounds, 2.9 assists, 1.0 steals, 0.7 blocks and 1.6 threes per game.
He’s also within range of having the second 50/40/90 season of his career (the first being 2006-07), currently shooting .479 from the field, .393 from downtown and .901 from the line (third in the NBA). Don’t be fooled into thinking that’s a common occurrence just because both Durant and Dirk are in the running this year. There have only been 12 such seasons that a player has pulled that off when shooting enough of each type of shot to qualify. It’s a pretty big deal. Speaking of big deals, Dirk is currently ninth in the league in player efficiency rating (23.2) and should be going to his 12th All-Star game.
Wild Card 1: Nikola Pekovic (nERD 9.1)
In what will likely be one of the most controversial picks of the two nERD-Star teams, Nikola Pekovic earns the final spot on the Western squad over some players whose omission might actually be deemed a criminal offense. The Big Montenegrin, however, is quietly having an excellent season and probably isn’t as farfetched as you might think as a candidate. He’s averaging career highs in points per game (18.2) and rebounds per game (9.2), while chipping in 1.0 assist, 0.6 steals and 0.5 blocks. Those peripherals don’t scream All-Star, but the efficient .532 shooting from the field (12th in the NBA) and .747 from the line do.
He ranks 17th in the Association in rebounds per game and eighth in offensive rebound percentage (13.0). He’s also 15th in the league in offensive win shares (3.8) and 16th in total win shares (5.2). That’s rather impressive, considering he plays on the 20-21 Tiimberwolves. It basically goes without saying that this pick won’t even come close to happening in real life, especially when you look at the list of snubs below and consider that many of them are all but guaranteed to make the real All-Star team. Even so, Pek is having an under-appreciated season and shouldn’t be as far out of the conversation as he currently is.
Wild Card 2: LaMarcus Aldridge (nERD 9.3)
Woah, that was close. LMA is having the kind of season that has people talking MVP, but he only made our nERD-Stars by a hair. Obviously the MVP talk is warranted, considering he’s averaging career highs in points (24.2), rebounds (11.6) and assists (2.9), while chipping in 0.9 steals and 1.0 block per game. He shoots very well from the line for a big man, (.816), but his value probably dipped as a result of his shooting a relatively average .476 from the field at such a high volume (a league-leading 21.1 attempts per game).
Our metric penalizes high-volume shooters whose efficiency rates aren’t great, which explains drops in value to people like LMA and Carmelo Anthony. That aside, Aldridge is tearing up the league, currently ranked sixth in scoring, fifth in rebounds per game and eighth in player efficiency rating (23.5). He was considered a borderline All-Star the last two seasons, but there’s little doubt that he’ll be a part of the Western squad this season. He has made the leap from star to superstar this year, leading his surprising Blazers to the fourth best record in the league at 33-11, and is about to receive all the accolades that come with such a status.
Notable Snubs: David Lee (nERD 9.0), Dwight Howard (nERD 8.7), Goran Dragic (nERD 8.7), Wesley Matthew (nERD 8.5), DeMarcus Cousins (nERD 6.5), Tony Parker (nERD 5.9), Mike Conley (nERD 5.8), Tim Duncan (nERD 5.7), Russell Westbrook (nERD 1.5), Kobe Bryant (nERD -2.5).
Wow. That list of snubs could actually be an All-Star team on its own and probably still beat the Eastern Conference team (seriously, the disparity of talent between the two conferences is bordering on ridiculous). Much like the Eastern Conference snubs from yesterday, this list contains players that are very likely to make the real team (Howard, Cousins), some who have a legitimate shot (Lee, Dragic, Matthews, Parker, Conley, Duncan), and some who would’ve made if it hadn’t been for injuries taking up a large part of their season (Westbrook, Bryant).
The biggest flaw in this whole idea might be the fact that DeMarcus Cousins didn’t make the team. He’s averaging career highs in points (22.6), rebounds (11.6), assists (3.0), steals (1.8), and blocks (1.2). What hurts Boogie with our metric, like Aldridge, is his high usage rate (league-leading 33.0) combined with relatively low efficiency numbers for a big man (field goal percentages of .488).
What’s important to note is that his nERD of 6.5 is still very good, but guys like Jordan and Pekovic get the nod because they are highly efficient players, which our metric favors. Also worth mentioning, there no Spurs on the team, despite the fact that they’re currently the third best team in the league. That’s probably a testament of their total team effort and contribution more than anything else.
This system is obviously not perfect. Some people will be quick to disapprove because of some glaring superstar omissions (Wade, Irving, Howard, Cousins) and some unexpected inclusions (Hill, Drummond, Varejao, Jordan, Pekovic). This is clearly not a system that is likely to be implemented in the near future (if ever at all), so it’s important to look at it with a critical eye rather than just dismissing it right away.
Maybe the names that surprised you are having a better season than you originally realized. Our attention tends to gravitate towards big numbers in the popcorn categories (points, rebounds, assists, etc.), but advanced stats have come a long way in increasing our understanding of the game. Efficiency in scoring, defensive ability, and a player’s overall contribution to his team are factors that don’t jump out as readily in a standard box score, but are important to consider nonetheless.
Almost no one in the world would pick Andre Drummond as an All-Star over Carmelo Anthony, but are we really right to consider the gap so big? When you put all the numbers on the table and consider the ways that each player can impact his team, both positively and negatively, there are a lot of things to consider that go beyond a 15-point discrepancy in scoring average.
I’m a company man, so of course numberFire’s nERD metric was my vehicle of choice for this ride to parts unknown. This same exercise would be interesting to try with any of the many other overall impact/efficiency stats out there, such as player efficiency rating, win shares, or player impact estimate. This method of selecting All-Star teams might not be perfect, but its flaws are different from the current practices and ultimately could be refined over time.
Regardless of how the teams are chosen, when the All-Star starters are announced tonight and the rest of the rosters are revealed next week, there will be controversy. There will be a lot of snubs and discussion about the current selection process and its validity. I’m not saying this idea is perfect, but I for one could certainly accept a formula being used that could evolve and improve over time instead of the system we currently have. At the very least, we could definitively say that any given year’s All-Star rosters represented the 12 statistically best players from each conference that season, without a million asterisks about team affiliation, previous career accomplishments, and injuries.
NumberFire is a sports analytics platform that uses algorithmic modeling to better understand sports. Follow NumberFire on Twitter at @numberfire and Facebook.