Metrics 101: Who's the NBA MVP?
Triple-doubles.
That's the last time they'll be mentioned throughout this article, because they shouldn't serve as the crux of any argument about the NBA MVP. Fun as they may be, there are plenty of metrics that tell us far more about the value of the league's leading candidates.
No one number can determine the best order on an MVP ballot. Context is necessary (as always), and plenty of different statistics help flesh out the resumes of James Harden, Russell Westbrook and the other contenders for the sport's most prestigious individual award.
The full argument for any one candidate would stretch out over many thousands of words, and there's no space or time for that here. Instead, we're boiling the arguments down to the key points, working from the honorable mentions to the five best choices for spots on official ballots but always keeping the focus on remaining objective wherever possible.
Most years, the MVP debate is tough. But that's not the case this year.
In 2016-17, differentiating between some of these superstars is nearly impossible.
Honorable Mentions
1 of 6
Jimmy Butler, Chicago Bulls
It was a season of improvement for Jimmy Butler, who scored three more points per game than he did in 2015-16 while bumping his true shooting percentage from 56.2 to 58.6 percent and still finding time to reattain his former excellence on the defensive end. Though this was a season of turmoil for the Chicago Bulls, their All-Star wing consistently functioned as a steadying force with his two-way prowess and ability to take over on the scoring end, particularly down the stretch of tight contests.
When Butler didn't play, the Bulls accumulated a minus-7.1 net rating, which would leave them tied with the Los Angeles Lakers for No. 30 in the season-long standings. But when he was on the floor, the Windy City played witness to a 3.0 net rating—a mark that would place No. 8 overall, just ahead of the Cleveland Cavaliers.
This isn't because he lucked into lining up with more talented teammates. After all, everyone else on the Chicago roster seemed to run together in a morass of mediocrity (or worse), while he stood out on a nightly basis.
"I think that's the most important thing. I thought winning is what this is about—period," James Harden said while trying to boost his own MVP candidacy above Russell Westbrook's, per ESPN.com's Calvin Watkins. "I'm not going to get in-depth with all that, but I thought winning was the most important thing. If you set your team up in a position to have a chance, at the ultimate goal, that's the most important thing."
But it's not that easy. If it were, Stephen Curry would automatically have a better shot than James Harden, since the Golden State Warriors won a dozen more games than the Houston Rockets. Importance to a team has to matter, and that's where another historic offensive season from Curry—he again averaged over 25 points with a true shooting percentage north of 62, which only nine other players have also done—drops slightly behind the pack.
Win shares are by no means a tell-all stat, but their intention has merit. By attempting to properly apportion real-life wins to individual players, they can show how valuable players were to specific teams. And according to win shares, as divided into a team's actual tally of victories, Curry finishes behind each of the most notable candidates and one of the other honorable mentions:
| Jimmy Butler | 41 | 13.8 | 33.7% |
| Giannis Antetokounmpo | 42 | 12.4 | 29.5% |
| Russell Westbrook | 47 | 13.1 | 27.9% |
| James Harden | 51 | 15 | 27.3% |
| Kawhi Leonard | 61 | 13.6 | 22.3% |
| Stephen Curry | 67 | 12.6 | 18.8% |
| Kevin Durant | 67 | 12 | 17.9% |
For the sake of comparison, Curry generated 17.9 win shares for last year's 73-win juggernaut. His 24.5 percent would've left him behind this year's leading candidates, but it's still a far more impressive finish.
Kevin Durant, Golden State Warriors
The Curry argument applies to Kevin Durant, as well. Except this is a slightly different situation, since the forward was a leading candidate for the coveted one-man award early in the season before slight regression and injuries yanked him from the top of the totem pole.
Durant's first season with the Warriors should be considered a massive success. He led the league in win shares per 48 minutes, posted the highest true shooting percentage of his remarkable career and made unmistakable strides on defense. Long regarded as an offensive juggernaut who could use his size advantageously on the preventing end, he was good enough to earn All-Defensive consideration while filling a versatile role that asked him to protect the rim on some possessions and switch onto guards on others.
But Durant's all-around profile isn't enough to overcome the knee injury that limited him to just 62 appearances. He finished 11th in ESPN.com's real plus/minus (RPM), but a lack of volume kept him from rising higher than No. 15 in RPM Wins. He wound up sixth in box plus/minus (BPM), but that same limited playing time prevented him from ascending past the No. 8 spot in NBA Math's total points added (TPA).
5. Giannis Antetokounmpo, Milwaukee Bucks
2 of 6
Per-Game Stats: 22.9 points, 8.7 rebounds, 5.4 assists, 1.6 steals, 1.9 blocks
Advanced Metrics: 26.0 player efficiency rating (PER), 4.16 real plus/minus (RPM), 425.68 total points added (TPA)
Giannis Antetokounmpo's advanced metrics all look incredible.
He ranks in the top 10 for player efficiency rating (PER), sits inside the top 20 for RPM despite playing with a group of teammates who were often learning on the job and trails only James Harden, LeBron James and Russell Westbrook in NBA Math's TPA. On a per-possession basis, he wasn't quite on the same level as the league's other superstars—and yes, he undoubtedly qualifies for such celestial status now—but he maintained that lofty level while playing more minutes than anyone not named Andrew Wiggins, Karl-Anthony Towns or Harden.
And despite the sterling nature of the more advanced statistics, his basic ones are still pretty darn good.
According to Stat Muse, Antetokounmpo is the first player in NBA history to finish in the top 20 for each of the major box-score categories: points, rebounds, assists, steals and blocks. He's also one of five players to pace his own team in each of the five, forming a quintet alongside Dave Cowens, Scottie Pippen, Kevin Garnett and LeBron James:
| 1st Place | Giannis Antetokounmpo (22.9) | Giannis Antetokounmpo (8.7) | Giannis Antetokounmpo (5.4) | Giannis Antetokounmpo (1.6) | Giannis Antetokounmpo (1.9) |
| 2nd Place | Jabari Parker (20.1) | Greg Monroe (6.6) | Matthew Dellavedova (4.7) | Khris Middleton (1.4) | John Henson (1.3) |
If you're looking for the definition of meaning everything to a team, well, there you have it.
4. Kawhi Leonard, San Antonio Spurs
3 of 6
Per-Game Stats: 25.5 points, 5.8 rebounds, 3.5 assists, 1.8 steals, 0.7 blocks
Advanced Metrics: 27.5 PER, 6.15 RPM, 383.56 TPA
Already, we've reached the point in the countdown where any remaining player would be a perfectly legitimate choice for MVP. Selecting Giannis Antetokounmpo would draw justifiably raised eyebrows, but Kawhi Leonard is a perfectly valid option for the top of the ballot.
In fact, he was the personal selection of ESPN.com's Zach Lowe, who addressed the concerns about Leonard's staggeringly poor on/off defensive splits in typically intelligent fashion:
"A ton of digital ink has been spilled about why San Antonio's defense was better with Leonard on the bench. Some of it is Parker and Pau Gasol. A lot of it is clearly luck. Opponents shot horribly on wide-open 3-pointers and foul shots with Leonard sitting. Maybe David Lee is really distracting. What has gotten lost: San Antonio's fatter points allowed per possession figure with Leonard on the floor, compiled mostly against the best opponent lineups, would still rank fifth overall at the team level. That is borderline elite.
"
The defensive splits aren't held against him here, though it's also interesting he's fallen from No. 9 in ESPN.com's defensive real plus/minus (DRPM) last year to No. 120 in 2016-17. The offensive burden he's assumed has had a legitimate impact on his off-ball defense, despite how easy it is to watch his on-ball work and think he should still be a leading Defensive Player of the Year candidate.
No, the explanation for his fourth-place finish here is simple: The top three have just been...better.
That's not a knock on Leonard. It's merely an admission that while he's playing at a Hall of Fame level, the three candidates ahead still managed to outpace him—by a razor-thin margin, in the case of our No. 3 finisher.
3. LeBron James, Cleveland Cavaliers
4 of 6
Per-Game Stats: 26.4 points, 8.6 rebounds, 8.7 assists, 1.2 steals, 0.6 blocks
Advanced Metrics: 27.0 PER, 7.52 RPM, 470.37 TPA
This isn't just about RPM or TPA. The former rewards LeBron James for the slightly elevated efficiency with which he played on both ends (despite taking some possessions off, he was a defensive demon in 2016-17), and the latter gives him credit for the absurd workload he took on, even if he occasionally sat out of a few contests.
James just meant everything to the Cleveland Cavaliers.
Leonard's on/off splits can be explained away by a number of factors, and it's undeniably impressive that the San Antonio Spurs scored an additional 10 points per 100 possessions while he was on the floor. But caveats simply aren't needed for the four-time MVP, who took his squad from a lottery afterthought to a championship contender.
According to NBA Math's FATS calculations, Cleveland played like a 26.2-win outfit whenever James was catching his breath on the pine rather than gracing the court with pinpoint passes and contributions in every area imaginable. When he played, that number of expected wins skyrocketed to 51.3—nearly double the original tally.
And lest you think this is at least partially because James tends to spend so much time alongside Kyrie Irving and Kevin Love, it's not. The other members of the Big Three certainly have positive impacts on the proceedings, but it's James who serves as the primary influencer.
When he plays without Irving and Love, the Cavs operate like a 55.4-win unit, suggesting that James operating as a one-man wrecking crew—admittedly, a small-sample-size warning is in effect here—might be the team's best option, fellow stars be damned. When those two play without him, though, Cleveland is expected to win just 30.2 games over the course of a full season.
That, in a nutshell, is impact.
2. James Harden, Houston Rockets
5 of 6
Per-Game Stats: 29.1 points, 8.1 rebounds, 11.2 assists, 1.5 steals, 0.5 blocks
Advanced Metrics: 27.3 PER, 4.75 RPM, 626.23 TPA
If MVP meant what it literally stood for, the conversation would've ended with LeBron James. But the meaning of the award is more arbitrary, allowing for wins and other factors to creep into the conversation. In reality, it's typically handed to the best player on one of the best teams, and narrative always matters.
However, even from a purely objective standpoint, Harden had a superior season.
While James' on/off impact is staggering, that's partially a function of the team around him. Harden doesn't boast the same type of differential, but he's responsible for the scheme in which he operates as well as the growing successes of his supporting cast. Unfortunately, it's impossible to quantify how the bearded guard's skill set allowed general manager Daryl Morey and head coach Mike D'Antoni to witness their dreams becoming realities.
On the flip side, it is possible to make sense of Harden's immense offensive impact. Chris Herring wrote this for FiveThirtyEight with a few days remaining in the regular season:
"Harden and fellow lefty Isaiah Thomas are about to become the first guards in history to average 29 points or more on 20 shots or less. Along with that, Harden is generating a league-high 27 points from assists per night. (He broke Steve Nash's record for the number of 3s a player has assisted in a season, which was 285, according to ESPN Stats & Information Group. Harden currently has 361.) In fact, with two games left, he could still eclipse Tiny Archibald's 44-year-old NBA record for most points produced per game, at 56.8.
"
Harden finished with 27.1 points created per game off assists and 29.1 scored with his own limbs, which doesn't quite move him past Archibald but does leave him in historic territory. Though Houston could replicate its system with a high-scoring bench mob, it was never better than when Harden was leading the show, racking up trips to the stripe, pull-up triples and impressive drop-off passes to rim-running bigs.
1. Russell Westbrook, Oklahoma City Thunder
6 of 6
Per-Game Stats: 31.6 points, 10.7 rebounds, 10.4 assists, 1.6 steals, 0.4 blocks
Advanced Metrics: 30.6 PER, 6.2 RPM, 890.62 TPA
Russell Westbrook broke a basketball statistic. Seriously.
Kevin Pelton wrote this for ESPN.com:
"To improve the quality of the rating for most players, BPM [box plus/minus] uses interaction effects that multiply a player's assist percentage by his usage rate and his rebound percentage. As you might guess, Westbrook's season is off the charts historically by both measures. Basically, Westbrook's season is way outside the sample on which BPM was trained to estimate player value, making its estimate of his value unreliable. BPM is treating Westbrook's versatility as exponentially better than anyone else's on record, and that's surely an exaggeration.
"
But still. Westbrook quite literally broke an advanced metric, which has produced fairly reliable results for as long as it's been possible to calculate—the numbers date back to 1973-74. And that means he also broke NBA Math's total points added, which is derived from box plus/minus and featured him with the best individual season since that aforementioned pre-merger campaign, taking down Michael Jordan's 1988-89 heroics in the process.
It's possible to spin this negatively, thinking that Westbrook's numbers are somehow overstated. But that would be denying his superiority in other metrics, basic per-game stats and the clutch, as well as downplaying the historic load he shouldered for the Oklahoma City Thunder.
Maybe he prevented some of the youngsters surrounding him from developing. But his one-man show still worked, propelling a limited roster into the playoffs while he posted the top usage rate in NBA history (by a wide margin) and the third-highest assist percentage in the annals, behind only two John Stockton seasons.
It's impossible to understand what the Thunder would've been without him, if only because he didn't give them enough opportunities to work while he wasn't racking up stats. And in this case, that's by no means a bad thing.
Adam Fromal covers the NBA for Bleacher Report. Follow him on Twitter: @fromal09.
Unless otherwise indicated, all stats from Basketball Reference, NBA.com, ESPN.com or NBA Math.
