As the NBA's metrics community gathers in Boston for the annual MIT Sloan Sports Analytics Conference, starting Friday, the growing impact of analytics on the league can be found in many places. From the number of teams employing analysts to advanced stats on NBA.com, it's hard to miss the influence of analytics. But the most interesting example might be playing out on the court as Kevin Durant and LeBron James battle for MVP.
Wait, the MVP race? While advanced metrics won't necessarily decide the vote, as has been the case with baseball's biggest awards in recent years, there's reason to believe that our newfound understanding of the game has helped lift both players to newfound levels of performance.
Unprecedented scoring and efficiency
Last month, Deadspin noted that both Durant and James are reaching historic levels in terms of their combination of scoring efficiency, as measured by true shooting percentage (TS%), and share in their teams' offenses, as measured by usage rate.
In the past, as shown by Evan Zamir, there has been a frontier beyond which NBA players have been unable to go. They could trade off fewer shots for better efficiency -- as James did last season -- or score more with a lower true shooting percentage, but never accomplish both at the same time. Until now.
I've measured the combination of usage and efficiency with a metric called adjusted true shooting percentage that estimates what the player's true shooting would be if he used plays at an average rate (20 percent of the time). Translating TS% from past seasons to this season's NBA average to account for fluctuations in offense levels league-wide, the cap on adjusted TS% is about .700, a mark nobody had reached before this season. As the chart below shows, James (.709) and Durant (.703) are both on pace to surpass it.
BEST SCORING SEASONS SINCE NBA-ABA MERGER
Of course, there's a month and a half left in the season, and both MVP candidates could regress to some extent. At the same time, given Durant set the previous record in adjusted TS% last season (edging past Adrian Dantley in 1982-83) and James was in the top 10 in 2012-13, this is far more trend than fluke.
Death of the volume scorer
Thing is, Durant and James aren't alone. While their efficiency is far off the charts, two more of the league's top 10 scorers in points per game -- Stephen Curry (.601) and James Harden (.603) -- have posted TS% of better than .600 this season. Blake Griffin (.588) and Kevin Love (.597) aren't far behind. Among the NBA's 10 leading scorers, LaMarcus Aldridge of the Portland Trail Blazers (.513) and DeMar DeRozan of the Toronto Raptors (.523) are the two with below-average shooting efficiency.
This season's top five in points per game collectively score with a TS% that's 13.0 percent better than league average, just ahead of last season (12.9 percent) for the highest relative mark since the ABA-NBA merger. That hasn't always been the case, as the chart shows.
As recently as three seasons ago, the top five scorers were just 6.0 percent better than average. And high scorers were notoriously inefficient in 2000-01, when the top five were 4.9 percent better than average and the top 10 were 3.6 percent better than average. Not coincidentally, that's when Allen Iverson of the Philadelphia 76ers was voted MVP.
About that time, statistical analysts first started to gain a foothold on the Internet and began touting the importance of efficient scoring. If anything, that message has probably sunk in too well, and volume scoring -- the ability to create shots at average efficiency or slightly worse -- is now somewhat underrated. Better still is the ability to create shots at above-average efficiency, and the new math has helped stars do just that.
Superior shot selection yields historic results
The advances statistical analysts have made in understanding shot selection, and the superiority of 3-pointers and shots at the rim to midrange jump shots, have trickled down to players, and not equally. Rethinking shot selection has helped star players trade out those long 2-point attempts -- long a staple because they are asked to create their own offense rather than getting set up by others -- for 3-pointers or forays to the hoop.
That's evident in the way Durant and James have evolved in terms of their shot distribution, via NBA.com/stats.
STARS' SHOT SELECTION
James, who averages a little more than 17 shots per game, has over the past two seasons cut about three midrange shots in favor of one more try at the rim and two more beyond the arc. Durant's less dramatic transformation this season has seen him take about one fewer midrange attempt per game out of his 20-plus.
It's unclear how much analytics have played a direct role in influencing the league's biggest stars. Last season, Sports Illustrated's Lee Jenkins reported on Durant's use of a personal statistical analyst, but Durant recently said he let the specialist go because he was concerned about overthinking things on the court. Instead, Durant prefers to focus on the X's and O's.
Though James won't be confused with sabermetrics pioneer Bill James, he has embraced the value of high shooting percentages.
"When you're a young player, you cast up low-percentage shots, and you're not really involved with the numbers as much as far as field goal percentage and things of that nature," LeBron James told Grantland's Kirk Goldsberry last season. "As I've grown, I've made more of a conscious effort to become a more efficient player and I think it's helped my team's success over the years."
Whether Durant and James specifically altered their games because of metrics or not, their lessons have surely had an impact. In a league in which efficiency is valued more than ever, Durant and James stand as the ultimate tribute. The new NBA, shaped by analytics, has helped them reach heights no scorers in modern league history have reached before.