The NBA Draft Is An Analytical Meritocracy

Teams have put their faith in analytics for the draft, but risk can never be fully removed


Like all major sports in the digital era, the NBA has put its faith in data analytics. The sport’s draft system necessitates proper analysis of prospects’ data, given the relatively limited supply of additional information for coaches to go on. No sooner had the dust settled in the Oracle Arena following the Cavaliers’ clutch win over the Warriors that attention turned to the draft and the potential picks began being evaluated and scrutinized.

Naturally, interest in the draft has always been huge, but the explosion of publicly available data gives fans extra material to discuss, objective information to ponder over and more substantiated reason to be excited about a prospect. Players will still be underrated and there’ll still be less-than-worthy prospects in round one - what the draft has become, though, is something of a brutal meritocracy. Number crunching has replaced gut feeling and young players push to improve their numbers in incredibly and increasingly specific areas of their game.

The thirst for reliable data in the decision making process is reflected by ESPN’s Draft Projection model which, as explained by FiveThirtyEight, ‘predicts how well a college player who is ranked among Chad Ford’s Top 100 prospects will perform - according to Statistical Plus/Minus (SPM) - during season two through five of his NBA career.’ The model is standard in the sense that it uses college data, physical data like weight and height and the player’s ranking in the Top 100 to build a picture of the prospect. Where it differs, though, is that it ‘acknowledges that NBA data on draft prospects is strongly left-censored, because very few prospects actually get a chance to play in the NBA at all, much less stick around long enough to get a meaningful sample.’

This is where the ESPN model is more reliable than its counterparts - the distinction is an important one. Each player is then put in one of four categories (though most players straddle two or three): Superstar, Starter, Role Player and Bust based partly but not entirely on the prospects projected SPM. The wealth of data available is frightening, and franchises can get a decent idea of how players will perform in future just by properly assessing the numbers. A player may have almost no chance of becoming a superstar, but it could be similarly unlikely that they become a bust - risk is both revealed and abated when data is properly analyzed.

Resultantly, clubs are taking decisions solely based on analytics, with increasingly positive results. The Denver Post use the example of Andre Roberson, drafted in the first round three years ago by Oklahoma City as a shooting guard, despite never really playing there for the Colorado Buffs. The gamble - based on Thunder general manager Sam Presti’s faith in analytics - has paid off, and Roberson is now a starter in a strong performing Thunder team.

If a player’s offensive rebounding percentage makes him more likely become a bust, clubs will know about it. If a player’s assist percentage gives them a genuine chance at becoming a superstar, clubs will know about this, too. In a sense, the draft is an analytical meritocracy, with no stone left unturned in predicting just how successful a college player could become. Risk will never fully disappear and neither will Roberson-style reward, but the NBAs seemingly growing faith in the numbers is validated by, well, numbers. 

University lecture small

Read next:

How Are Higher Education Institutions Using Analytics?