I had previously written about this topic in February 2020. This article is an update and expansion of what I wrote then.
With the rise of analytics and the global popularity of the NBA and basketball overall, NBA General Managers seem to have more resources at their disposal than ever before. Presumably, this paired with decades of history to learn from would have led to teams getting better at picking the best available player when it’s their turn to make a selection in the draft. We would hope that teams would have become less likely to select busts early in the draft or even just pick a decent player when a better player was available. However, NBA GMs haven’t gotten better at drafting over time — they’ve gotten worse.
If NBA GMs were absolutely perfect at drafting, the best player would always go first, the second best player would always go second, and so on. Perhaps there would be a bit of deviation from this order as GMs draft players who fit holes in their rosters, especially later in the draft when better teams are drafting. (A team with good starting and backup point guards might draft a wing who they expect to be slightly worse than an available point guard, for instance.) But in general, we will use NBA teams’ proficiency for drafting the best available player as a proxy for their overall drafting competence.
How well do NBA teams draft?
One could be tempted to look at much value NBA teams usually get from a particular draft pick. If a team with the 3rd pick drafts someone who ends up outperforming the average 3rd overall pick, you might say they did done a great job, whereas a team with the 4th pick that draft someone who underperforms the average 4th pick is considered to have failed. But not all drafts are equal, and a team that picks the best player in the draft with the 3rd overall pick did a good job even if that player underperforms the average 3rd overall pick, and a team who picks the 12th best player with the 6th pick messed up even if that player does better than the average 6th pick.
To measure how well NBA teams draft, we’ll need some objective (or as objective as possible) metric to determine which players had the best careers. There are a lot of metrics that attempt to measure a player’s total positive impact — Player Efficiency Rating, Wins Produced, Win Shares, and many others. For our purposes, we’ll use win shares, as it gives us an intuitive and mostly reliable idea of a player’s total career contributions on both offence and defense. We’ll look at the total win shares over a player’s career, as this signifies a player’s total ability to perform on the court, including his ability to stay on the court — the best ability is availability.
To measure how well GMs performed at picking the best available player (or at least the available player who would end up with the most career win shares), we can use Spearman’s rank correlation coefficient (“ρ”), which will allow us to compare the rank of the career win shares of a player among the other players in his draft class to the position he was selected. If teams drafted perfectly according to this metric, the player in a draft class who would go on to have the most win shares would be picked first, the player who would go on to have the second most win shares would be picked second, and so on. A perfect draft gets a score of one, and a draft done randomly would get a score around zero. This ensures that GMs aren’t penalized in draft years without top talent; as long as they drafted one of the best players available, the metric will report that they did well, regardless of whether the best player available was LeBron James or Hedo Türkoğlu.
Years where GMs did a better job of picking the top talent available have scores closer to one, whereas years where teams weren’t reliably picking the best available players will have scores that are lower. If teams have improved at drafting (and developing) players, we would expect to see our Spearman’s rank correlation coefficients get bigger over time.
We’ll look at the first rounds of NBA drafts between 1979 (the year the NBA added the three point line) to 2018 (so we have several years of data for draftees). These years will give us an idea of drafting in the modern NBA era, and limiting ourselves to the first round can cut out a lot of noise from later draft picks, whose career win shares often have as much to do with whether they are given opportunities to play almost as much as their actual ability. Having one or two diamonds in the rough appear out of the second round can really screw up how well this metric thinks teams did at drafting in a way that seems unfair to me — second round picks seem to act more like dice rolls on prospects who are particularly unlikely to achieve their best-case scenario career.
NBA teams have gotten worse at drafting
Looking at Spearman’s ρ for drafts from 1979 to 2018, we can see that NBA teams clearly have not gotten better at drafting. The higher ρ, the better they did, and, with a few exceptions (2010, 2012, and 2018), teams did noticeably worse at drafting than they typically did in the 80s and 90s.
Looking at the 5 year moving average of ρ makes seeing what happened easier; there is a clear trend: teams got significantly worse at drafting through the late 90s and early 00s, then stabilized somewhat in the late 2000s, though some of the most recent drafts (especially 2017) have looked quite catastrophic for NBA GMs.
There’s one glaring explanation for this — drafting high schoolers straight into the NBA is very risky, and teams were willing to take the risk from 1995 until rule changes prevented teams from drafting players straight from high school after 2005. The timing of the decline certainly matches; the 5 year average peaks in 1998, three years after Kevin Garnett popped the seal on high schoolers going straight to the NBA, and bottoms out in 2004, shortly before the NBA required players to be at least 19 during the year the draft is held and a year removed from high school to be drafted.
Letting NBA teams draft high schoolers might be bad news for NBA GMs
If we compare the years where at least two players were drafted out of high school in the first round to the years, we can see that having just two high schoolers drafted in the first round was associated with a worse drafting performance by NBA teams:
- 1979-2005, years where <2 players were drafted from high school in the first round: ρ = .50
- 1979-2005, years where ≥2 players were drafted from high school in the first round: ρ = .35
- 2006-2018, years where teams cannot draft players straight from high school: ρ = .35
Pre-2006, there is a notable gap between how well teams did at drafting in the first round depending on how many high schoolers they drafted. If I were an NBA GM, I might be a little nervous about the rumors that the NBA will let teams draft players out of high school again — high school draftees make my job harder and increase the chances that I’ll do something that might seem reasonable at the time but ends up looking really stupid in retrospect. Completely whiffing on a draft pick gets me fired or at least puts me on the hot seat, whether I mess up by picking a high schooler who ends up being bust or by not picking a high schooler who ends up being a superstar.
But then again, NBA GMs aren’t doing so great at drafting nowadays, anyway…
NBA teams still haven’t recovered from the 90s
While more high schoolers in the draft seemed to make the draft harder, NBA teams’ ability to draft hasn’t recovered in the one-and-done era. If anything, NBA teams have been slightly worse at drafting since 2006 (ρ = .35) than they were even from 1995-2005 (ρ = .41).
When NBA teams could draft high schoolers, they usually didn’t; from 1995 to 2005, 29 players coming straight from high school were drafted in the first round. That’s only about one in ten first round picks. Drafting players straight from high school may have only been part of the reason teams have gotten worse at drafting. What else could be going on?
- The draft getting harder has more to do with drafted players being younger in general rather than than drafted players coming straight from high school in particular. The one-and-done era has seen more players enter the NBA before the age of 22, and seeing just one year of a player in college (or the G-League, or the Euro league, etc.) might not help that much. While younger draftees have tended to do better than older ones, predicting which younger player will do best is hard.
- Drafting high schoolers is only part of the reason that the draft got harder in the late 90s; the growth of the sport, especially internationally, increased the potential talent base enough that the risk of missing a great player got higher and the talent gap between a top-5 prospect and a late 1st round prospect got smaller, leading to more instances where players outperform those who were drafted before them.
- Regardless of whether eliminating teams’ ability to pick high schoolers made drafting easier, other factors made drafting harder through the 2010s, especially the need for GMs to compare players who were facing all kinds of different competition in different leagues. How is a GM supposed to compare a great role player on a Euro league team to a star at a Big 12 school? GMs have to make more apples-to-oranges comparisons than they had to make before.
I suspect that there is a little bit of truth to all of these factors and that adding in the additional challenge of high school draftees might make disastrous drafts like the one in 2017 look much more normal.
What about that 2017 draft?
Holy crap! A negative correlation between how early a player was drafted and how well they’ve done relative to their draftmates!? I couldn’t let this go unnoted.
A few years later, the 2017 draft looks like an unparalleled disaster for NBA teams’ draft boards. It was such an outlier that I had to cut it out of the graph I used above (although barely, I admit). It sticks out like a sore thumb when compared to all the other drafts:
I won’t linger on this draft because it deserves its own deep dive, but having a draft where teams could have randomly picked among the first 30 players and done better than they actually did (a miserable ρ = -.01 that is technically worse than random selection) demanded some comment. The world really was falling apart in 2017.