An NBA insider from a middle class franchise (the Hawks) sounds off about NBA draft practices. The system is a mockery of fairness. He’s so sick of how the top pick in the draft is manipulated by the league to allocate the best players for a handful of privileged franchises. The year is 1963.
“This is the worst rule we have in effect and we should do everything possible to abolish it.” Kerner referred to the rule allowing a team to select a player on the first round draft who has attended a college within a 50-mile radius of the team’s city with no other club having a crack at him.
UPI wire story, April 23, 1963
Territorial picks
I’m still somewhat in shock about the Mavericks getting the number one pick in the 2025 lottery after the sham trade of Luka to the Lakers. Of course I knew the NBA was capable of it; this whole Stern series has examples of what they’re capable of. But it was still a breathtaking moment when they went and actually did it.
You’re probably familiar with the post-1985 roll call of suspicious picks. (We’ll get back to them.) But you might not know about the roots of the system and the institutional memory of draft manipulation buried in the very origins of the league.
The NBA Territorial Pick system was introduced in the earliest days of 1950, “the year in which the BAA was renamed the NBA.” Instead of all college players being subject to the normal draft, teams were allowed to call dibs on one player from any college inside a 50-mile radius of their home arena. That player would become their first round pick and other teams could not select him no matter how high in the draft they were. The stated idea was that since college basketball was initially so much more popular than pro basketball, it would be great if they could carry over the fanbase for a star who had been playing already for four years in a particular market.
It had a huge impact. The picks were used to secure blue chip no doubters. “Of the 23 territorial picks, 12 players have been inducted into the Basketball Hall of Fame.” Oscar Robertson, Wilt Chamberlain, Bill Bradley and Tom Heinsohn were all territorial picks. As you might expect the Philadelphia Warriors especially feasted on the strong Philly college basketball scene. (And they somehow got Wilt through a pre-emptive high school territorial pick.)
It was eliminated in the spring of 1963 with final phaseout for the 1966 draft. (That angry quote above was a clear case of trial-ballooning a done decision.)
NBA sources said the decision to eliminate a team’s right to claim players at colleges within a 50-mile radius was clearly aimed at the five-time champion Boston Celtics, and their great star Bob Cousy. Cousy retired after the Celts took their fifth straight NBA title this year and will coach Boston College next season. There were fears in the league that Cousy might be able to recruit and develop talent at BC that would go to the already powerful Celtics in the territorial draft.
AP wire story, May 1, 1963
And that was that. Lights out for the territorials. The specter of a Cousy-as-Pied-Piper leading a vicious circle of Boston domination was too much. Shaken by the idea of blue chip franchises hogging championships, the NBA put away childish things and the draft has been a model of fair play ever since.
Of course on here Earth 1 we entered the Lottery Era in 1984 (coincident with the David Stern Era) and a new potential for manipulation was built into an intentionally opaque system. Let’s take a look at the most suspicious results, gently sorting them into two categories.
Territorial “right team” for the consensus #1 pick
- 1985 : Patrick Ewing to the Knicks – (14.3% chance) – I’ve actually never been fully convinced on this one. (I’m the UFO nut who knows Roswell is strictly for casuals.) But it’s been hashed plenty out there without me going over it more.
- 1992/93 : Shaq (and then Penny) to the Magic – (15.1%, 1.5% chances) – It certainly seems like establishing Orlando was an NBA priority. Shaq recently claimed that David Stern asked him if he wanted to go “where it was cold or it was hot.” (He picked hot.)
- 2003 : LeBron James to the Cavs – (22.5% chance) – 100% territorial. In Northeast Ohio, nothing is given. Except maybe this pick.
- 2008 : Derrick Rose to the Bulls – (1.7% chance) – Another blatant territorial set-aside. Rose was a Chicago household name after back to back city and state high school championships.
- 2023 : Victor Wembanyana to the Spurs – (14% chance) – “Blessed” team that already has a track record for international development and has developed a French superstar (Tony Parker.) They weren’t ever going to send Wemby to Charlotte. “Overseas Chill Guy” is the Spurs territory.
Compensation for some loss
Not always about the consensus pick himself, (although all these happened when there was a clear #1), more that some mysterious balancing of the wheel of fate occurs… for the privileged team.
- 2011 : Kyrie to Cleveland after they lose LeBron – (2.8% chance)
- 2012 : Anthony Davis to the Pelicans (then Hornets) – (13.7% chance) – The New Orleans NBA franchise is owned by the NBA after George Shinn’s disastrous run ended during the Great Recession. In December 2011 their only star Chris Paul is traded to the LA Clippers for not a lot. In April 2012 a franchise buyer comes through. 30 days later the balls bounce their way and they get another superstar.
- 2014 : Cleveland – (1.7% chance) – The Cavs get another #1 just as LeBron is thinking of returning.
- 2019 : Pelicans – (6% chance) – Zion to the Pelicans after they lose Anthony Davis. You funnel talent to LA and you get a #1.
- 2025 : Cooper Flagg (likely) to the Mavs – (1.8% chance) – You funnel talent to LA and you get a #1.
Compensatory Picks
The idea of compensatory picks in exchange for player loss is not unique to the NBA. The idea has been in play in various forms in both MLB and the NFL for years. MLB in particular has had it go through several revisions starting with the advent of free agency in the 1970s. The general spirit is that if you lose a particularly good free agent you get a high draft pick as compensation from the team that signed him.
The details of the rules aren’t very interesting but it’s worth nothing that the two major forms of manipulation are recognizable mutations of existing rules: one that was in the NBA for decades and one that other major leagues have employed.
It’s not a coincidence that all the examples listed above and the lottery area itself start with Stern’s official reign and the dawn of the cap era post 1983. As we detailed in the last piece about the cap, the ABA era of the 1970s represent an actual capitalism experiment that was so distressing to management. They’ve done everything to cover it up again.
Theory
I’m sure you can summon more than a few dark examples of an institution officially ending a practice only to have it linger on for years, sub rosa and de facto. Mormon church polygamy. Legal covenant-based exclusion becoming casual real estate redlining norms. Hazing rituals.
I went looking for some lighter management theory reading on institutional memory and the ghosts of long-ago practices emerging again and again over time. The research intern (LLM) recommended a famous paper entitled “Clio and the Economics of QWERTY” by Paul A. David from the AER in 1985. (pdf)
The piece is interesting but it’s mostly about lock-in and path dependence for processes. The analogy to territorial picks is not perfect because QWERTY is not a hidden ancient influence, it’s right there at our literal fingertips. I did find this section very evocative however. The author is describing how even a market where keyboard buyers don’t have preferences will converge on a de facto standard if there’s even a small initial preference in the pool of potential typist-trainees (potential users) or the users themselves:
From the viewpoint of the formal theory of stochastic processes, what we are looking at now is equivalent to a generalized “Polya urn scheme.” In a simple scheme of that kind, an urn containing balls of various colors is sampled with replacement, and every drawing of a ball of a specified color results in a second ball of the same color being returned to the urn; the probabilities that balls of specified colors will be added are therefore increasing (linear) functions of the proportions in which the respective colors are represented within the urn. A recent theorem due to W. Brian Arthur et al. (1983; 1985) allows us to say that when a generalized form of such a process (characterized by unbounded increasing returns) is ex-tended indefinitely, the proportional share of one of the colors will, with probability one, converge to unity.
In other words a seemingly mild intervention (pull a random ball, and then put that ball back with another of its color) results in a field completely dominated by one winner. Maybe it’s the image of pulling balls from a drawing that make the parallels to draft lottery interference jump out.
To my amateur eye this looks like another appearance of the famed “Matthew Effect”, named for the recurring gospel verse summed up as “Unto those who have much, much will be given. But for those who have little, all will be taken away.” It comes up all the time in biology, finance and social networks: any place a power law exerts itself. (So basically any place.) Seemingly simple “natural” processes with slight bias produce emergent distributions with runaway winners. For our purposes it seems like any touch, any interference is going to by nature end up “touching” those who have “much” and creating a vortex of activity around privileged teams. You can’t interfere for the sake of neutrality. That just doesn’t work.
What should be done
Why do we need a draft at all? Before in the old feudal days of the reserve clause and no free agency the NBA wanted to keep the talent pipeline handouts as even as possible since there was very little ability to redress imbalances post hoc. But now, we have the salary cap, which in theory is a built-in balancer when combined with the free agent market. Assuming the cap works, why not let teams just bid for amateur talent? Cap space is cap space, if you have 30 separate self-interested teams with an equal purchasing power the talent pool will spread itself around pretty fast. Think how amazing it would be if a bad team could clear cap space and instantly try to address its deficiencies by signing two top-10 players in the incoming rookie class.
The reason of course is that the main cap is just one part of salary control, the other part is the strict rookie scale and reserve clause that rookies are still subject to. Teams can pay under market value for young players and reap the rewards of their service before they have to pay them. (This lets bad owners skate by not paying their fair share: They can field a bad team, draft high, get their fans excited about a young talent they don’t have to pay, let them leave, field a bad team, repeat.)
The union is also partly to blame here. A fairly justified critique of unions is their emphasis on seniority in favor of merit and willingness to subject new employees to worse conditions. The NBAPA is no different. Rookie contracts before free agency are typically four years long with only team options in those years. So in effect the full salary cap is meaningless for young players. They’re in a separate bubble with a much smaller cap tied to the rookie scale and a 1950s-vintage reserve clause lockup.
Once a veteran player breaks through and proves valuable enough to re-sign he can get a rich negotiated contract that take advantage of the full cap. It’s tricky to figure out how much the rookie scale affects the median player. Obviously some players’ careers end before four years. But many of those players would not command high salaries anyway. The real earnings shift right now is from young above-average players who are paid less than market to veteran minor stars who often “soak up” cap with larded on extra years paid above their market value.
I am still trying to find a full dataset for the median career length of NBA players in the last decade. I spot-checked three rosters from the 2021-22 season (to allow for four years of run-out for then rookies) and I was surprised to find that the median full career length for all listed players was 7 years. (Data from basketball-reference.com) That is quite a bit higher than what random SEO-trap webpages and LLMs estimate but since they don’t have sources I have no real reason to believe the lower number is true. If the median career length is really above 4 then there’s even more reason for the NBAPA to support the death of the draft.
A world without the draft would require veterans to take a little hit but would be amazing for fans and the median player: no tanking, more rewards for scouting and talent eval and one more lever of NBA manipulation removed. To steal a turn of phrase: we should do everything possible to abolish it.