Professional sports leagues face a structural problem that most industries don't: they need their weakest members to get stronger. In a normal market, weak competitors go bankrupt and exit. In a sports league, a team that's perpetually terrible drives down attendance, television ratings, and the competitive credibility of the entire enterprise. The league needs some mechanism to channel talent toward struggling teams, and the draft — in which teams select from a pool of incoming players in an order determined by the previous season's standings — is the primary tool for doing so.
The basic draft concept is straightforward: the worst team picks first, the best team picks last. This gives struggling franchises first access to the most promising new talent, which should, over time, create competitive balance. The problem is that it also creates a perverse incentive. If finishing last guarantees the first pick, and the first pick is a potential franchise-altering player, then a team with no realistic chance of winning the championship has a rational incentive to lose as many games as possible. This strategy — known colloquially as "tanking" — has plagued every major league at various points. Teams trade away their best players mid-season, rest starters, and make decisions that are transparently designed to maximize losses, all in pursuit of a draft position that rewards them for failing.
The NBA introduced its draft lottery in 1985 specifically to combat tanking. Instead of giving the first pick to the worst team outright, the league created a system in which the bottom teams are entered into a lottery with weighted odds — the worst team has the best chance of winning the first pick, but it's not guaranteed. The weights have been adjusted several times since then, most recently in 2019, when the NBA flattened the odds so that the three worst teams each have a 14% chance at the first pick, down from a system where the worst team had a 25% chance. The flattening was a direct response to teams like the Philadelphia 76ers, who embarked on a multi-year tanking strategy so blatant that the process became known as "The Process" and its architect was eventually fired — though the strategy, by most accounts, worked.
The lottery introduces randomness into the draft order, and the randomness serves a specific function: it weakens the link between losing and reward. Under a straight worst-to-first system, tanking is a dominant strategy because the relationship between losing and draft position is deterministic. Under a lottery, the relationship is probabilistic. The worst team might get the first pick, or it might get the fifth pick, and the difference between those outcomes is significant enough that tanking carries real risk. A team that deliberately loses forty games to improve its odds might end up with the same pick it would have gotten by trying to win. The randomness doesn't eliminate the incentive to tank — a 14% chance at the first pick is still better than a 1% chance — but it reduces the payoff enough to make the strategy less appealing.
The NFL uses a different approach, assigning draft order strictly by record with no lottery component. The league has been less plagued by tanking, partly because NFL rosters are so large (53 players) that a single draft pick, even the first overall, is less transformative than in the NBA (where a roster is 15 players and one star can change everything). The NHL uses a weighted lottery similar to the NBA's. Major League Baseball recently introduced a lottery for the top picks as well, after years of debate about whether tanking was distorting competitive integrity.
The philosophical tension at the heart of draft lotteries is between fairness and incentive design. A purely fair system — where the worst team always gets the best pick — creates terrible incentives. A purely random system — where every non-playoff team has an equal shot at the first pick — is fair in process but unfair in outcome, because a team that barely missed the playoffs could leapfrog a team that genuinely struggled. The weighted lottery is a compromise: it tilts the odds toward the teams that need the most help while introducing enough uncertainty to make gaming the system unreliable.
Whether the compromise works depends on what you're optimizing for. If the goal is competitive balance, the evidence is mixed — the NBA's lottery hasn't prevented dynasties, and some lottery winners have squandered their picks while some lower-seeded teams have thrived. If the goal is reducing the most egregious forms of tanking, the lottery has been more successful, particularly since the 2019 odds reform. If the goal is creating a process that fans and teams perceive as legitimate, the lottery has mostly succeeded, though occasional conspiracy theories about rigged drawings (the 1985 "frozen envelope" rumor has never fully died) reveal that trust in the randomization is not automatic.
What draft lotteries demonstrate at scale is the same principle that applies to any random selection process: the value of randomness lies not in producing optimal outcomes but in producing legitimate ones. A random process can't guarantee the best result, but it can guarantee that no one gamed the system to get there, and in competitive environments where trust is essential, that guarantee is worth the tradeoff.