Thread by @Afinetheorem

On the NBA and economic theory: dynamic mechanism design tells us how to redo the draft. A good rule of thumb (due to Reny?) is that stochastic mechanisms like draft lotteries usually aren't best. How to prevent tanking while still having worst team get top pick? Simple! 1/x

Theory is beautiful. Myerson Revelation: look only at mechanisms that get teams to reveal quality "type" (we/ dynamic caveats!). Better teams will get some information rents else they tank. Let's find mechanism with no tanking that maximizes prob worst team gets top pick. 2/x

Let V be value of winning it all, W<<V value of getting top pick, p(n,w,l) prob of winning it all after n games. True quality is random walk over season from baseline unknown to mech designer. Costs 0 to give up in a game, c to try. If both try win prob based on true quality. 3/x

Immediate that variance in p grows over season, so incentive to tank grows also. Also then immediate that early wins and losses more informative about true quality since in game 1 worst team still has "chance" to win it all. 4/x

Optimal mechanism: get points for each loss. In game 1, a loss is worth 1 Draft Point. As season continues, loss is worth less, becoming negative late. Draft position based on Draft Points. Everyone wants to win late in season. Early season used to handicap truly bad teams. 5/x

Would make for some very exciting late season games (in NBA and other sports). Your dreadful team will need to go full out the last week or they will lose the top pick, not gain it! A Dallas Mavericks 2025 strategy would have been brutal for them. 6/x

Cool thing is that w/ a bit of econ theory you can actually *prove* the optimality of this rule. Mechanism design beautiful b/c we have theoretical "principles" that make "what is the best way to do X among any way including those I didn't even think of" a tractable problem! 8/8

@BillSimmons @ZachLowe_NBA @mcuban A mechanism perhaps of interest re the Draft ^

(btw, there are similar ideas to award 'draft points' only after a team is eliminated. PWHL does this. Not bad - gives incentive to try hard, weights teams who were bad early by giving them 'more games' to accrue points - but mechanism above is better on incentives and fairness.)

PS - If you want to improve your econ intuition, "fairness of worst team gets best pick" depends on avg. performance, "try hard in every game" depends on marginal incentives, with transferable utility you can solve both problems more cleanly than w/ randomization.