Lila Soto: Picture a GM, Tuesday morning, spreadsheet open — their team is 28 and 42, eight games back of the tenth-worst record. And the spreadsheet is basically telling them: win more games, hurt yourself.
Iris Holm: That's not a hypothetical.
Lila Soto: Not at all. That's tanking — losing games on purpose to land a better draft lottery position. And the NBA just passed a reform, 29-1, that was supposed to make that spreadsheet look different. Adam Silver behind it, Board of Governors, the whole thing.
Iris Holm: The 3-2-1 lottery. The three worst teams now get reduced odds at the number one pick.
Lila Soto: Right — and I mean, that's a genuinely surprising move. You'd think the worst team gets the most help. Here they get less. But then economists actually modeled it out, and — mm, the expected value of a losing season still beats a forty-five-win bubble season. The math is just... identical.
Iris Holm: The league adjusted the odds. The payoff matrix didn't care.
Lila Soto: Which is kind of the whole thing we're trying to work out today — not just what tanking is, but why every serious attempt to fix it leaves the same spreadsheet open on Tuesday morning.
Iris Holm: And whether the league has quietly decided that's acceptable.
Lila Soto: But wait — the 1985 lottery was supposed to stop that spreadsheet from existing at all, right? Like, that was the whole pitch.
Iris Holm: Think of it as a raffle. The more you lose, the more tickets you hold. The NBA didn't get rid of the raffle in 1985 — it just made it fancier.
Lila Soto: And before that, the worst team just... automatically got the top pick. No randomness.
Iris Holm: Deterministic. Which created its own corruption problem — teams visibly throwing seasons. So the league added ping-pong balls. One thousand and one four-number combinations drawn from fourteen balls, weighted by record. Worse record, more combinations. The Orlando Magic won back-to-back number one picks in '92 and '93 that way — Shaquille O'Neal, then the trade for Penny Hardaway. That's the dream the lottery is selling. But here's what the league didn't account for — actually, I mean, maybe they did and just didn't care — losing went from being *certainly* rewarded to *probably* rewarded. That's not a disincentive. That's just a different spreadsheet.
Lila Soto: So they solved visible corruption by encoding the incentive deeper into the math.
Iris Holm: The Philadelphia 76ers proved it. 'The Process' — years of deliberate losing, openly, publicly. So visible the league had to reform again in 2019.
Lila Soto: Which means the 1985 mechanism failed by its own terms. The 76ers basically broke it in front of everyone.
Iris Holm: That's the original sin. The league fixed something visible and introduced something invisible. And invisible incentives are harder to kill.
Lila Soto: And that invisible part is where the EV math actually bites. Because even after 2019 — even with the three worst teams each sitting at 14% instead of 25% — the franchise cornerstone value is so asymmetric that the probability drop barely registers. Like, you're multiplying a smaller number by something enormous.
Iris Holm: Quantify 'enormous.'
Lila Soto: Shaq. Orlando Magic, 1992. A moribund franchise becomes a national brand, a championship contender — and then they get the number one pick *again* in 1993. Back to back. That compounding effect is what a GM is actually pricing in when they open that spreadsheet.
Iris Holm: So the expected value calculation isn't just one pick — it's the possibility of sequential picks.
Lila Soto: Right — and mm, that's the part that gets buried. Nine more wins buys you marginally better playoff seeding, maybe a little ticket revenue. But losing those nine games inches you toward an Orlando Magic outcome. And the Orlando Magic outcome doesn't just change the roster — it changes the *franchise* for a decade.
Iris Holm: Which means the 14% figure is almost beside the point. The payout on the right side of the equation is so large that even a low probability times that upside clears a high probability of a middling outcome.
Lila Soto: And I guess — yeah, what's underneath that is the franchise cornerstone narrative itself. The league built the mythology that one transformational pick justifies everything. LeBron's lottery moment, Shaq to Orlando. That story is so embedded that even the reform can't dislodge the math it created.
Iris Holm: The league is competing against its own legend.
Lila Soto: And the part that might actually make all of this worse — the game theory layer under the individual EV — it gets genuinely strange. We'll get to that.
Iris Holm: Strange doesn't cover it. The game theory layer means individual EV isn't even the problem anymore. If one team tanks, the expected value of *staying* competitive drops for every other team watching. So they tank too. Not because they ran the same spreadsheet — because they watched someone else run it first. That's a Nash equilibrium. No odds adjustment touches the payoff matrix.
Lila Soto: So the tanking spreads — like, it's contagious by logic.
Iris Holm: Amplifies. Beyond what any individual incentive predicts. The league fined the Utah Jazz $500,000 for tanking-adjacent lineup decisions. Indiana Pacers $100,000. Enforcement already failed. And now the 3-2-1 reform expands the lottery to *sixteen* teams.
Lila Soto: Wait — sixteen? I thought it was fourteen before.
Iris Holm: Fourteen before. Sixteen now. Which means — actually, this is the part that I keep turning over — you've created a sweet spot. Fourth or fifth worst. Tank just enough to get lottery odds, avoid the three-worst penalty zone. Strategic mediocrity. And that is *much* harder to prosecute than whatever got Utah fined.
Lila Soto: Because it's invisible. You're not resting starters, you're just... not quite trying hard enough to win close games. How do you fine that? And — mm, okay, here's something that I think makes this even harder to unwind. The Memphis Grizzlies hold a traded pick restriction on the Utah Jazz's 2027 pick, from the Jaren Jackson Jr. trade. So the Jazz's tanking calculus is now literally embedded in Memphis's asset valuation. No lottery reform touches that.
Iris Holm: The incentive is in the contract structure. Not the odds.
Lila Soto: Right — and that's kind of where the Rookie Matching Mechanism comes in as an idea. The Gale-Shapley algorithm — teams and players submit ranked preferences, get matched. Its architects say it's the most structurally coherent alternative. But even they admit it can't fix tanking *and* player-team mismatch simultaneously without creating new distortions.
Iris Holm: Which closes with what would have to be true for any fix to work: you'd need to change the payoff architecture — salary cap penalties tied to draft position, something with real teeth — not just the ping-pong ball count. The NBA keeps adjusting the raffle. The problem is the prize.
Lila Soto: The thing I keep sitting with — and I genuinely can't settle it — is whether the 29-1 vote means Adam Silver actually believes the next tweak works. Or whether competitive balance is just the story the league tells broadcast partners while accepting tanking as the cost of giving small markets one shot at transformation every decade.
Iris Holm: Those are different problems with different solutions. And I don't think the league has separated them.
Lila Soto: Yeah — and I mean, that's kind of the uncomfortable place to land, right? Because if the cure requires abandoning competitive balance entirely, or letting rookie free agency concentrate every top talent in four cities, then the disease might actually be... load-bearing. The tanking is doing something the system needs.
Iris Holm: The odds-flattening trade-off. Weaken the link between need and pick position, and you've also weakened the draft's whole reason for existing. That's not a side effect — that's the paradox.
Lila Soto: And no one's resolved it. Not the play-in inclusion idea, not the Rookie Matching Mechanism, not thirty-something years of lottery formats. Every proposal trades one distortion for another.
Iris Holm: Which is probably where we leave it. Genuinely open.
