How the NBA Works

The strangest fact about the NBA

The NBA needs four groups of people.

Remove any one of the four and there is no league. Now look at what each group holds in writing.

The league has spent eighty years writing enforceable protection for every group the business needs — except the two that cannot negotiate back. Owners are protected from other owners. Players are protected from owners. Small markets are protected from large ones. Broadcasters are protected by contract. Fans and host cities are protected by the fine print of a revocable license and the hope of the next election. Every row of that table is sourced further down this page.

The players’ row proves the bottom two rows are a choice, not a law of nature. In January 1964, the game’s stars refused to take the floor for the first nationally televised All-Star Game until the league committed to a pension. The commissioner personally guaranteed it fifteen minutes before airtime. The real pension came three years later, under threat of a playoff strike. One refusal and one threat, three years apart, turned the league’s most replaceable-seeming group into contractual partners who now collect roughly half of a $12 billion revenue stream. By right, not by grace. Fans and cities have never run that play — the Fan Compact tells the 1964 story in full.

The flywheel — and who fuels it

Here is the whole machine in one loop. Follow it clockwise from the top, because the top is where every dollar enters.

The top half is the story the league tells about itself, and it is true. The game went global. The league made its players some of the most-followed people on earth, and their fame became a ~$76 billion media book. The bottom half never makes the broadcast. Thirty franchises — deliberately fewer than the cities that could support them — convert rising valuations into leverage over the towns they play in. The leverage extracts public capital. And the capital lands in the one place the players’ hard-won revenue split cannot reach: the owner’s balance sheet.

Investors like to call NBA franchises beachfront property, and the phrase is truer than they mean. Beachfront owners famously pay for their own houses. In this league the town builds the house — and in the deal on Portland’s table, the rent is a dollar a year.

The numbers, 2020 → 2026

Six years of numbers tell one story: the league got much richer, and the growth concentrated in money that has nothing to do with any particular building.

BRI is not revenue. Basketball Related Income ($11.68B) is the defined pool the players share. League revenue ($14.3B projected) is larger, because whole categories of money are kept out of the pool. Which categories, and why a city should care, is Part Five.

The growth left the building

In 2024-25, central and national revenue — the money that follows a franchise no matter where it plays — was 38% of team revenue and climbing; under the new media deal it would have been 44%. The league is drifting, on purpose, toward the NFL’s centralized model. Meanwhile the one major stream that was genuinely local collapsed. The successor to Bally Sports shut down in April 2026, thirteen NBA teams became local-TV free agents with roughly $180M in payments at risk Verified, and the league’s reported fix is a national, geofenced streaming hub targeted for 2027-28.

Now hold that fact against the arena question. A team collects its $143 million national media check in any building, in any city. Expansion fees, the Nike contract, league sponsorships, betting-data rights — none of it depends on which arena a team occupies. What an arena actually touches is tickets, suites, concessions, parking, and local sponsorship: the shrinking share of NBA economics. Cities are being asked to pay record public money for the part of the machine the league itself is moving away from.

What an arena deal actually is

Public debate prices arena deals on one number — “City X paid $300 million, City Y paid $500 million” — and that framing is how cities lose. A subsidy headline is not a deal. A deal is four ledgers, and an honest comparison reads all four:

So the benchmark question is never “how much did the public pay.” It is: what did the public invest, what risk did it carry, what rights did it surrender, and what did it get back? Ask it that way and thirty arena deals stop blurring into headlines. They sort into three families. Tap any team on the map; the full thirty-team detail lives in the deals analysis.

The inverted deals: the public owns the building and pays the team

In several markets the money now runs backward. Indiana’s public authority charges the Pacers a dollar a year in rent (plus a $3.45M parking fee) and pays the team roughly $14.5M a year to operate the public’s own arena — and the team keeps the naming revenue. Memphis is renovating FedExForum with a planned $550M of entirely public money while the city and county pay the Grizzlies escalating operating subsidies; the long-term lease extension remained unsigned as the money began to flow. Detroit’s arena authority granted a 35-year rent-free concession on a building whose debt is serviced by captured property taxes, school millage included, through 2048. These are the deals quoted back as “market” whenever a new city asks what to expect.

The drafted deals: same league, same threats, written terms

Other cities faced the same league, the same threats, the same consultants — and wrote instruments. Milwaukee put non-relocation into state law, binding any future buyer of the Bucks, with liquidated damages reported still at $200M in year 30, real escalating rent, and owners who contractually absorbed — and actually paid — $24.1M of overruns. Sacramento’s executed agreement sets damages beginning at $580M, bans even relocation negotiations for the first 28 years, and hands the team’s name, colors, and trophies to the city on default. Oklahoma City’s new lease runs to 2053, rent paid into a capital fund, with a roughly $1B early-exit penalty. Houston’s 2003 template put $8.5M a year of real rent against the bonds and gave the city 5% of naming rights (capped at $200K a year). Miami’s county, owning its building, captured most of a $117M naming deal. Dallas voters capped the public share at the ballot box in 1998, and the cap has held for a quarter century. None of these cities defeated the auction. Every one of them proved the auction’s terms are draftable.

The private proof: where cities decline, owners pay

And where the public declined to pay for the building, the owners simply paid. Chase Center: privately built, pays property tax. Intuit Dome: privately built. Crypto.com Arena: privately built, except a small public piece the deal required to be repaid from arena revenues. United Center: privately built, $10.8M in property taxes in 2024. Toronto self-funded a C$350M renovation while extracting C$800M in naming value. Philadelphia’s next arena is announced as a ~$1.5B private joint venture. This family is the quiet answer to “this is how it works everywhere”: where cities decline to bid, owners fund arenas the way every other industry funds its factories. And the franchises appreciate anyway.

The rulebook, translated into money

June 6, 1946. The Commodore Hotel, New York. A room of hockey-arena operators meets to invent a basketball league, mostly to fill their buildings’ empty dates. Before a single game is played, the first agenda produces a salary cap — $40,000 per roster — and territorial rights: fifty miles, exclusive, from day one. The franchise fee is $1,000. Verified

Cost control for owners and monopoly for arenas came before the basketball did. Everything since is elaboration. The rules below read like sports policy. They are economic architecture. The chips under each one grade who it protects — and the chips are the finding.

Twelve rules, one direction. None of them is irrational; each is expertly built to solve a real coordination problem — for someone. The cap solves owner-versus-owner bidding. The tax solves rich-owner dominance. Revenue sharing solves small markets; pooled media solves fragmentation. The subsidy auction and the relocation threat are coordination problems too — the cities’ and the fans’. They remain unsolved because the league profits from them unsolved.

Cities are not customers. They are investors.

Everything above sharpens into one distinction the arena debate almost never makes. Fans are the league’s customers — underprotected customers, but customers. Cities are something else. When a government puts $573 million of construction capital into a building a private operator monetizes, it is not buying a product. It is providing capital. And capital, everywhere else in the economy, comes with rights.

Private capital would never accept this deal, and the proof is that it never has. A private investor contributing $573M toward a ~$1.3B enterprise takes equity, board seats, information rights, covenants — ask the private-equity firms that now hold stakes across the league’s franchises. Player labor, the other non-owner input, negotiated its rights: a defined share, audit power, an arbitration system. Public capital alone arrives naked. No share, no audit right, no governance, no enforceable stay — and in the NBA’s case it is barred from equity by written Association policy (Article 5(j)), while the labor contract routes subsidy dollars around the players’ split and into owner equity (Article VII). The system is not neglecting to give public capital rights. It is designed to receive the capital without them.

Portland is the live illustration: roughly $573M of public capital (true commitment $848M) into a franchise that sold, months before the vote, for $4.25 billion — a franchise whose value has grown ~2.2× since 2020, appreciation that accrues entirely to the owner and is realized, at sale, outside the player split and outside any public claim. What the public’s money builds him →

The conclusion is stronger than any complaint about subsidies: public capital should have what every other input in this league already has — defined, written, enforceable rights. Not charity. Not gratitude. Terms.

What to do with all of this

The obvious next question is what leverage a city actually holds, and the honest answer is sobering. The research behind these pages mapped every legal lever any American city has ever pulled against a league. Exactly one has ever produced cash from an NBA relocation: a well-drafted lease, enforced by specific performance — and lease leverage expires with the lease. Everything a city will ever enforce must be written down before the public money moves, because the league’s constitution guarantees the city nothing, forever, on purpose. The cities that understood this — Sacramento, Milwaukee, Oklahoma City, Houston — wrote instruments. The cities that didn’t are the ones now paying their tenants.

But even the best-drafted deal leaves the auction standing, because the auction is a coordination problem, and coordination problems are not solved one negotiation at a time. The league itself is the proof. Every structure in Part Five — the cap, the tax, revenue sharing, pooled media — is the owners solving their coordination problems collectively. The players solved theirs with a refusal in 1964 and a strike threat in 1967. The two groups still negotiating alone are the two at the bottom of the heatmap.

That is the entire case for the Fan & Community Compact: five league-wide rules that do for cities and fans exactly what the salary cap does for owners — put a floor under a ruinous auction. Not a boycott. Not a lawsuit. A compact, in a league already governed by compacts: owners have a constitution, players have a collective bargaining agreement, networks have eleven-year contracts. The Compact is the missing document, the one for the people who pay for everything.

The short visual version of this argument is at /who-has-power. The Portland case that surfaced it — six months inside 8,000 pages of one city’s records while the votes were still open — is at /everyone-but-you. Every number on this page is indexed, badged, and correctable at the evidence library.

Read & sign the Fan Compact →