The NFL's Letter Didn't Move a Single Contract. A Phone Call From the CFTC Moved Three Categories in Three Weeks.
The league has now written to Kalshi and Polymarket twice and been ignored twice. The federal regulator has never published a rule and has cleared more markets off the exchanges than every state attorney general combined.
September 6, 2026 at 6:38 PM EDT
6 min read
Editor's note: TrueEdge builds odds tools and earns affiliate commissions from licensed sportsbooks. This piece argues a position on prediction-market regulation, and our commercial interest runs toward the sportsbooks. Weigh it accordingly.
The NFL wrote to Kalshi and Polymarket in March asking them to stop listing four kinds of contract. It wrote again this week. In between, the exchanges removed three categories of contract — and the league had nothing to do with any of them.
Sabrina Perel, the NFL's chief compliance officer, put the frustration on the record: "It is deeply concerning that bets within the objectionable categories that we identified months ago have been and continue to be listed as contracts on exchanges." The AP moved it Thursday, September 4. One account dates the letter itself September 3; the league has not published it, so we are not asserting a day.
What actually cleared those markets was the Commodity Futures Trading Commission, working by request rather than by rule, and it is worth walking the calendar because the calendar is the story.
Three weeks, three retreats, no rulemaking
August 14. Kalshi pulled every sports "mention" market — contracts on whether a broadcaster says a particular word — after the CFTC opened a review into whether they are manipulable. Mention markets on Trump, on earnings calls and on news broadcasts stayed up. Only the sports ones went. Neither the exchange nor the agency confirmed the mechanics; the reporting rests on two people with direct knowledge, and one of them described the contracts to NPR as "potentially very easy to manipulate."
August 26. Polymarket's US entity self-certified a product line called NFL American Football Player Participation Contracts, with Patrick Mahomes' return from an ACL tear as the worked example, and told the CFTC it intended to list no earlier than August 27. Kalshi was already running the same shape of market. This was a deliberate test: the CFTC's June proposal says injury contracts are contrary to the public interest, and self-certification is the procedure that lets a designated contract market list first and argue later.
August 31. Polymarket withdrew the certification. Its spokesperson said the company "withdrew the filing to be in full compliance with the CFTC." Kalshi kept its version live — Mahomes trading around 94% to play in Week 1 — and kept markets on when injured players would return.
September 1. Kalshi pulled them too, at the CFTC's request. Gone: injury-duration contracts and Week 1 participation contracts. The venue had run markets on the availability of Luka Dončić, Anthony Edwards and Malik Nabers this year.
September 3 or 4. The NFL sent its second letter, listing categories that the CFTC had already emptied two of.
Four days of federal jawboning did what five months of league correspondence did not. That is not a compliment to the CFTC. It is the problem.
What the league is actually asking for
The four categories, as the NFL frames them: contracts a single person can move (a kicker missing, an incomplete pass, a running back's yards on his first carry, a receiver's targets, what an announcer says, which celebrity is in the building); contracts the league calls inherently objectionable (injuries, availability, fan safety, player misconduct); contracts on officiating (flags thrown, replay outcomes, crew assignments); and contracts on things already knowable to someone (the first play called, roster moves, trades, draft picks, coaching decisions).
The March version came with a cleaner statement of the theory from Jeff Miller, an NFL executive vice president: "We're trying to stay as far as we can from some of those sorts of inside information wagers that could exist in this space." That is a real concern and not a novel one. A contract on whether a specific player suits up is priced by information that roughly forty people possess and nobody is legally barred from trading on.
The league's leverage problem is that it has none. It renewed with DraftKings and FanDuel and added Fanatics on August 27, and it declined to license any exchange. Kalshi is the official prediction-market app of nine MLB clubs. Polymarket has league-level deals with MLB, the NHL, MLS, Serie A and LaLiga. The NFL is the last big holdout, which means it is the one league that cannot get a contract pulled by picking up the phone to a commercial partner. So it writes letters, and the letters get filed.
The part the exchanges should be worried about
Kalshi's argument to the Supreme Court, in substance, is that a nationwide financial exchange cannot be run by fifty separate gambling regulators and that Congress gave the CFTC exclusive jurisdiction so it would not have to be. New Jersey filed its certiorari petition on September 2, asking the opposite question after the Third Circuit sided with Kalshi in April and the Ninth Circuit ruled against it 3–0 on August 28.
Set the merits aside for a paragraph and look at what the federal alternative looks like in practice. The CFTC proposed a rule on June 10 — amendments to Regulation 40.11 plus a new Appendix F, with a 90-day review process and defined public-interest factors — that would treat injury and officiating contracts as contrary to the public interest. Chairman Michael Selig's line in the announcement was that "the CFTC will protect the integrity of our regulated markets without standing in the way of responsible innovation." Comments closed July 27. The rule is not final. It will not be final before kickoff on Wednesday. And in the gap, three product categories came off two exchanges on the basis of requests that were never published, under a standard that was never adopted, with no comment period, no reasoned explanation and no route to appeal.
That is a worse deal than the one Kalshi is describing to the Court, not a better one. A regulator that can clear a product line with a phone call in August can decline to clear anything in October, and the only thing that changed is who is sitting in the chair. Exclusive federal jurisdiction is a real argument. Exclusive federal discretion is not the same argument, and it is what the exchanges are currently living under.
The strongest case against us
Here is the other side, put as well as we can put it. Injury markets are genuinely bad. They price a medical record, they create an audience with money on whether a hamstring holds, and they aim harassment at team doctors and trainers who never signed up to be a market-moving source. Officiating markets are worse — the people whose decisions settle the contract are unpaid participants in a market they cannot trade. Every day those contracts stayed listed was a day of real exposure. A rulemaking that finishes in 2027 is not a remedy for a season starting September 9. If informal pressure emptied the shelf in three weeks, the process complaint is a luxury.
That is right on the merits, and it is why the outcome here is better than the alternative in which nothing moved. But process objections are not luxuries when the same discretion runs both directions. Notice which category survived every one of these interventions: contracts knowable in advance. Kalshi has been listing Week 1 starting quarterbacks, next-team markets for individual players, coach firings, and when rookie passers will debut. Those are the purest insider-information products on the board, they sit squarely inside the NFL's fourth category, and nobody has asked for them to come down. There is no published reason why injury markets crossed a line and roster-news markets did not, because there is no published standard at all — which is exactly what you would expect from regulation by request.
What would change our mind
If the CFTC finalises the June rule before the Super Bowl with injury and officiating contracts prohibited on the face of it, the last three weeks stop looking like discretion and start looking like a regulator enforcing a policy slightly ahead of its paperwork, which is ordinary and defensible. And if either exchange discloses what it was actually told — the request, the date, who made it — the transparency objection largely dissolves.
Neither has happened. What we have instead is a league writing letters nobody answers, an agency moving markets nobody can read, and a season that opens Wednesday.