Issue 10
Opening Take
The Commodity Futures Trading Commission (CFTC) is looking at how prediction markets pull traders in. Front Office Sports reported last week that the agency is reviewing incentive programs for both traders and market makers over concerns that some promotions are misleading. An August advisory had already flagged offers like risk-free trades, unlimited rebates, and guaranteed payouts.
We've followed the product side of this closely here. Issue 5 covered a $3M Kalshi hedge against Lane Kiffin's bonus. Issue 8 covered MLB licensing its data to build team performance indexes, and last issue CME Group began trading NHL team futures. Issue 8 also noted Kalshi's $6 billion weekend, with parlay-style contracts making up 64% of one Saturday's volume. Now the regulator is asking what's pulling that volume in.
Sportsbooks have been here before. Ohio fined DraftKings, Caesars, and BetMGM $150,000 each in 2023 over bonuses described as free or risk free. And the New York Times found a DraftKings model that scored customers by how much they were expected to lose after a bonus. The platforms are new, but the acquisition math looks familiar.
The team performance indexes we've followed here only work as a benchmark if people trust the data and the trading around it. The CFTC's own proposal would restrict contracts tied to things like a player's injury because they're easy to manipulate. Meanwhile the NHL, MLB, and MLS have embraced prediction markets, the NBA is expected to follow, and the NFL is still the holdout. If promotions are driving much of the volume, are we looking at a price or a customer count?
What I'm Watching
- Paramount and Warner Bros. Discovery are now one company. Paramount closed its $110 billion acquisition of Warner Bros. Discovery on Oct. 6, creating Skydance Corp. CBS Sports and TNT Sports now sit under one roof, with rights to every major league except the NBA, which puts the new company alongside ESPN for breadth. David Berson, formerly CEO of CBS Sports, will lead Skydance Sports. The harder part is the balance sheet: roughly $80 billion in debt, a credit rating cut to BB (below investment grade) by Fitch, and widespread layoffs expected. RedBird Capital added another $4 billion at close, bringing its total investment to about $6 billion. Disclosure: RedBird IMI, a joint venture partner of RedBird Capital, is Front Office Sports' primary investor, which Front Office Sports notes in its own coverage. Read more
- LIV Golf is swapping sovereign money for a lender's. Saudi Arabia's Public Investment Fund (PIF) stopped funding LIV after the 2026 season, and LIV filed for Chapter 11 bankruptcy. BC Partners Credit is now providing $300 million in financing if LIV emerges, with an initial $4 million loan signed Monday. CEO Scott O'Neil called BC a "white knight" at Sportico Invest London and described a move away from a Saudi-style business plan toward a conventional one. BC's Ted Goldthorpe says he sees valuations of $100 million or more per team, though LIV has never sold a team stake. The next bankruptcy hearing is Oct. 8 in Trenton, New Jersey. Can a lender's math replace patient capital? Read more
- EuroLeague said no to the NBA. EuroLeague's 13 shareholder clubs met in Cernobbio, Italy on Oct. 6, and a supermajority rejected the NBA's offer to collaborate. Sources say the offer valued EuroLeague far below the $3 billion it's seeking, and it would have required EuroLeague to pause its expansion from 20 teams to as many as 24 by 2027-28. EuroLeague says it has no intention of pausing. Formal negotiations are expected, while the NBA says it will launch a European league next year regardless, with team bids reportedly in the $500 million to $1 billion range. How much leverage does an established league have against a brand with more global pull? Read more
Insight of the Week
45/35/20. That's the weighted blend of EvanMiya (45%), KenPom (35%), and Torvik (20%), three of college basketball's best-known rating systems, that predicted game spreads better than any of the three on their own. The analysis covers more than 17,000 games across the last three seasons. Its author, Evan Miyakawa, builds the EvanMiya model and discloses it.
EvanMiya and KenPom were virtually tied on game spread accuracy, with Torvik third. The number I'd focus on is how far off all of them still are. A typical spread miss is about 11 points, compared with about 15.6 if you assumed every game was a toss-up, and the models explain roughly 41% of the variation in game margins. Most of what happens in a game is still unexplained.
College basketball tips off in under a month. My favorite time of the year, and these are the three rating systems I follow. The takeaway for the analytics crowd isn't which model wins. It's that combining independent models beats picking one. Which of the three do you check first?
Looking Ahead
If you're attending any of these, reply and let me know. I'd love to connect.
Odell