DraftKings NFL handle climbs 15% as prediction unit triples volume

Jason Robins stood at a Wells Fargo conference podium and said something that deserved more attention than it got: DraftKings' prediction market volume has grown roughly two and a half times since July. He said this in the same breath as a 15% year-on-year sportsbook handle increase for the first two weeks of the NFL season, and the room apparently treated both numbers as part of the same good-news story. They are not the same story.

The sportsbook figure is a mature business performing well in a strong season. The prediction market figure is a different animal — compressed into a few months, accelerated by World Cup activity over the summer, and now apparently sustaining into football season. Robins described it as a "huge growth story," which is the kind of phrase executives use when they are still surprised by what they are seeing in their own data.

What interests me is the market share claim. Robins said DraftKings holds close to a double-digit share of the broader sports prediction market, with a higher share specifically in NFL. For a product that launched in December 2025, that is a faster capture rate than most people in this industry would have forecast nine months ago. The company offering roughly three times as many NFL markets as competitors is part of how that happened — depth of coverage creates liquidity, and liquidity attracts the traders who then deepen it further.

The consensus read on DraftKings' prediction market entry was that it would be a slow burn, that Kalshi and Polymarket had enough of a head start in brand recognition and infrastructure to hold their positions, and that a traditional sportsbook operator would struggle to attract the kind of user who thinks of themselves as a trader rather than a bettor. Robins' numbers suggest that read was at least partially wrong.

I have a view on where this lands. The consensus still underweights how much distribution matters in thin markets. DraftKings already had the app on millions of phones. Converting existing sports bettors into prediction market participants is a lower-friction sale than building a prediction market user base from scratch. Kalshi built its reputation among people who specifically sought it out. DraftKings is reaching people who didn't know they wanted this product until it appeared next to the parlay they were already building. Those are structurally different customer acquisition paths, and the second one scales faster once the product reaches a minimum quality threshold.

The spending question that some investors raised at the conference is real but probably premature. Robins indicated the company could increase marketing spend through the end of 2026 to support 2027 growth. The adjusted EBITDA target of around a billion dollars for 2026 remains intact. The prediction market business is not yet large enough to move that number materially, which means it is still in the phase where growth is cheap to fund and the cost of building market depth comes before the revenue that depth generates.

Missouri's cease-and-desist letters, which now cover DraftKings alongside five other platforms, are the variable nobody at a Wells Fargo conference wanted to discuss in detail. A state enforcement action does not resolve quickly, and the legal question of whether sports event contracts fall under the Commodity Exchange Act or state wagering law is not settled. DraftKings has more regulatory surface area than a pure-play prediction market operator — more states, more licences, more political exposure. That creates a different kind of risk calculus than the one Robins was sketching on Tuesday.