DraftKings and FanDuel tighten grip on New York as handle slides

In August, DraftKings put $614 million in handle through its New York operation and still managed a 9.2% hold. FanDuel ran $599 million through its books and held 10.7%. Those two numbers, sitting next to each other in the New York State Gaming Commission's monthly report, tell the story the headline misses: the market contracted, and the two firms at the top of it got stronger.

New York's total online sports betting handle fell to $1.73 billion in August, down from $2.04 billion in the same month a year earlier. That is a real decline. But strip out the single Fanatics bettor who ran hundreds of millions through the platform in August 2025 and inflated the comparative baseline, and the year-over-year picture tightens considerably. Fanatics' own handle this August was $213.2 million — and its gross gaming revenue still rose nearly $3 million from a year ago. The headline number on the decline is softer than it looks.

The more important signal is in the hold. Operators collectively held 9.7% of handle for the month, roughly a full percentage point above August 2025. In a market where volume is contracting, the surviving dollars are being won at a higher rate. That is not a market in distress — that is a market consolidating around operators with pricing power.

BetMGM and Caesars held above 10% and 9% respectively on combined handle under $235 million. BetRivers posted a 10.5% hold on just over $10 million. Those are thin books running efficient operations, not operators gaining ground. The gap between the top two and everyone else is widening in both volume and revenue.

The commission's report flagged that some portion of the decline may reflect prediction market growth pulling dollars away from licensed sportsbooks — but it could not quantify that. I have seen similar attribution problems in other state markets where new platforms enter: the effect is real but diffuse, and regulators are always six months behind measuring it. The honest answer is that August's calendar was weak — no marquee NFL regular season, no World Cup, college football's Week Zero drawing a fraction of September volume — and the baseline was inflated. Those two factors explain most of the gap before prediction markets get any credit.

What the August data cannot tell us is whether DraftKings' simultaneous launch of a prediction market platform changes the competitive math in New York or simply adds a second revenue line to the same customer relationship. DraftKings led licensed sportsbook handle in New York for a second straight month while running a prediction market product in parallel. If that product is pulling from the same wallet, the hold rate compression will show up in the September report. If it is pulling from a different wallet entirely, DraftKings has found a way to grow total revenue in a contracting licensed market.

FanDuel's hold was down from 11.9% a year ago to 10.7% this August. That is the one number in the report I keep coming back to. Market leaders with pricing power do not usually see hold compression when volume falls — they see it when competition intensifies at the margin. Something is pressing on FanDuel's take rate, and the report does not say what.