High Roller's Prediction-Markets Bet Is Priced to Work Before the Launch Proves It

In five days, a tiny online casino operator is about to bet its entire future on a product that has never generated a dollar of revenue. On September 15, High Roller Technologies (NYSE: ROLR) starts the invitation-only beta of ROLR, its U.S. prediction-markets platform, with a full public launch targeted by the end of October. The same week, management pitches the story to institutions at the H.C. Wainwright Global Investment Conference in New York. The timing is not accidental: the company needs investors to believe before the numbers arrive.

Whether they should comes down to a single arithmetic fact. At about $6.20 a share and roughly 11 million shares outstanding, High Roller trades for a market value near $65–70 million even after sliding about 18% from its post-earnings high. Against that sits a casino business being deliberately run into the ground — revenue was $2.8 million in the second quarter, down 52% from a year earlier, on pace for barely $11 million annually after $30 million in 2024 — and a prediction-markets operation that so far books no revenue at all. Nearly the entire valuation is a bet on a platform that is still being tested.

A shrinking casino funds a new act

High Roller earned its way as a small Las Vegas operator of real-money online casino brands such as High Roller, Fruta, and Kassuuu. That business is now being wound down, not because it is failing management but because management chose a different future: exiting "certain online casino markets," as it put it, and redeploying into prediction markets.

The second quarter shows the cost of that choice. Net revenue fell to $2.8 million from $5.8 million a year earlier, and the net loss from continuing operations widened to $2.4 million from $1.2 million. Adjusted EBITDA, which was near break-even a year ago, swung to negative $1.8 million. General and administrative expense actually rose 19% to $3.4 million as the company spent to prepare for prediction-market entry. The transition is funded by cash and restricted cash of about $18 million at the end of June, drawing on a roughly $26 million raise completed in January — enough runway for a business this small, though the burn continues.

The new act is built on a partnership. In January High Roller signed a binding deal with Crypto.com, and in April it closed the definitive agreement: a 24-month exclusive arrangement in which Crypto.com | Derivatives North America acts as the futures commission merchant providing licensing, custody, and liquidity, while High Roller owns the customer-facing platform and holds a perpetual worldwide license to the backend technology. ROLR US has already been approved as a member of the National Futures Association and registered as a Guaranteed Introducing Broker. The company has also lined up marketing partners, including an exclusive deal with Forever Network, which reported more than 20 billion impressions in 2025.

The opportunity is real; the share-price math is early

Prediction markets are genuinely booming, which is why this story attracted attention in the first place. Trading volume roughly quadrupled to about $64 billion in 2025, and third-party forecasts put the U.S. market on a path toward $1 trillion to $1.5 trillion in annual contract volume by decade's end.

That is a large and fast-growing pool. But a large pool is not High Roller's pool, and volume is not automatically revenue. The company has not disclosed the economics of its split with Crypto.com, how much fee it takes on each trade, or any customer-acquisition target. The public proof so far is a free-to-play "prediction challenge" meant to test engagement — a marketing exercise, not a revenue line.

The competition is also better capitalized than High Roller. Robinhood already offers event contracts through its Derivatives unit; Kalshi and Polymarket are entrenched; and the American Gaming Association has used the moment to argue that sports event contracts are, in substance, sports bets — a regulatory cloud the company will have to navigate as it courts the same customers.

This is the mismatch worth naming. The stock trades as if the prediction-markets ramp has already happened, while the evidence — no revenue, no unit economics disclosed, a casino base that is shrinking rather than cushioning the transition — says the outcome is still uncertain. When a multiple runs ahead of proof rather than behind it, the buyer is paying for the potential, not for demonstrated results.

The proof window is the next two quarters

What keeps this from being a pure caution lecture is that the thesis is falsifiable on a short clock. The public launch is targeted by the end of October, which means the fourth quarter should produce the first real prediction-markets revenue, and the first half of next year will show whether the volume compounds. Management has said the beta will be judged on registrations, deposits, trading patterns, and settlement — the exact metrics that would demonstrate paid demand rather than narrative.

The honest read is that High Roller is a high-risk option on a large market, and the market has already priced in meaningful success before the proof exists. A buyer here is paying a near-whole valuation for a launch that could still stumble on execution, regulatory friction, or simple user apathy in a crowded field. A watcher gets the better end of the trade: the next two quarters will tell whether the ramp converts to real, disclosed economics — and the multiple will reprice accordingly, in whichever direction the evidence points. It is too early to call the prediction-markets bet a winner, and the valuation is too rich to ignore the risk.