Robinhood Q2 2026 Earnings Breakdown: Prediction Markets Grew 10x, Crypto Halved, and the Stock Still Fell

TL;DR

• Record revenue of $1.308 billion, up 32% year over year, just ahead of the $1.306 billion estimate. Diluted EPS of $0.62 crushed the $0.45 estimate and rose 48% year over year.
• Transaction revenue jumped 44% to $776 million, and the composition is the story: prediction markets up more than 10x to $156 million, equities up 95% to $129 million, options up 29% to $342 million.
• Crypto revenue fell 38% to $100 million. The line that made Robinhood a growth stock in 2024 is now smaller than the prediction markets line that barely existed a year ago.
• Despite all of it, shares closed down 3.15% at $89.84 and fell a further 4% in extended trading.
• The reason is what management did not say. There is no explicit revenue guidance, and the 2026 adjusted operating expense plus share-based comp range was narrowed to $2.675 to $2.775 billion. Tightening the cost range while declining to guide revenue reads as caution about the second half.

What Robinhood Actually Reported

| Line | Q2 2026 | Change | |---|---|---| | Total net revenue | $1.308B | +32% y/y, record | | Diluted EPS | $0.62 | +48% y/y, vs $0.45 est | | Transaction revenue | $776M | +44% | | Options | $342M | +29% | | Prediction markets | $156M | up more than 10x | | Equities | $129M | +95% | | Crypto | $100M | -38% | | FY26 adj. opex + SBC | $2.675-2.775B | narrowed |

Read the transaction lines top to bottom and you are watching a business change species in real time.

The Board

Every line grew except the one Robinhood used to be famous for.

The Mix Shift Nobody Was Modelling

A year ago the bear case on Robinhood was simple: it is a leveraged bet on crypto volumes wearing a brokerage costume. When crypto cools, revenue halves, and the multiple goes with it.

Crypto just cooled. Revenue fell 38% to $100 million. And total revenue set a record anyway.

That is a genuinely important result, and it is the strongest argument the bulls have ever had. Three lines absorbed the shock:

Prediction markets: up more than 10x to $156 million. From a rounding error to bigger than crypto in four quarters. This is now a top-two transaction revenue line.

Equities: up 95% to $129 million. Doubling equity transaction revenue in a year is not a market-conditions story, that is share capture.

Options: up 29% to $342 million. Still the largest single line, still compounding. Options remain the highest-margin flow Robinhood touches, which is why we keep writing about what calls and puts actually are: a meaningful share of Robinhood's profit comes from retail traders using instruments they have not been taught.

The diversification argument is no longer theoretical. It has been stress-tested by a 38% decline in the original growth engine and it held.

So Why Did the Stock Fall?

Because a beat with no guidance is not a beat, it is a question.

Robinhood did not provide explicit revenue guidance. What it did provide was a narrowed full-year range for adjusted operating expenses plus share-based compensation, at $2.675 to $2.775 billion.

Narrowing a cost range is normally a confidence signal: management knows where the year lands. But narrowing the cost range while saying nothing about the revenue line, in a quarter that set a revenue record, points one direction. If the second half were tracking above plan, the easiest thing in the world would be to say so.

There is also the crypto problem, which is not solved by being diversified away from it. Crypto revenue at $100 million and falling 38% means the trajectory is still down, and a business does not get credit for growing despite a declining segment forever. Eventually it has to stop declining.

And the tape did not help. HOOD reported into a session where the Fed held and the market fell 2%, with the Nasdaq closing more than 10% off its high. High-beta retail-facing financials are not what anyone wants to hold overnight in that environment.

Scoring Our Own Preview

Our July 25 preview named the three lines that would decide the print: crypto revenue, net interest income and account growth. Two of those three were the right lens. Crypto was the swing factor, exactly as flagged, and it came in weak, exactly as feared.

What the preview underweighted was prediction markets, which went up more than 10x and is the single most important number in this release. That is a fair miss to own: we were watching the segment that could break the quarter and not closely enough at the segment that could remake the business.

The preview also suggested a covered call for holders wanting to harvest elevated implied volatility into the report. That one worked. The stock fell, the premium was kept, and the vol crush after the print is exactly what a call seller gets paid for.

The Bull Case and the Bear Case

Bull case. This is now a diversified retail trading platform, not a crypto proxy, and it proved it in the worst possible quarter for that proof. Revenue records with crypto down 38%, EPS up 48%, equities doubling, prediction markets scaling from nothing to $156 million. At $89.84 you are buying a company whose revenue base got structurally more durable this quarter, priced as though it got riskier.

Bear case. No revenue guidance in a record quarter is a tell, and the narrowed cost range reinforces it. Prediction markets growing 10x is thrilling and entirely unproven as a durable revenue stream: it is a new product in a favourable political and sporting calendar, with regulatory risk nobody has priced and comparisons that get brutal next year. Crypto is still shrinking. And Robinhood remains a high-beta bet on retail engagement, which is the first thing to evaporate in a genuine drawdown, and the Nasdaq just entered one.

Our read: a buy, but a volatile one, sized as a satellite position rather than a core holding. The mix shift is real and the market is underrating it. What the market is correctly rating is that this business still lives and dies on retail risk appetite, and that appetite is currently being tested by a 30-year Treasury yield above 5.19%. Do not own this expecting it to be defensive. Own it because you think retail trading volumes compound and prediction markets are a real product, and be honest that both of those are bets.

The Options Angle

• At roughly $89.84, one contract is about $8,984 of notional, which keeps every standard structure available to a mid-sized account.
• Post-earnings implied volatility collapses, so this is the wrong moment to buy premium and a reasonable one to have already sold it.
• Cash-secured puts in the low $80s are the cleanest expression of "I want this cheaper," and the wheel suits a stock this volatile with no dividend to protect.
• The next real catalyst is the conference call on Thursday July 30 at 5:00pm ET, which is unusually late relative to the release. Management commentary on prediction markets and on why there is no revenue guidance is the thing to listen for, and it can move the stock more than the print did.

The One-Line Read

Robinhood set a revenue record while its crypto business fell 38%, which is the diversification proof the bulls have wanted for two years, and the stock fell anyway because management narrowed its cost range and refused to guide revenue: own it for the mix shift, size it for the beta, and listen to Thursday's call for the guidance the release withheld.