Robinhood's Q2 2026 print was, by the numbers, about as clean a beat as a company can turn in. Diluted EPS of $0.62 blew past the $0.4494 estimate — a 38% surprise. Revenue hit a record $1.308 billion, up 32% year-over-year. Net income jumped 48% to $573 million. Gold subscribers hit a record 4.8 million. Thirteen separate business lines are now each doing $100 million-plus in annualized revenue.
And the stock still fell — about 3% after hours on the print, then further the next morning. Beats don't guarantee a rally anymore when the market is this picky about how the growth is showing up.
The Headline Numbers
| Metric | Q2 2026 | Q2 2025 | YoY |
|---|---|---|---|
| Total net revenue | $1,308M | $989M | +32% |
| Net income | $573M | $386M | +48% |
| Diluted EPS | $0.62 | $0.42 | +48% |
| Adjusted EBITDA | $741M | — | +35% |
| Gold subscribers | 4.8M | 3.4M | +39% |
| Total platform assets | $369B | $279B | +32% |
Every one of those is a record. Revenue of $1.308B landed almost exactly on the $1.306B consensus (a rounding error, essentially), while EPS blew through estimates by 38% — the kind of gap that usually gets a print called an unambiguous beat. It wasn't treated as one.
Why the Stock Fell on a Genuine Beat
Two things did the damage, and neither is about the headline print.
Crypto revenue fell 38% year-over-year, to $100 million from $160 million. That's the second straight quarter this segment has gone from marquee growth driver to drag, and it's exactly the kind of volatility investors have learned to price defensively after the 2024–25 crypto cycle. Our take: this isn't really new information — anyone who watched Robinhood's crypto numbers swing through the last two years should have expected this line to be lumpy — but "expected" doesn't mean "ignored," especially in a market already nervous about growth durability.
Management tightened FY2026 opex/SBC guidance to $2.675–2.775B (from $2.7–2.825B) — but gave no forward revenue guidance at all. That's a genuinely disciplined move on the cost side, and management is right to frame it as captured efficiency funding two new business lines (Rothera, WonderFi) rather than a spending freeze. But paired with silence on the top line, in a market looking for reasons to sell growth stories, "we're being careful with costs" reads to some as "we're not sure revenue keeps accelerating." That's a stretch of an inference, but it's the one the tape made.
The result: shares slipped roughly 3% in the after-hours session on the print itself — a genuinely modest reaction given the crypto miss, actually, which tells you the market mostly bought the "record quarter" framing. The bigger move came the next morning.
The Fed Made It Worse
Robinhood's earnings landed the night before Fed decision day, and the following morning's session was ugly for reasons that had nothing to do with Robinhood. The Fed under new chairman Kevin Warsh held rates steady at 3.5%–3.75% for a fifth straight meeting, but three regional presidents dissented in favor of a hike — the most one-directional dissents since 2016 — with the Middle East conflict clouding the inflation picture. Long-end Treasury yields jumped despite the hold, semiconductor names sold off hard on AI-spending jitters, and the Dow shed over 1,100 points intraday. Robinhood shares fell a further 4% that morning to around $89.46, with no fresh company-specific news driving it — pure sympathy selling alongside SoFi (down 10% that same morning) and Affirm (down 4%).
Stack the two moves and Robinhood lost something like 7% across two sessions on a quarter that beat EPS by 38%. That gap between the fundamentals and the price action is the real story here: one leg was Robinhood-specific (crypto softness, no revenue guide); the other leg was the entire high-beta fintech complex getting sold on a hawkish-leaning, uncertain Fed day. Reading the whole move as a verdict on Robinhood's quarter would be a mistake.
What's Actually Working
Strip out the crypto line and the macro noise, and the underlying business is diversifying in exactly the direction management has been promising:
Prediction markets are now a real line, not a novelty. Event-contract revenue hit $156 million, up more than 10x year-over-year, on 13.6 billion contracts traded (also up more than 10x). This was a rounding error a year ago.
Thirteen business lines now clear $100M+ in annualized revenue, including Robinhood Legend (~18 months post-launch) and the Credit Card business — real evidence the "everything app" strategy is landing, not just a slide-deck aspiration.
Trump Accounts launched July 4 and already show 7M+ sign-ups and ~$1.5B deposited — a brand-new distribution channel disclosed for the first time this quarter, worth watching for how fast it converts into Gold subscriptions and net deposits.
Agentic Trading — Robinhood's AI-agent trading product — hit ~100K accounts and $100M+ in AUC just months after a May 2026 launch, and the company is already pushing it further: a same-week X post touted "P&L, meet your agent," a new feature making every trade's outcome visible to a user's AI agent.
One number worth a gut-check: $129 million of net income — $0.14 of the $0.62 EPS — came from a one-time gain on deconsolidating Robinhood Ventures Fund I. Back that out and adjusted-ish EPS is closer to $0.48. Still a real beat, still up meaningfully year-over-year, just less dramatic than the headline "+48%."
What Robinhood Told Its Followers
Robinhood's own X activity around earnings barely engaged with the numbers. The main earnings-day post was a bare-bones "Robinhood Markets has just released financial results for the second quarter of 2026," pointing followers to the call. The tweet that actually drove engagement that week wasn't about earnings at all — it was the Agentic Trading feature launch: "P&L, meet your agent... Live for everyone in the U.S. today." Same playbook as SoFi's own post-earnings messaging: lead with product, let the numbers speak for themselves in the filing.
What to Watch
- Whether the crypto line stabilizes or keeps shrinking. A third straight quarter of crypto declines would be harder to write off as noise.
- H2 revenue. Consensus implies about $2.77B in H2 2026 versus $2.375B in H1 — a real acceleration the company hasn't explicitly guided to. Watch whether management addresses that gap on future calls.
- Expense growth in Operations (+97%) and G&A (+51%). Both are growing much faster than the opex/SBC outlook implies is sustainable — worth tracking against the "disciplined spend" narrative.
- Trump Accounts and Agentic Trading conversion. Both are brand-new distribution channels; the next couple of quarters will show whether they're durable or a launch-quarter spike.
The Bottom Line
Robinhood beat EPS estimates by 38%, hit records across nearly every metric that matters, and still lost roughly 7% across two trading sessions. About half of that is a legitimate reaction to real questions — crypto's second straight down quarter, and management staying quiet on forward revenue while tightening cost guidance. The other half is a market that was already selling high-beta growth names on a messy, hawkish-leaning Fed day that had nothing to do with Robinhood specifically. Separate those two threads and the operating business looks considerably healthier than the two-day chart suggests.
Robinhood Markets (NASDAQ: HOOD) reported Q2 2026 GAAP diluted EPS of $0.62 on revenue of $1.308B, both up sharply YoY and both records. Shares fell roughly 3% after hours on the print and a further ~4% the next morning amid a broad fintech selloff tied to Fed policy uncertainty.