Flows and event tape
A lot of today was plumbing, not macro.
Workday (WORKDAY)+18% on talk of a private-equity acquisition. The move wasn’t a careful bid model; it was positioning. Shorts covered, event-driven money stepped in, and momentum followed. Takeaway: there’s still a real bid for deal convexity even with the discount-rate argument unresolved.
Reddit (REDDIT) was index math. Shares rose 11% after hours after news it’ll be added to the S&P 500 on August 18. Same script: people front-run passive flows, underweights get forced to chase, and retail adds volume. The tape treated it as forced buying first, fundamentals later.
One quieter tell from the 13F stack: Seth Klarman’s Baupost disclosed a new position in Pershing Square and added to Amazon and Alphabet in Q2. Not a day-trade catalyst, but it’s a reminder that patient capital is still fine living in liquid compounders—and sometimes prefers doing it through a vehicle like Pershing Square instead of trying to thread needles in single names.
Earnings dispersion
Earnings season is paying for clarity and punishing anything that muddies the path to leverage—especially in smaller names where confidence is most of the equity.
- Marpai (MARPAI)down after GAAP EPS -$0.18 on $4.17M revenue. The quarter didn’t improve the “losses are narrowing” story. Stock got hit.
- Talon Metals Corp. (TALON)up on GAAP EPS $0.02 and $51.5M revenue (flagged as a revenue beat). Profitable plus beat still gets rewarded.
- Diginex Limited (DGINEX)down after GAAP EPS -$1.20 and $3.62M revenue (a revenue miss). A top-line miss is brutal when investors are already skeptical.
- Nu Holdings (NU)up on a Q2 beat with revenue above expectations, pointing to stronger results from increased customer engagement. That’s a better duration driver than promo-led growth.
- Gemini (GEMINI)flat/down on another loss despite higher revenue. Without a credible bridge from growth to shrinking losses, “scaling” just means scaling the problem.
Macro and regulation
Rates stayed heavy. The US Treasury sold 30-year bonds at the highest yield since 2001. Call it supply, term premium, inflation uncertainty—whatever the driver, the message is the same: long duration isn’t cheap, and refinancing risk stays in play for levered or story-heavy names.
Consumer stress is showing up in the financing channel. $211B in new car loans last quarter, with an average payment of $785/month and a typical term of nearly six years. That’s affordability by extension. It works until it doesn’t, and it rarely breaks on a calm day.
At the top end, demand still looks fine: US airfares +25% YoY, supported by strong demand “especially from affluent consumers.” The two-speed consumer remains the cleanest frame—credit tightens at the bottom while premium categories keep pushing price.
Crypto policy stayed half-formed. The SEC delayed a meeting that was expected to clarify new crypto asset regulations while legislation remains stuck. Separately, a judge ordered Kalshi to stop offering most prediction market contracts in Washington state. The definitional fights—what’s a security, what’s a contract, who regulates what—are still playing out in real time, sometimes state-by-state, and that’s friction for growth.
What mattered today
- Event-driven flows still run the tape when there’s deal optionality or index mechanics forcing hands (WORKDAY, REDDIT).
- Earnings dispersion is unforgiving: investors are paying for a credible path to leverage, not “revenue up.”
- Long-end yields keep duration expensive, while consumer credit stretches via longer terms.
- Regulatory uncertainty in crypto and prediction markets keeps adding headline risk.
Today wasn’t about a new macro truth—it was about who got cornered by flows, and who showed the market a believable bridge to cash flow.