Cathie Wood Goes On a Selling Spree: 3 Stocks She Just Sold

Ark Invest trimmed positions in AMD, Robinhood, and Roku as part of “profit-taking” despite strong one-year gains (AMD +295%, HOOD +22%, ROKU +58%). The article flags valuation risk in AMD (74x this-year earnings, 41x next-year) even as the latest-quarter results showed revenue up 38% to $10.3B and adjusted earnings up 45% y/y. Robinhood’s growth is decelerating (revenue +15% in the latest quarter, weakest since late-2022), while Roku’s sale reflects deal progress—Fox’s announced offer values Roku at $160/share initially, implying about a 5.6% return from $148.09 close over the next 6–12 months, with Wood likely reallocating capital.

Analysis

ARK’s selling looks more like a liquidity/position-sizing signal than a thesis shift, but that still matters because AMD and HOOD are both crowded momentum names. When a widely followed allocator starts distributing stock after a big run, the near-term effect is usually multiple compression before fundamentals show up; that is especially true for AMD where valuation already implies near-perfect AI execution. Relative-value still favors NVDA over AMD until AMD can prove sustained sequential AI revenue acceleration and better operating leverage.

HOOD is the most fragile of the three because its revenue mix is most sensitive to retail risk appetite, options activity, and crypto volatility. That makes the next 1-2 quarters more dependent on market tape than user growth; if realized volatility stays muted, top-line momentum can keep disappointing even with active accounts intact. The flip side is also important: a sharp equity pullback or crypto spike would quickly reflate the name, so this is a trading vehicle, not a stable compounder.

ROKU is effectively a spread trade now, not a fundamentals trade, and the edge is limited unless the deal ratio moves. FOXA is the cleaner expression if one wants exposure to the transaction closing, while ROKU holders are mostly collecting a modest carry unless Fox stock strengthens. The contrarian read is that the market may be over-interpreting routine portfolio rotation as bearish news; the real signal is modestly negative for high-beta growth factor flows, not a wholesale fundamental break.

Key Decisions for Investors

• Long NVDA / short AMD for 1-3 months as a relative-value AI pair; entry on any AMD strength after flow-driven selling. Thesis breaks if AMD prints clear sequential AI revenue acceleration or guides materially above current consensus.
• Short HOOD on rallies or buy 1-2 month puts if implied vol is reasonable; the trade works if transaction revenue keeps lagging because equity vol and crypto remain subdued. Cover if HOOD reaccelerates revenue above mid-teens growth or if market volatility spikes sharply.
• Do not chase ROKU outright; if seeking merger-arb exposure, use FOXA or wait for the ROKU/FOXA spread to widen to a more attractive level before entering. Falsifier is a materially tighter deal spread or Fox stock rally that lifts implied consideration.
• Trim or hedge high-beta growth baskets that are crowded with passive/quant ownership over the next few weeks; this is a factor-flow warning, not a stock-specific disaster. If the Nasdaq keeps grinding higher and these names stabilize, treat the signal as noise rather than a structural short.