3 Unstoppable Stocks I Bought Last Month

Nvidia: revenue surged 73% in the latest quarter with net margin at 56%; shares trade around 21x this fiscal year and <16x next year, the lowest 2027 earnings multiple among the Mag 7 despite China trade restrictions. Reddit: daily active users 121.4M (+19% YoY) and revenue +70% last quarter, with free cash flow tripling and forward P/E ~33x (next year ~24x) even as the stock is down roughly 50% from recent peaks. Zillow Group: revenue grew ~15–16% each of the last two years and is trading ~18x this year and ~14x next year, but the company remains exposed to a weak housing market (U.S. existing home sales ~4M, a 30-year low) which could materially impede the turnaround.

Analysis

Nvidia sits in a rare position where secular demand (AI training + inference) is accelerating but the profit pool is unusually concentrated; that concentration creates both pricing power today and a crystallizing target for incumbents and foundry partners to claw back margin over 12–24 months. Expect gross/net margin compression to be the primary re-rating channel rather than demand destruction — a 300–800bp margin slide would materially reset earnings even if revenue growth stays elevated, so monitor share gains in networking and chiplet roadmaps from rivals as a leading indicator.

Reddit’s dataset monetization is a high-leverage, low-capex revenue stream that scales with AI players’ need for labeled, community-curated text; this can drive asymmetric upside in the next 6–12 months but is simultaneously a one-way funnel to larger AI platforms that could synthesize Reddit content instead of directing traffic. Treat current FCF improvement as durable only if product hooks (ads, subscriptions) maintain engagement against an increasingly summarized search behavior.

Zillow is a classic macro-exposed growth asset: operational metrics (lead yield per listing, conversion rates) imply material operating leverage once transaction volumes rebound, but timing is driven by rates and inventory — think 12–36 months for a clear inflection. The optionality here is binary: a relatively small market-share improvement in a rising transaction cycle can re-rate multiples quickly, while a prolonged affordability shock keeps cash flows under pressure.

Key Decisions for Investors

• Pair trade (6–12 months): Long NVDA equity (size 2–4% notional) vs short INTC equal notional. Rationale: express concentrated GPU moat vs legacy CPU capex miss; hedge macro/tech beta. Risk: competitor tech or foundry advantages compressing NVDA margins — stop-loss if NVDA/INTC spread tightens by 20%. Target: 30–80% asymmetric upside on NVDA leg vs limited downside through short INTC funding.
• Event-driven trade (3–9 months): Buy RDDT shares or 9–12 month LEAPS (buy OTM call if liquidity allows) size 1–2% with a 10–15% OTM put as hedge. Rationale: monetize AI dataset demand and ad yield improvements; unwind if AI platforms materially reduce referral traffic. Target: 30–50% upside; max drawdown cushioned by put hedge.
• Macro-tactical (12–36 months): Long Z (or ZG) via buy-and-collar: purchase shares size 1–3% and sell near-ATM 12–18 month calls while buying 6–12 month 10–15% OTM puts. Rationale: capture upside from housing cycle normalization while funding protection. Target R/R ~2:1 if transaction volumes recover; exit or increase hedges if mortgage rates remain >6% beyond 12 months.
• Risk-off hedge (rolling): Maintain a 0.5–1% notional long position in data-center related alternatives (e.g., NDAQ-listed instruments tracking cloud infra or modest short of non-AI-capable incumbents) to protect portfolio tech exposure should AI capex stall for 6+ months. Rationale: preserves liquidity and limits beta in a rapid derating scenario.