1 Stock to Buy, 1 Stock to Sell This Week: Marvell, Kohl’s

U.S. stocks closed at record highs, with the S&P 500 up about 1% for the week, the Dow up 2.1%, and the Nasdaq up 0.8%, as investors welcomed progress in U.S.-Iran talks and a strong earnings season. Next week’s key catalysts are the core PCE inflation print, Q1 GDP, and several major earnings reports, including Marvell on Wednesday and Kohl’s on Thursday. Marvell is expected to post EPS of $0.79 on $2.4 billion of revenue, while Kohl’s is projected to report a $0.21 per-share loss on $2.99 billion of revenue and weaker guidance, with both stocks facing sizable post-earnings volatility.

Analysis

The near-term market setup is being driven less by macro data than by positioning around two binary catalysts: a geopolitics de-escalation headline that supports risk appetite and a compressed earnings calendar that can still move single names violently. That combination favors quality growth with secular demand visibility, while exposing lower-quality consumer cyclicals to harsher-than-normal punishment because there is little macro noise to offset company-specific disappointments. In that sense, the market’s breadth-positive tone is fragile: indices can grind higher even as dispersion inside retail and hardware widens.

Within semis, the bigger second-order winner is not just Marvell but the broader AI networking supply chain, especially adjacent vendors that benefit if hyperscalers keep expanding connectivity budgets. The key risk is that supply constraints can initially look like a positive because they validate demand, but they can also cap near-term upside if bookings convert into revenue slower than the market expects. If guidance remains strong but not explosively better, the stock likely becomes a vol event more than a fundamental re-rating event after an extended run.

Kohl’s is the cleaner short because the issue is not one quarter, but a structurally weak format in a discretionary environment where traffic sensitivity is high and margin leverage is negative. The most important second-order effect is competitive: any weakness there supports share gains for off-price and value-oriented peers, while also reinforcing that consumer trade-down is still the dominant retail behavior. If management sounds more cautious than expected, the downside could extend beyond the earnings gap because investors will likely reprice the business on a lower terminal sales base rather than a temporary miss.

The contrarian angle is that consensus may be overconfident in the durability of the AI earnings premium and underconfident in how quickly retail misses can cascade into estimate resets. For Marvell, a good print may not be enough unless the guide implies incremental acceleration into the next two quarters; for Kohl’s, the market may already be positioned for weakness, but a modestly worse guide can still trigger a sharp gap because implied volatility is elevated and the equity is low-quality balance-sheet optionality. In short: long the best AI infrastructure exposure, but keep it tactical; short the weakest consumer names only into event windows, not as a structural blanket short.

Key Decisions for Investors

• Long MRVL into the print only if it holds above the prior high into the close before earnings; target a 2:1 reward/risk using a 2-3 week horizon, with upside continuing only if guide implies sequential acceleration beyond the current AI networking spend cycle.
• Buy a call spread in AVGO vs. long-only semi beta for 1-2 months: if Marvell confirms AI networking demand, Broadcom is the cleaner second-order beneficiary through custom silicon and networking adjacency; prefer defined-risk options over stock after a big run in the group.
• Short KSS into earnings or via put spread expiring 2-4 weeks out; the setup favors a gap-down event because the market is pricing weakness but not full terminal-value compression if guidance resets lower again.
• Pair trade: long COST / short KSS for the next 1-2 months to express consumer divergence; the spread should work if discretionary weakness persists and traffic migrates toward higher-quality value/necessity formats.
• Avoid chasing DELL, HPQ, and CRM ahead of prints unless you have a strong view on guide revision; these names look like lower-conviction event risk with less asymmetric payoff than MRVL or KSS.