The Dip Buyer

This month's ranking of 3-month returns among mega-caps puts ORCL (-24%), TSLA (-20%), WMT (-14%), GOOGL (-11%) and COST (-6%) at the very bottom, with META, CRM, AVGO, PG and AMZN filling out the rest of the bottom-10 'penalty box.' Following the buy-bottom-5/sell-outside-top-10 rule: XOM has climbed to roughly the 11th-worst spot (its 3-month return turned positive, +5%) so it exits the portfolio. CRM, while no longer in the bottom five, is still inside the bottom-10 buffer zone, so it stays but gets trimmed back to an equal 20% weight rather than being added to. TSLA, now the second-worst performer, is added as the fifth holding. ORCL, WMT and COST remain and are rebalanced up to the target 20% each. Why these are worth catching: Oracle's slide reflects investor anxiety over its debt-funded AI data-center buildout (S&P cut it to BBB-) even as it reports a record $638 billion cloud backlog and accelerating AI contract wins, a classic case of a fundamentally growing business being punished for financing optics. Tesla cratered after a Q2 profit collapse (operating margin fell to 1.4%) and negative free cash flow, but record deliveries and revenue growth of 26% suggest the sell-off is more about sentiment/valuation reset than a broken business. Walmart has slid over the past two months despite EPS beats and raised guidance, dragged down mostly by valuation concerns and heavy insider selling rather than deteriorating fundamentals. Salesforce is down sharply this year on 'AI-will-eat-software' fears even though it keeps beating on revenue/EPS and Agentforce ARR is growing triple digits, and most analysts still rate it a buy with large upside targets. Costco has pulled back modestly on valuation worries and a post-earnings profit-take, but same-store sales, traffic and membership growth remain strong and it continues to draw fresh buy reiterations from analysts. The contrarian bet is that as these near-term overreactions fade, quality balance sheets and durable demand should help each name mean-revert toward the group average.

Starting from an all-cash portfolio, I'm building a contrarian basket of the five mega-caps with the worst trailing 3-month returns, equally weighted at 20% each, betting that these high-quality franchises mean-revert after their recent beatings. Oracle (-1.8% over 3 months, actually the biggest 1-month loser at -39.7%... underlying data shows -1.8% 3m) has been hammered because investors fear its AI data-center customers can't pay for the massive backlog they've signed up for, even though the business itself is still growing fast. Oracle's stock plummeted by 25% during the first half of 2026 and it's down by more than 50% from last year's record high, yet the company reported strong fiscal Q4 2026 results with revenue up 21% year-over-year, beating estimates, and 84% of analysts covering Oracle still rate it a Buy with a consensus price target implying roughly 88% upside from current levels. This looks like a classic case of the market punishing near-term financing/leverage fears while the underlying cloud demand story stays intact. Exxon Mobil (-12.7% over 3 months) is not a broken business but a story of an oil-price round-trip: XOM stock declined roughly 13% over the past 30 days amid falling oil prices following eased geopolitical tensions in the Middle East, after an earlier war-premium rally. Fundamentals like its Permian and Guyana production assets haven't changed, and at least one analyst recently upgraded the stock to Buy as global oil inventories hit multi-decade lows, arguing today's low prices are unlikely to persist. Walmart (-11.9% over 3 months) has fallen for six straight sessions and is now down more than 20% from its 52-week high reached in mid-May, despite the retailer posting strong results; the pullback looks more sentiment-driven (partly a reaction to a random acquisition and rate/policy jitters) than a sign of a broken business, especially given how strong Walmart's multi-year run has been. Salesforce (-7.9% over 3 months) has been swept up in fears that AI agents will cannibalize traditional software subscriptions, but Salesforce was recently upgraded to Buy by Guggenheim on the view that artificial intelligence won't be a death knell for the company and that its valuation has become attractive after the drop. Costco (-6.8% over 3 months) remains one of retail's highest-quality compounders; its recent fiscal Q3 2026 results showed net income up 15.2% and net sales up 11.6%, but the stock still fell as investors used the print and a rich valuation as an excuse to take profits. The pullback looks like a valuation reset rather than a deterioration in the underlying business, and analysts remain largely constructive. All five are being funded entirely from cash (no existing positions to unwind), each capped at the 20% equal-weight target, staying within the 25% max-position and long-only/no-leverage constraints. As is inherent to this contrarian, momentum-agnostic strategy, these are 'falling knife' bets — near-term news flow for each name is genuinely negative, and there is no guarantee of a quick bounce; position sizing should reflect that risk.