Baird reiterates Oracle stock rating at Outperform, $215 target

Oracle reported Q4 fiscal 2026 revenue of $19.2 billion, up 21% year over year and slightly above the $19.1 billion consensus, while non-GAAP EPS of $2.11 beat the $1.96 estimate. Oracle Cloud Infrastructure revenue reached $5.8 billion, up 93% year over year and ahead of Street expectations, though applications revenue was slightly below estimates and capex guidance came in above forecasts. Baird kept an Outperform rating and $215 target, but said the setup may be less favorable after the stock's nearly 15% weekly decline.

Analysis

The setup is less about the quarter itself and more about the market having already priced Oracle as an AI infrastructure winner with near-perfect execution. When a stock has run on multiple expansion before the next catalyst, even a clean print can become a de-risking event if the company shifts the debate from demand scarcity to funding intensity. The key second-order issue is that heavy capex plus financing needs can compress future free cash flow conversion just as investors are trying to underwrite a longer-duration AI backlog.

The beneficiary set is broader than ORCL. Any hyperscaler or GPU-adjacent supplier that can grow with less balance-sheet strain may now screen better versus a vendor leaning harder on external capital to support growth. That relative shift matters because the market is likely to reward “asset-light AI monetization” and punish models where revenue visibility comes with rising leverage, working-capital drag, or capex front-loading.

The main risk is a two-stage reset: first, estimate revisions cool after the analyst-day narrative is absorbed; second, the market starts discounting fiscal 2027 cash generation instead of fiscal 2026 revenue growth. If management reinforces capex discipline or shows accelerating deal conversion at analyst day, the drawdown can reverse quickly; if not, the stock could underperform for several months as sell-side targets converge toward more realistic margin and financing assumptions. The move looks more like a valuation and positioning unwind than a structural thesis break, which argues for using strength to fade rather than chasing a momentum reversal.

Contrarian read: consensus may be underestimating how much of Oracle's AI upside is already embedded in expectations for gigantic deal flow, while underestimating the cost of turning those signed contracts into durable earnings. The market tends to reward top-line surprises in the first inning of an AI cycle, but later rewards capital efficiency and incremental ROIC. That makes the near-term asymmetry less attractive than the headline growth suggests.

Key Decisions for Investors

• Trim ORCL into analyst-day strength over the next 1-2 weeks; risk/reward favors taking down exposure before management has to reconcile AI demand with fiscal 2027 capex/funding needs.
• Initiate a relative-value short ORCL / long MSFT or AMZN pair for the next 1-3 months; thesis is that higher-quality AI monetization with stronger balance sheets should re-rate versus capex-heavier growth.
• If options liquidity is sufficient, buy ORCL put spreads 4-8 weeks out to express a controlled downside view into the analyst-day catalyst; target a 1:2 to 1:3 premium-at-risk profile.
• For accounts needing AI exposure, rotate part of the ORCL position into lower-funding-risk beneficiaries of enterprise AI spending rather than pure infrastructure builders; prefer names where AI demand does not require near-term balance-sheet expansion.
• Set a trigger to cover any ORCL short on a decisive upward revision to fiscal 2027 free cash flow or explicit capex moderation; absent that, use rallies to add to the short on a 2-4% intraday pop.