IBM just had its worst day ever. Oppenheimer thinks investors should stay away from the stock

IBM sank ~25% (worst day ever) after preliminary Q2 results missed expectations, posting adjusted EPS of $2.93 vs $3.01 consensus and revenue of $17.2B vs $17.86B. Oppenheimer downgraded IBM to Perform from Outperform, citing that the bull thesis will take longer and that the stock is likely range-bound near term. The note warns it may be difficult for IBM to meet full-year goals or achieve double-digit CC software growth in 2026/27, increasing downside risk as shares are down ~27% YTD.

Analysis

This is less a one-quarter miss than a credibility break in a mature incumbent whose valuation depends on management’s ability to bridge the gap between “turnaround” and actual re-acceleration. When a stock is already de-rated, the next leg lower usually comes from guide risk and multiple compression, not another small earnings revision. The immediate beneficiaries are best-of-breed infrastructure and data software names that can sell around IBM’s bundle, with CFLT a plausible second-order winner if enterprise buyers keep favoring modular platforms over broad vendor stacks.

The key near-term catalyst is the formal earnings print and, more importantly, whether management defends full-year margins, free cash flow, and software growth assumptions. If that guidance is walked down, IBM can trade like a low-growth cash machine rather than a transformation story, which typically means another 5-10% downside over the next 1-3 months as longer-duration holders de-risk. If guidance is merely reaffirmed, the stock can bounce, but the ceiling remains low until investors see two clean quarters of sequential improvement.

The contrarian view is that the selloff may be too blunt if the weakness is concentrated in a few lumpy lines and cash generation holds up. IBM still has a large installed base and can stabilize faster than the market assumes, so the thesis is falsified if Red Hat-style growth re-accelerates, FCF guide is unchanged, and bookings commentary improves on the July print. Absent that, the risk/reward still favors fading rallies rather than catching the falling knife.

Key Decisions for Investors

• Short IBM into any post-downgrade relief bounce; use the July 22 print as the event risk. Risk/reward is favorable if the market starts pricing a guide cut, with a stop if management reaffirms FY outlook and software growth.
• For a cleaner relative-value expression, long CFLT / short IBM for 1-3 months. The thesis is that enterprise spend rotates toward modular data infrastructure while legacy bundle vendors lose pricing power; exit if IBM stabilizes and CFLT’s next update softens demand.
• Buy IBM puts only after a failed bounce toward the pre-gap area; prefer 1-2 month tenor to capture guide-risk asymmetry. This avoids paying peak implied volatility immediately after the shock.
• Watch the July 22 earnings call for three falsifiers: maintained full-year revenue growth, unchanged free cash flow, and evidence that Red Hat is re-accelerating. If all three hold, cover shorts and expect a range-bound tape instead of another leg down.