Will any Oracle insider make an open-market stock purchase by June 30, 2026?

The Q3 update materially strengthens the NO case. CEO Magouyrk continued discretionary selling at ~$155 — near the stock's lows — during a record quarter with 84% IaaS growth. This is the most bearish insider signal possible: selling at the bottom of a 40%+ decline during the best operational quarter in 15 years. The selling pattern now spans 6+ months with zero purchases across the most bullish guidance period ($90B FY27 target), the deepest stock decline, AND the strongest earnings beat. If insiders won't buy after this combination, there is no plausible catalyst for a purchase within the remaining 3.5 months. The class action (lead plaintiff deadline April 6) further constrains windows. Two blackout periods (Q3 reporting through mid-April, Q4 earnings in June) leave approximately 4-6 weeks of open trading, during which insiders who have been systematically selling have no demonstrated inclination to reverse course.

The prior assessment at 0.10 was already low, but Q3 data justifies a meaningful downward revision. The key new information: continued CEO selling at lows during a record beat, and Kehring's position now reduced ~75% from peak. The confidence upgrade from MEDIUM to HIGH on GOVERNANCE_ALIGNMENT reflects a pattern that has been tested against multiple scenarios and held. The behavioral signal is unambiguous — these insiders have had every conceivable reason to buy (stock at multi-month lows, record revenue growth, massive backlog) and instead chose to sell further. The Barrows v. Oracle class action with April 6 lead plaintiff deadline likely means enhanced trading restrictions through at least Q2 2026. Legal counsel in active securities fraud cases typically imposes a blanket moratorium on discretionary insider transactions. Combined with Q4 earnings blackout in June, the practical buying window collapses to perhaps 3-4 weeks in late April/May.

Considering the remaining YES paths: (1) New board appointment making a symbolic purchase — possible but unlikely during active litigation and typically small enough that legal counsel would still advise against it. (2) Ellison making an open-market purchase — as a 10%+ owner with $40B+ exposure, he has no need and the class action names executives, creating legal risk. (3) A mid-level officer making a small purchase — the most plausible path, as someone not named in the class action might have fewer restrictions. However, the 6+ month pattern of universal selling creates strong social/organizational pressure against breaking rank. The resolution window now has only 3.5 months remaining versus the original 4.5, and the two blackout periods are a larger percentage of remaining time. Net: the prior was 0.10 and every new data point pushes lower.

The Q3 earnings update is the critical new input. Oracle posted record revenue of $17.19B (+22% YoY) with cloud infrastructure growing 84% — the strongest operational quarter in at least 15 years. Despite this, insiders continued selling at ~$155, near the stock's low. This eliminates the hypothesis that insiders were waiting for better fundamentals to buy. They now have both a record quarter AND a deeply discounted stock, and still chose to sell. The GOVERNANCE_ALIGNMENT confidence upgrade to HIGH is well-justified. With 3.5 months remaining, two blackout periods, and the class action lead plaintiff deadline on April 6, the effective buying window is extremely narrow. The only remaining uncertainty is whether a new insider could be appointed and purchase, but the litigation environment makes this very unlikely.

Applying a systematic analysis of remaining buying windows: Q3 earnings filing/quiet period extends through approximately mid-April. Lead plaintiff deadline April 6 may trigger enhanced restrictions. That leaves late April through mid-May (~3-4 weeks) as the primary open window. Q4 FY26 earnings are expected in mid-June, creating a blackout from approximately late May through mid-June. If Q4 earnings fall after June 15, there may be no post-Q4 open window before the June 30 resolution date. So the question reduces to: will any insider who has been selling for 6+ months, during a record quarter, at stock lows, under active securities litigation, suddenly reverse course during a 3-4 week window in late April/May? The behavioral evidence says no. Kehring has reduced his position 75% — he is not a buyer. Magouyrk sold at lows — he is not a buyer. The class action deters everyone else.

While the evidence strongly points to NO, I assign slightly more uncertainty than some runs because: (1) We don't know the full universe of Section 16 insiders — Oracle likely has 15-20 reporting insiders, and we have detailed behavioral data on only 7-8. A VP-level officer we haven't tracked could make a small purchase. (2) The class action restricts named defendants more than other insiders — a director not named in Barrows v. Oracle might have fewer constraints. (3) Stock price could move dramatically in either direction, and a sharp decline to new lows might trigger a symbolic purchase by a board member trying to signal confidence. However, these scenarios are all low probability individually and the 6-month pattern is overwhelming evidence. The Q3 record beat is the nail in the coffin — there is simply no remaining catalyst that would be more bullish than what insiders already witnessed without buying.

The Q3 update is decisive. Record quarter with 84% IaaS growth, stock at lows around $155, and insiders STILL selling. This exhausts every plausible reason an insider might buy. Class action lead plaintiff deadline April 6. Two earnings blackouts. Effective open window: ~3-4 weeks. Probability of any purchase in that narrow window given 6+ months of universal selling: very low.

CEO selling at ~$155 during a record quarter is the single most informative data point. If the CEO won't buy at the bottom after their best quarter in 15 years, no insider will. Kehring at 75% position reduction is functionally exiting. The class action freezes discretionary trading. The remaining 3.5 months with blackouts leaves essentially one narrow window. The probability of any purchase approaches the base rate for a completely unexpected event (new board member, merger-related insider change) which is very low.

Updating from the prior 0.10 aggregate: every new data point from Q3 pushes probability lower. Continued selling during record quarter, confidence upgrade to HIGH, narrowing resolution window, approaching class action deadline. The only non-negligible YES scenarios require events not in the current information set (unexpected board change, dramatic corporate action). Assigning 0.06 to account for unknown unknowns while reflecting the overwhelmingly bearish insider behavior pattern.