MU

HIT :: earnings event :: tier RESOLVED

Registered claim

Micron (MU) Q3 fiscal-2026 print on 2026-06-25 confirms the HBM-fully-booked + multi-year-prepayment structure with explicit dollar disclosure of prepayment balance and/or named hyperscaler customer counterparty. Beat-and-raise on 2026 HBM revenue guide; Q4 FY26 (Sept) guide raised by ≥10% vs prior consensus.

Registered 2026-05-10.

Resolution spec

Spec registered 2026-06-12.

• [A] CORE :: MU Q3 FY26 print confirms the HBM prepayment structure with explicit dollar disclosure of the prepayment balance OR a named hyperscaler counterparty.
  kind: dated binary event :: event date 2026-06-25 :: evidence: primary source
• [B] confirm :: FY2026 HBM revenue guide raised vs prior guide at the Q3 FY26 print.
  kind: dated binary event :: event date 2026-06-25 :: evidence: primary source
• [C] confirm :: Q4 FY26 revenue guide >= 10% above pre-print consensus (frozen consensus snapshot, due 5 days before the print per policy).
  kind: numeric threshold :: q4_fy26_revenue_guide_vs_consensus_pct >= 10
  baseline 42.26 (consensus snapshot, frozen 2026-06-20)
  note: FQ4 FY26 revenue consensus (units: $B) = consensus-implied residual: FY2026 annual revenue consensus $113.26B (stockanalysis.com, 36 analysts, FY ending 2026-08-31) minus FQ1 FY26 actual $13.64B + FQ2 FY26 actual $23.86B (Micron primary-source results releases) minus FQ3 FY26 guide $33.5B = $42.26B. Caveat: Q3 leg uses company guide as the analyst-consensus proxy; ~±$1-2B model risk. Fetched 2026-06-20 per daily Step 8.6.

Verdict rule: HIT if leg A, and either leg B or leg C. PARTIAL if leg A, and neither leg B nor leg C. MISS if not leg A.

As frozen: HIT = A && (B || C); PARTIAL = A && !(B || C); MISS = !A

Tier history

• 2026-05-15: HIGH (initial). Trigger: Hold tier=high within photonics frame. MU HBM 2026 fully booked + multi-year prepayments + Mizuho PT to $740 + Bernstein commentary support the prediction directionally. SK Hynix HBM4 MP began Feb 2026 + M15X cleanroom complete two months ahead + HBM4E samples H2 2026 confirm cycle tightness. The DRAM-ETF speed-crash hypothetical (0.50 substack contradicts) is correctly classified low-weight against the broader cycle evidence.

Evaluated daily since registration: 41 evaluations. The 1 that moved the tier is listed above.

Resolution receipt

Verdict: HIT. Resolved 2026-06-25.

Rule: HIT if leg A, and either leg B or leg C. PARTIAL if leg A, and neither leg B nor leg C. MISS if not leg A.

As frozen: HIT = A && (B || C); PARTIAL = A && !(B || C); MISS = !A

• [A] PASS: MU FQ3 FY26 call named NVIDIA (Vera Rubin) as the HBM4 36GB 12-high counterparty (satisfies the named-hyperscaler-counterparty branch) AND disclosed 16 Strategic Customer Agreements backed by ~$22B in cash deposits + commitments (~$18B deposits + ~$4B LCs) - explicit dollar magnitude of the customer-commitment structure. MU characterizes the deposits as not prepayments, a balance-sheet nuance, but the named-counterparty branch is independently and unambiguously met.
• [B] PASS: FY2026 HBM revenue outlook raised: MU completed price-and-volume agreements for its entire calendar-2026 HBM supply incl. HBM4 (>$1B HBM4 shipped, ramping ~2x HBM3E) and raised the HBM TAM to ~$100B by 2028 (from ~$35B in 2025); the 2026 HBM revenue trajectory is revised upward vs the prior (Q2) guide.
• [C] PASS: actual_raw=50 vs frozen baseline 42.26 → 18.32% >= 10. FQ4 FY26 revenue guide midpoint $50.0B (in same $B units as the frozen consensus baseline $42.26B).