Jul 9, 2026

Regime & Verdict#

• Regime: UPTREND (SPY 10d +2.52%, 5d +0.80%; QQQ 10d +1.78%, 5d −0.26%) · directional_tradable = TRUE · s1_standdown = FALSE (both momentum legs live, no crash cap)
• Vol-state: SPY sitting ON the zero-gamma flip (spot $751.32 vs flip $751.74) — regime-transitional, a 2nd-moment conditioner only, no direction read. VIX 15.84, declining.
• Breadth: price breadth 59.84% green vs. options-flow breadth soft-dissenting at 36.8% bullish — the uw market-regime composite calls for book-wide HALF position sizing on this divergence. This is a ceiling on top of (not a replacement for) the excess-implied size — it is a no-op tonight: every sized name already caps at or below half from its own lane ceiling (OI_FADE HIGH-lane names cap at half by construction; S2/S4 never clear starter).
• Event risk: June CPI 2026-07-14 + PPI 2026-07-15 (both Tier-1) fall inside the h10 window opened by any short taken this week and inside the S2 h3–5 window. Big-bank earnings (JPM/BAC/C/WFC) also land 7/14, inside h10, but don't touch any sized name — banks fail the vol-book IV-rank≥80 gate this run. A separate OI_FADE earnings cluster (EQR/TMHC/CSX/RJF 7/22, STLD 7/20) lands inside h10 and — combined with missing fundamentals verdicts on all five names — is held at watch, not sized. FOMC (7/28–29) is ~13 trading days out, just outside every h10 window this week.
• On a beta-heavy tape the gate stack matters more, not less (RESEARCH/50 §5.1: ungated additive-confluence HIGH book realized −8.2%, 0-for-6).

Bottom line: 5 sized names (3 half, 2 starter), all in OI_FADE — the only lane clearing the +1% HIGH bar tonight. UDR, MAT, and NKTR are clean CONFIRM cases with no event risk inside h10 and no cluster/tail flags — sized at the HIGH-lane half ceiling. FROG is tail-adjusted to starter on a genuine CAUTION fundamentals tension (beats + 25% growth vs. insider selling). MARA is the most interesting case tonight: CONFIRM fundamentals and same-day OI-print corroboration make it look like the cleanest setup, but beta 5.38 plus a live bullish TX-power catalyst is an extreme, explicitly-flagged squeeze tail — capped at starter with a mandatory defined-risk structure (no naked short), regardless of signal quality (no additive confluence). CPRX is a clean VETO: 4/4 beats, +11.65% revenue, zero debt fight the short, and the flow is a dated S&P SmallCap600 removal (technical, not thesis). Five more OI_FADE candidates (EQR/TMHC/CSX/RJF/STLD) sit at watch on missing fundamentals verdicts stacked with earnings inside h10. MUR/BP (Energy cluster pair) and NVDA/MSFT (S2) also sit at watch — S2's MSFT is independently cut by its own same-day-negative-news tail-cap rule regardless of its CAUTION verdict. All ten S4 names are advisory-lane and missing fundamentals verdicts — held at watch across the board. MOM_SHORT/MOM_LONG fired no named tickers this run (aggregate cohort stats only) — flagged back to the orchestrator to emit names next run.

Directional Book (excess-scored)#

Cut to watch/dropped (not sized, detail below): OI_FADE — CPRX (VETO), EQR/TMHC/CSX/RJF/STLD (earnings inside h10 + missing verdicts), MUR/BP (Energy cluster, missing verdicts). S2 — NVDA (CAUTION, −1 tier), MSFT (same-day negative news tail-cap cut, independent of its CAUTION verdict). S4 — all ten names (BCE/AMRZ/SYY/BKD/EXE/KD/RARE/LYB/SPG/VALE) held at watch, missing fundamentals verdicts (advisory lane). MOM_SHORT/MOM_LONG — no named tickers surfaced this run.

Vol Book (non-directional)#

• NFLX (primary) — earnings 2026-07-16 postmarket, IVR 98.7, implied move 1.47%. Liquidity PASS ($75.47, $3.73B ADV). Crush survives cost. Small size, wide wings, two-sided tail; net-of-cost expectancy not independently recomputed — do not enter until it prices net-positive.
• TSM (primary) — earnings 2026-07-16 premarket, IVR 94.5, implied move 1.94%. Liquidity PASS ($436.96, $6.56B ADV). Crush survives cost. Same structure discipline as NFLX.
• ASML / JNJ / ABT / CTAS / CAG (secondary, half-size wide-wing advisory) — IVR 93.9 / 84.5 / 96.3 / 95.1 / 83.5 respectively, earnings 7/15–7/16, all premarket. Implied moves 2.50% / 1.17% / 1.01% / 1.47% / 1.71%. All clear the $50M liquidity floor comfortably (thinnest is CAG at $317.7M ADV). CTAS/CAG/ASML/JNJ report on PPI day (7/15) — the print itself doesn't gate a delta-neutral book, but expect wider realized ranges into the print.
• Banks: none clear the IV-rank ≥ 80 gate this run — the 7/14–15 bank earnings cluster produces no vol candidate tonight (contrast STT last week).
• SPY-0DTE: reduced-conviction half-of-half. IVR 14.4 (LOW vol), sitting on the zero-gamma flip — thin net-of-cost edge (~+3bps/day). Keep structures small; the flip proximity means a modest spot move can flip local gamma sign intraday.
• QQQ-0DTE: STAND ASIDE. Front-month backwardation gate fails, 30d signal reads PREMIUM_BUYING (not selling), and GEX data is unreliable tonight. No edge to sell premium; the unsampled left tail is not compensated.

Watch / Stood-down#

• OI_FADE VETO (dropped entirely): CPRX — 4/4 EPS beats, +11.65% revenue, 37% margin, zero debt directly fight the short; the flow driver is S&P SmallCap600 index-removal effective 2026-07-15, a dated mechanical event, not a thesis-relevant positioning signal.
• OI_FADE earnings-inside-h10 + missing verdict (held at watch, flagged back to orchestrator for next-run spawn): EQR (earn 7/22, REIT-cluster partner to UDR — corr≥0.70 pairing noted, moot while watch), TMHC (earn 7/22), CSX (earn 7/22), RJF (earn 7/22), STLD (earn 7/20). All five would additionally require a defined-risk-through-print structure once/if a fundamentals verdict clears, given earnings land inside the h10 window.
• OI_FADE Energy cluster, missing verdict: MUR, BP — corr≥0.70 pair per Phase-B pairing note, flagged back to the orchestrator; would collapse to one theoretical position (oi_fade_energy_cluster_0709) once verdicts clear.
• S2 — NVDA held at watch (CAUTION, −1 tier): genuine one-sided DP concentration (98.5% buy-side, $691M) with clean earnings/growth fundamentals otherwise, but insider MSPR −98.61 into the accumulation is enough of a flag to hold at watch rather than size starter tonight.
• S2 — MSFT held at watch (independent tail-cap cut): strong DP concentration (96.6% buy-side, $707M) is exactly the S2 setup, but a same-day negative news cluster triggers the lane's own explicit tail-cap rule ("cut on any same-day news/catalyst") regardless of flow quality — this is a sufficient cut on its own, and compounds with (does not sum with) the separate CAUTION fundamentals verdict (insider MSPR −83.89).
• S4 sentiment-contrarian — all ten names held at watch, missing fundamentals verdicts (advisory lane, flagged back to orchestrator): BCE, AMRZ, SYY, BKD, EXE, KD, RARE, LYB, SPG, VALE. SYY carries an extra Phase-B flag — a real downgrade requiring diligence, not noise — elevated risk vs. the rest of the cohort even once a verdict lands. KD's liquidity is the thinnest of the group ($55.9M ADV) but still clears the $50M floor. S4 is advisory-only by lane construction regardless (half-cap ceiling even at full conviction), so this cohort was never going to clear starter tonight even with clean verdicts.
• MOM_SHORT — no named candidates fired. Phase B reported aggregate cohort stats only (29 liquid near-52w-low names, 6 sector cohorts flagged for corr≥0.70 collapse, bond-ETFs deweighted as rate-beta not weakness) but no ticker list reached Phase D this run. Not S1-crash-gated (s1_standdown=FALSE) — the gap here is a data-pipeline one, flagged back to the orchestrator to emit names next run.
• MOM_LONG — no named candidates fired. Same gap: 378 liquid near-52w-high names reported in aggregate (median≈0, edge outlier-carried; 9/20 top names healthcare, sector-cap noted) but no ticker list reached Phase D. Basket-only by construction regardless (never per-name HIGH, tail-driven +2.35% mean) — flagged back to the orchestrator.

Risk#

• Correlation clusters applied: (1) REIT pair UDR–EQR — EQR isn't live (event-risk + missing verdict), so UDR sizes standalone; re-evaluate the pair once EQR clears. (2) Energy pair MUR–BP — both held at watch on missing verdicts, would collapse to one position once live. No other pairwise correlations ≥0.70 identified among today's five sized names — UDR/MAT/NKTR/FROG/MARA span Real Estate, Consumer Cyclical, Healthcare, Technology, and Financial Services with no shared theme.
• Extreme tail cap — MARA: beta 5.38 is the highest in tonight's book by a wide margin, and the bullish TX-power catalyst is a live, ongoing news thread, not a one-off. This is exactly the squeeze-tail case the gate stack exists to catch — CONFIRM fundamentals and same-day OI corroboration do not buy back the tail risk (no additive confluence). Sized at starter with a mandatory defined-risk (put-spread) structure — no naked short on this name under any circumstance tonight.
• Event overhang: CPI (7/14) + PPI (7/15) sit inside every h10 window opened this week and inside the S2 h3–5 window. Applied as defined-risk-through-print on UDR/MAT/NKTR/FROG/MARA (OI_FADE) — not an automatic downgrade, since none of these are macro-directional bets. The separate OI_FADE earnings cluster (EQR/TMHC/CSX/RJF/STLD, 7/20–22) is a harder gate — earnings inside h10 stacked with a missing fundamentals verdict is enough to hold all five at watch outright, not merely defined-risk.
• Liquidity floor: every sized name and every vol-book name clears the $50M ADV floor with real margin (thinnest sized name is MAT at $62.9M ADV). No liquidity-driven drops tonight — verified directly against the Stock Screener parquet (2026-07-09 snapshot) via DuckDB rather than trusting Phase B's pre-applied floor.
• Regime-cap discipline: the book-wide HALF ceiling from the uw market-regime composite (options-flow breadth soft-dissent) is a no-op tonight — every sized name already caps at or below half from its own lane ceiling. The hard rule (never up-size above the excess-implied size) held throughout — FROG's and MARA's tail-cap downgrades are both reductions, never upgrades.
• Missing-verdict discipline: seven OI_FADE names (EQR/TMHC/CSX/RJF/STLD/MUR/BP), both S2 secondary candidates, and all ten S4 candidates are sitting on the sideline at least partly because a fundamentals verdict hasn't been run yet — flagged back to the orchestrator, not silently dropped or silently sized. This is the largest missing-verdict count of any run to date; worth a data-pipeline note for the orchestrator (see below).
• Data-pipeline note (not a gate, a process flag): MOM_SHORT and MOM_LONG reported only aggregate cohort statistics tonight, no ticker lists — Phase D cannot size, cluster, or fundamentals-gate a basket it doesn't have names for. Recommend Phase B emit the actual name lists alongside the cohort stats next run.
• Hedge note: five shorts (three half, two starter) against zero live longs tonight (both S2 candidates held at watch) — net book exposure is short and moderate, concentrated entirely in the OI_FADE lane. In an UPTREND regime with no s1_standdown, this is a deliberate short bias built entirely from idiosyncratic flow+fundamentals confirmation, not a index-directional bet — consistent with invariant #2 (no additive confluence, one lane scores at a time) and invariant #3 (say so when the book is thin/skewed).

---

Envelope: decision.json (schema v2, validated — python3 scripts/validate_decision.py --file analyses/scan/2026-07-09/decision.json → VALID). Conviction write-back persisted to conviction_2026-07-09.json for next run's adverse-flow exit check (UDR, MAT, NKTR, FROG, MARA). Regression gate not run tonight — no lane/threshold changes made.