Market scan · Aug 7, 2026

RegimeUPTREND/REBOUND-THRUST

breadth63.8% green5d+3.51%10d+4.65%Directional tradableS1 stand-down

Vol state

LONG-GAMMA REGIME CHANGE vs 08-05/08-06 (both were short-gamma). VIX 14.90, a FRESH CYCLE LOW (LOW tercile, cuts 16.35/18.43). SPY GEX +$1.85B, spot 772.84 vs flip 772.36 = +0.06% ABOVE flip (long gamma, razor-thin cushion -- roughly a 0.5pt give-back flips it short). QQQ GEX +$853M, spot 722.39 vs flip 721.08 = +0.18% above flip (long gamma). The 0DTE-VRP mechanical gate is GREEN for the first time in 3 sessions (sell_premium=true structurally, size_scalar 0.5 both names) but the LOW-VIX-tercile CONDITIONAL expectancy is NET NEGATIVE for both SPY (-0.059% gross -> ~-0.16% net) and QQQ (+0.016% gross -> ~-0.08% net) once costs are applied -- see vol_book, STAND DOWN on economics not mechanics
the full-sample net-positive VRP headline is carried entirely by the MID/HIGH VIX buckets. || S1_STANDDOWN MARGIN: ret5 +3.510% vs the +2.5% trigger = +1.01pp. Trajectory 08-05 +3.03pp -> 08-06 +1.12pp -> 08-07 +1.01pp, i.e. -0.11pp today vs -1.91pp the prior session -- decay has FLATTENED, not reversed
dd15 -1.28% shows no pullback underneath the thrust yet. This is the up-thrust/rebound leg of the standdown rule (fires in both mean-reversion regimes per the gate stack), so it zeroes relative-weakness shorts AND is the OI_FADE lane's own Hard Rule #2 trigger tonight. || BREADTH-VS-FLOW DIVERGENCE FLAGGED: 321 adv / 180 decl, pct_green 63.82% (green tape), but options-flow bullish only 37.7% -- flow is not confirming the breadth. Soft qualitative caution layered on top of MOM_LONG, not a formal gate.

3 event-risk item(s) flagged.

Tier mix

No calls cleared the board on this scan.

No call cleared HIGH/MEDIUM/ADVISORY tier in this scan — no new directional edge.

SPY

vol_0dtebuy premium

Vol state: long_gamma_low_vix_net_negative
Net expectancy: -0.002

Stand aside Mechanical 0DTE-VRP gate is GREEN for the first time in 3 sessions (sell_premium=true structurally, size_scalar 0.5, dealers LONG gamma: spot 772.84 vs flip 772.36 = +0.06% above flip -- a regime CHANGE vs 08-05/08-06 short gamma). BUT conditioned on tonight's LOW VIX tercile (14.90, cuts 16.35/18.43), the tail-cap directive to size for the unsampled left tail and quote NET OF COST flips this: SPY's LOW-VIX-tercile expectancy is -0.059% gross -> ~-0.16% NET -- the full-sample net-positive VRP headline is carried entirely by MID/HIGH VIX buckets, not this one. SPY's +0.06% cushion above flip is razor-thin; roughly a 0.5-point give-back flips dealers back to short gamma intraday, exactly the unsampled left tail this cap exists for. STAND DOWN on economics, not mechanics.

QQQ

vol_0dtebuy premium

Vol state: long_gamma_low_vix_net_negative
Net expectancy: -0.001

Stand aside Same mechanical GREEN (dealers long gamma, spot 722.39 vs flip 721.08 = +0.18% cushion), same LOW-VIX-tercile override: QQQ's expectancy is +0.016% gross -> ~-0.08% NET once costs are applied -- gross-positive, net-negative, precisely the class of trade the net-of-cost tail-cap discipline exists to catch. STAND DOWN on economics.

ENVX

vol_earningsbuy premium

Vol state: rejected_fat_tail_not_crush
Implied move: 19.90%

Stand aside REJECTED at the contract level: implied ~19.9% against an underlying already realizing ~93-102% annualized with a chronically fat tail -- rich IV is compensation for a genuine tail, not a crush edge. Orchestrator-corroborated after the fact: ENVX ran +14.9% over the next 5 trading days and +19.0% over 10. No position.

TRMB

vol_earnings

Vol state: unverified_pending_deep_dive

Stand aside Flagged by the lane but NOT deep-dived at the contract level before this Phase D run. Same discipline as a missing fundamentals verdict: hold at watch rather than size on an incomplete read. Flagged back for the next vol-book pass; not sized tonight.

Nothing in this scan landed at WATCH, STOOD_DOWN or DROP.

laneOI_FADEStood downShortregime fit0.00

final_size=none. horizon_validated=10, validated_excess=+0.0071 (re-baselined 2026-08-01; standing forward-decay watch -- the 2026-08-01 new-evidence cohort ran -3.73%, and the post-fix h10 cohort matures 08-03->08-07, closing TODAY; the 08-08 /calibration-audit is the live test). REGIME-BLOCKED, not mechanism-blocked: USFD and UTHR both cleared the ENTIRE mechanical stack (rel_build 2.42/1.31, persistence 1.074/1.031, earnings clear 11-05/11-04, catalyst_split LIVE 94.5%/67.2% post-print) and are liquidity-verified via scripts/liquidity.py against the 2026-08-07 screener panel: USFD close $108.88, $-ADV $276.8M PASS; UTHR close $538.84, $-ADV $240.9M PASS -- both figures reproduce the lane's reported numbers exactly, issue_type Common Stock on both. Both are blocked SOLELY by the lane's own Hard Rule #2 (stand down on a regime-classifier strong-rebound-thrust flag), which fires because s1_standdown=true tonight under UPTREND/REBOUND-THRUST -- the up-thrust/rebound leg the standdown rule is explicitly written for. Fundamentals on file for the record (moot while regime-blocked): USFD CAUTION (Q2 beat + guide raise + 3 same-day PT hikes partially JUSTIFIES the OI build, would cost -1 tier); UTHR CONFIRM (Q2 Tyvaso miss, stock -7.5% on the print, insider MSPR -31.76 selling 19/20 months, fundamentals SUPPORT the fade, would cost 0 tier). Correlation gate 3 INCOMPLETE, FAIL-CLOSED: USFD is ABSENT from data/prices.parquet (frozen 785-name universe per the prices-parquet-coverage-gap memory) -- the USFD-UTHR pair could not be computed. Do not treat the silence as a pass; if the standdown ever lifts, verify UTHR's correlation exposure via a non-parquet source before sizing both together. NXPI cleared every scripted gate but was cut UPSTREAM by the lane itself on the M&A/single-day-news check (live unresolved Ambarella acquisition talk + 98.6% single-day build) -- MECHANISM-blocked before reaching the regime question, never handed to Phase D.

USFDUTHR

laneMOM_SHORTStood downShortregime fit0.00

final_size=watch. horizon_validated=10, validated_excess=-0.0008 (NEGATIVE baseline, invariant #6's one recorded exception). TTD ($13.80, 2.2% of 52w range, liquidity PASS $247.2M ADV via scripts/liquidity.py, Common Stock) is TRIPLY blocked: (1) REGIME -- s1_standdown=true zeroes relative-weakness shorts tonight, and UPTREND/REBOUND-THRUST is exactly the up-thrust/rebound leg the gate is written for (margin +1.01pp over the +2.5% trigger, decay flattening 08-05 +3.03pp -> 08-06 +1.12pp -> 08-07 +1.01pp); (2) invariant-#6's structural PROVISIONAL watch-only cap on ALL new MOM_SHORT starters pending a DURABLE-N(>=30) forward-positive window -- the 07-18 audit's first matured forward window ran 0-for-9, -5.3% excess; (3) the lane's own recorded baseline is negative, the binding read even though hit-rate is up. POST excluded upstream on a prior VETO on file; PLTD excluded upstream as an inverse leveraged ETP (issue_type filter worked correctly here) -- neither reached Phase D.

TTD

laneMOM_LONGBasket watchLongregime fit1.00

final_size=watch. horizon_validated=10, validated_excess=+0.0023 mean, MEDIAN NEGATIVE, tail-driven, n=483. Barred on THREE independent counts, none of them regime (this is an in-regime long lane in an uptrend): (1) TIER MATH -- +0.23% sits BELOW the sizing map's +0.30% starter bar; the Phase C read is CONFIRMED correct on its own, this basket cannot clear even a starter tier on excess alone. (2) TAIL CAP -- basket-only, never per-name HIGH; the negative median on top of the mean shortfall is the binding distributional read, not the headline mean. (3) DURABLE-N -- a 3-name basket is thin against the lane's separate DURABLE-N bar, re-enabled only via a /calibration-audit recommendation. Sector cap (gate #7) ITSELF passes clean: ABNB Consumer Cyclical / NTRA Healthcare / FIVN Technology, no sector >33%. Correlation gate 3: ABNB-NTRA computed at corr=0.55 (data/prices.parquet adjclose daily returns, 2026-05-01->08-07, n=68 trading days) -- below the 0.70 cluster threshold, no collapse. FIVN is ABSENT from data/prices.parquet (frozen 785-name universe) -- FAIL-CLOSED per the coverage-gap memory: FIVN's correlation to ABNB/NTRA is UNVERIFIED, not confirmed-independent. Moot tonight since the whole basket is watch-only regardless; flag for any future promotion. Fundamentals: ABNB CAUTION (-1, real beat/raise but insider MSPR -70.98 selling 17/20 months into a 42.6x P/E at a fresh high); NTRA CAUTION (-1, real beat/raise but insider MSPR -61.28 and net margin -9.05%, still unprofitable); FIVN CONFIRM (0, beat+raise+S&P SmallCap 600 inclusion, insider neutral, caveat GAAP EPS guide cut and index-flow-assisted). Event risk: CPI 08-12 and PPI 08-13 both land inside this h10 window -> an ADDITIONAL -1 tier if this basket is ever sized (ABNB/NTRA stack to -2 total, FIVN to -1) -- moot while watch-only. Breadth-vs-flow divergence (63.82% green tape, 37.7% bullish flow) is a soft qualitative caution layered on top, not a formal gate.

ABNBNTRAFIVN

USFDUTHRTTDABNBNTRAFIVNTRMB

Lane glossary (5)

vol_0dte: Vol book — 0DTE variance-risk-premium premium-selling. Non-directional, delta-neutral, 0 directional points.
vol_earnings: Vol book — sells volatility into an earnings IV-crush. Non-directional, delta-neutral, 0 directional points.
OI_FADE: OI-flow fade — fades persistent multi-day net call-OI building; heavy call-OI build precedes underperformance. The most robust lane measured (short, h10).
MOM_SHORT: Momentum, short leg — near-52w-low relative weakness. Regime-gated against momentum-crash/squeeze rebounds (h10).
MOM_LONG: Momentum, long leg — near-52w-high relative strength. Tail-driven; basket-only, never sized per-name (h10).

Regime & Verdict#

Regime: UPTREND/REBOUND-THRUST · ret5 +3.510% · ret10 +4.650% · dd15 −1.280%
Vol-state: REGIME CHANGE vs the prior two sessions. VIX 14.90, a fresh cycle low (15.86 → 16.50 → 15.81 → 15.15 → 14.90), LOW tercile (cuts 16.35 / 18.43). Dealers are now net LONG gamma in both indices: SPY spot 772.84 vs zero-gamma flip 772.36 (+0.06% above), QQQ spot 722.39 vs flip 721.08 (+0.18% above). Aggregate GEX positive in both (SPY +$1.85B, QQQ +$853M) — that sign sets range/wing-width only; spot-vs-flip sets the regime, and tonight it says long gamma, where on 08-05/08-06 it said short. SPY's cushion is razor-thin: a ~0.5pt give-back flips it back to short gamma while QQQ holds.
Breadth: 321 adv / 180 decl, pct_green 63.82%, avg +0.70% — a genuinely green tape. But options-flow bullish is only 37.7% (2,368 of 6,281). Breadth-vs-flow divergence: YES — flow is not confirming the price breadth. Rotation is into Technology/Industrials/Consumer Defensive, out of Communication Services/Consumer Cyclical/Healthcare. uw risk market-regime reads TRANSITIONAL/reduce-size.
directional_tradable = TRUE (regime admits lanes) · s1_standdown = TRUE
Stand-down margin: ret5 +3.510% vs the +2.5% trigger → +1.01pp. Trajectory: 08-03 +0.01pp → 08-04 +1.61pp → 08-05 +3.03pp (peak) → 08-06 +1.12pp → 08-07 +1.01pp. The decay has FLATTENED — this is the change from yesterday. Last session's −1.91pp collapse became −0.11pp tonight, and ret10 kept rising (+4.12% → +4.65%). Yesterday's read was "roughly 1.1pp more decay ends the stand-down"; tonight that no longer looks imminent. The thrust stopped decelerating.

Bottom line: no directional edge today. ZERO new starters for an EIGHTH consecutive session (07-29 → 08-07); the book stays FLAT for a fifth. The vol book is EMPTY for a third — but tonight it stands down on ECONOMICS, not mechanics.

Two things genuinely changed tonight, and they point in opposite directions:

OI_FADE recovered to two mechanically-clean names (USFD, UTHR) after 08-06's pure zero. Both cleared the entire mandatory stack and are blocked only by the lane's Hard Rule #2 regime stand-down. UTHR is the strongest single candidate on file in over a week — CONFIRM fundamentals, clean catalyst survival, no M&A confound. Regime-blocked, not mechanism-blocked.
The 0DTE mechanical gate flipped GREEN for the first time in three sessions (dealers rotated long gamma) — and the sleeve still stands aside, because conditioned on tonight's LOW VIX tercile the net-of-cost expectancy is negative. The mechanics finally cooperated; the economics did not.

Directional Book (excess-scored)#

Ticker	Lane	Dir	Horizon	validated_excess	regime_fit	invalidation	gates
—	—	—	—	—	—	—	—

Empty is a valid and common output (RESEARCH/20 §2.0). Nothing scored; nothing sized.

Vol Book (non-directional)#

EMPTY — third consecutive session, but for a NEW reason.

0DTE-VRP — mechanical gate GREEN, stood down on net-of-cost economics. sell_premium=true, size_scalar=0.5 for both SPY and QQQ — dealers are long gamma (spot above flip in both), no VIX spike, and no front-end backwardation on the near-ATM basis the gate actually uses. The binding read is the state-conditioned expectancy: tonight's VIX 14.90 sits in the LOW tercile, where the gross mean is SPY −0.059% and QQQ +0.016% — charge the 0.10% round-trip and both go net negative (SPY ≈ −0.16%/day, QQQ ≈ −0.08%/day). The headline full-sample net-positive figure (+0.123% / +0.26%) is carried entirely by the MID/HIGH-VIX buckets. Selling cheap vol into a suppressed tape is exactly the trade that pays gross and loses net. Methodology note: the raw iv-term-structure full-0DTE average shows SPY 1.34× / QQQ 2.09× VIX, which looks backwardated, but that metric is contaminated by deep-OTM 0DTE contracts. The near-ATM 6-strike calc on the next tradeable expiry is under the 1.25× gate. This is not a repeat of 08-06's QQQ 1.54× flag on the metric the gate uses. Gamma-sign discipline held per 5b3daec — aggregate sign drove the wing width, spot-vs-flip drove the sell decision, kept separate.
Earnings IV-crush — ENVX examined at contract level and REJECTED. This corrects yesterday's provisional flag. Pulled the raw trade tape (the hot-chains file is missing the ATM put leg): spot $4.70, ATM straddle ≈ $0.93–0.94 → implied move ≈ 19.9% for the 08-14 expiry, term structure genuinely backwardated (193% front vs 118–155% back). But the underlying is chronically fat-tailed — trailing realized vol ~93–102% annualized (daily 1σ ≈ 6.3%), with at least three

9–15% single-day non-event moves in the trailing 90 sessions. Independently corroborated: ENVX ran +14.9% in 5 days / +19.0% in 10. Wings sized off a 20% implied move get blown through by ordinary chop, not just the print. Rich IV here is paid-for tail, not free premium. Yesterday's "mean abs 3.3%, no >13% print in 10 quarters" characterization does not survive contact with the panel.
TRMB — adequate flow ($472k combined), iv_rank 91.4, not deep-dived. Explicitly held at watch rather than silently dropped; candidate for tomorrow.
10 of ~15 screened names excluded on liquidity (premium in the tens/hundreds of dollars).

Watch / Stood-down#

OI_FADE — two clean names, both REGIME-blocked (the session's most important item)#

Ticker	Price	20d $-ADV	rel_build	persist	catalyst_split	Fundamentals	Binding constraint
UTHR	$538.84	$240.9M	1.31	1.031	LIVE (67.2% post-08-05 print)	CONFIRM	Hard Rule #2 (regime)
USFD	$108.88	$276.8M	2.42	1.074	LIVE (94.5% post-08-06 print)	CAUTION (−1)	Hard Rule #2 (regime)

UTHR is the cleanest short on file in over a week. Fundamentals actively support the fade: Q2 revenue missed on flagship Tyvaso against competitive pressure, the stock fell ~7.5% on the print, and insider MSPR is −31.76 with net selling in 19 of 20 months. The crowd is building calls into a name whose growth driver just cracked.

USFD is the weaker of the two despite the larger build: its Q2 was a genuine beat with FY guidance reaffirmed and three same-day price-target raises. That makes the call-OI build partly justified rather than pure crowd-chasing — CAUTION, −1 tier.

NXPI — cut on MECHANISM, not regime. Cleared every scripted gate (rel_build 1.61, earnings 10-26 clear, catalyst_split LIVE 98.6%) but the build is 98.6% single-day, coincident with a market-wide rally, while NXPI is named in unresolved M&A speculation as the acquirer of Ambarella (FT, 07-31). This is the AZN precedent from 08-05 — the 7-gate stack passes rumor-driven builds and catalyst_split reads them LIVE. Cut.
Phase A's mega-cap pre-screen was invalid and was discarded. NVDA/AAPL/AMZN/META/PLTR et al. show rel_build 0.04–0.26 against a top-15 cohort floor of ~1.3 — 2–20× below the bar. Huge absolute call OI, trivial relative build: the mega-cap beta trap CLAUDE.md warns about. None should have been candidates.
EXLS / UCTT re-checked (both were the 08-05 near-misses): EXLS rel_build decayed 0.964 → 0.956 and never re-inflated — mechanism dead. UCTT collapsed 0.982 → 0.28 — decayed out of the cohort entirely. Neither is a live candidate.

MOM_LONG — 3-name basket, blocked on TIER MATH#

ABNB ($178.07, 99.4% of 52w range, Consumer Cyclical) · NTRA ($322.10, 99.5%, Healthcare) · FIVN ($33.99, 99.5%, Technology). Sector cap (gate #7) satisfied at 33% each; HALO dropped to enforce it.

Binding constraint is arithmetic, not regime: MOM_LONG's measured excess is +0.23%, below the +0.30% MEDIUM-tier bar — watch-only before any other gate is consulted. Reinforced by the negative median (tail-driven distribution) and the risk-sizer's DURABLE-N bar, which a 3-name basket does not clear.

All three catalysts are real, which is worth recording even though nothing sizes:

ABNB (+17.43% today, the tape's top mover) — a genuine re-rating, not a squeeze: Q2 beat +7.29%, bookings accelerating, FY26 raised. But insider MSPR −70.98, worst in the batch, selling in 17 of 20 months, into a 42.6× P/E at a fresh high. CAUTION (−1).
NTRA — revenue +37.7% YoY on record oncology volumes, 7 same-day PT hikes; but insider MSPR −61.28 and still unprofitable (net margin −9.05%). CAUTION (−1).
FIVN — beat, raised sales guide, S&P SmallCap 600 inclusion as mechanical demand, insider neutral. CONFIRM (0) — the only clean long. Caveat: GAAP EPS guidance was cut, and part of the move is index flow whose durability past the inclusion window is unproven.

MOM_SHORT — TTD only, triple-blocked#

TTD ($13.80, 2.2% of 52w range, −21.90% today) is blocked by (1) s1_standdown, (2) the invariant-#6 watch-only cap on new starters, and (3) the negative −0.0008 baseline. POST excluded on a prior VETO (07-29); PLTD excluded as an inverse leveraged ETP — the ETP leak the issue_type filter exists to catch.

Advisory lanes — surfaced, structurally never size#

S2 (liquidity-reversion): TXN (DP $980.8M, 96.2% buy-side, ADV $2.25B) and RTX ($529.5M, 95.1%, ADV $1.14B). Both passed the full gate stack including the trading-day earnings window.
S4 (sentiment-contrarian): 26 names post-gate, led by LPX (PCR 25.59), BRZE (25.21), TENB (17.18), HUBB (16.26), URBN (15.88). 11 ETFs cut, 1 on earnings, 6 on prior verdicts. Plus a separate 18-name ivrank_chg_5d h3 tilt list (AEO, RBRK, PL, CM, HPQ…). S4 remains the only lane positive forward as well as historically. Data-quality flag: the lane returned hallucinated company-name annotations (e.g. "SN — Snap Inc", actually SharkNinja; "MTZ — Mitsubishi UFJ", actually MasTec; "CPRT — Carpetright", actually Copart). The PCR figures come from the panel and are sound; the name labels are not and have been stripped.

Risk#

Correlation clusters: trailing-3-month pairwise (2026-05-01→08-07, n=68) — ABNB–NTRA 0.55, everything else ≤0.32. Nothing clears the 0.70 collapse threshold. FAIL-CLOSED FLAG: USFD and FIVN are absent from data/prices.parquet (the frozen 785-name universe — prices-parquet-coverage-gap). The USFD↔UTHR pair and FIVN's exposure to ABNB/NTRA are unverified, not confirmed-independent. Moot tonight since neither lane sizes, but this must be resolved before either is ever promoted.
Event calendar inside the h10 window: CPI 2026-08-12 and PPI 2026-08-13 (both Tier-1), plus FOMC minutes 08-19 at day 8 of 10. Any h5/h10 position opened tonight would carry both prints — worth an additional −1 tier on anything sized. NFP landed this morning and is already in the tape.
Tail caps applied: MOM_LONG's negative median (tail-driven, basket-only, never per-name HIGH); OI_FADE's short-on-an-up-biased-tape starter cap; the vol sleeve's unsampled left tail, which is what the net-of-cost LOW-VIX read is protecting against.
Hedge note: book is flat — nothing to hedge. The breadth-vs-flow divergence (63.8% green tape vs 37.7% bullish flow) is a distribution tell carried into sizing discipline, not a veto.
What to watch tomorrow: (1) whether the stand-down margin resumes decaying or holds near +1.0pp — it flattened tonight, which delays rather than accelerates OI_FADE's re-opening; (2) whether USFD and UTHR hold their builds through another session; (3) SPY's spot-vs-flip cushion, thin enough that the vol regime could flip back overnight.

Standing note: tomorrow is the due date for /calibration-audit (2026-08-08). The post-hygiene-fix OI_FADE h10 cohort matured 08-03→08-07 — that audit is the first that can grade the current engine, and the test of the −3.73% forward-decay signal. Because the lane produced zero starters across that entire window, neither USFD nor UTHR adds a resolvable row to it.

Aug 7, 2026