Briefing reference: /Users/dadbot/Desktop/ClaudeBod/projects/dadbrain/Analysis/briefings/2026-08-05-pm.md PM: ECHO
Candidate triage from the Scanner watchlist (Directional Catalyst Instruments):
| Candidate | Bias | Verdict |
|---|---|---|
| XLE (EIA Wednesday 10:30 + Hormuz deal window TODAY) | Conditional DOWN (deal → crude breakdown) | Live — backtested, the week's strongest sector catalyst. EIA at 10:30 lands after my 10:00 signal close — the signal captures pre-print positioning, and the 15:20 gate stack self-corrects (VWAP persistence must agree with the 10:00 direction). Down-morning → LONG at 15:30; up-morning → SHORT. The historical bind remains vol_ratio: XLE has never cleared 1.2 live (best 1.192 on 7/23) — EIA-day volume is the one thing that could change that. |
| XBI (MRNA PDUFA decision TODAY) | Conditional UP (long if approved) | Live — backtested. Early approval → clean sector up-program → SHORT at 15:30. Rejection with first-half-hour ≤ −5% → chaotic-catalyst exclusion. Decision timing is unknown (anytime on the action date): a late-afternoon approval after my 15:20 gates is a tail risk on an XBI short — the 1.5×ATR stop is the control. vol_ratio bind: XBI live vols 0.386–1.119, never cleared 1.2. |
| SPY (macro data day: ADP 8:15, ISM 10:00) | UP (modest) | Secondary — backtested, the 8/3/8/4 repeat. vol_ratio cleared here two sessions running (1.3547, 1.4096) — the participation regime is real. But +0.36% futures is marginal vs my ±0.25% floor, ADP lands inside my signal window (whipsaw → direction-changed ≥2× disqualifier), and the BMO mega-cap cluster (DIS/UBER) contaminates the index program — constituent-level programs do not concentrate in SPY by 15:20. |
| AMD (−7 to −9% gap, beat-and-dump) | DOWN | Watch — NEW_CANDIDATE. The marquee gap. The LONG-at-15:30 fade requires the down-program to persist below VWAP all day. The 7/24 INTC learning stands: single-stock post-earnings reversals often complete in the first 90 minutes — if AMD snaps back before 10:00, the setup is already gone. Whether today is a real down-program or a headline yo-yo is set by the semis read-through (NVDA at the open). |
| SPCX (−10 to −12% pre-market, lockup Thu) | DOWN | Dropped — event/lockup-driven, weak program quality per the watchlist; supply-overhang sessions are not exhaustion programs. |
| NVO (earnings tank) / SHOP (miss if confirmed) / SPOT (aftermath) | DOWN | Watch — NEW_CANDIDATE. Smaller, cleaner gap-down fades than AMD if their down-morning persists to 15:20; same INTC structural caveat. |
The call: setup present — possible, same structural class as 8/4 and richer on candidates. The binding-gate story is now favorable: vol_ratio — the gate that bound me through the entire July dry spell — has cleared two consecutive sessions (1.3547, 1.4096) on the first clean trend-up days after the two-tier tape resolved, and today's volume environment (heaviest earnings day of the week, marquee gap-downs, data morning) should keep participation elevated. The modal qualifying path is the index fade — SPY short on an up-morning that survives ADP — the same trade that won 8/3 and stopped out 8/4; the sector paths (XLE short/long, XBI short on early approval) are the cleaner program structures if their catalysts fire; the single-name gap-fades (long AMD/NVO/SHOP) are real but carry the INTC early-completion caveat plus the first-session new_candidate spread observation. The honest caps: HIGH event risk with the data prints inside my signal window (ADP 8:15, PMI 9:45, ISM 10:00 all land before the 10:00 signal close — a hot ADP/ISM flips the morning direction), mega-cap earnings contamination of the index program, and the 8/4 execution lesson (on ≥1.6× ATR days the 1.5× ATR stop is clip-prone — ATRs are still elevated). That is 0.45 — same as 8/4: more candidates, but a messier tape and the same clip-risk.
Direction: SHORT — the modal session is a modestly-up index tape (futures +0.36%, momentum intact, record closes); the highest-probability qualifiers are up-morning instruments (SPY, XBI on approval, XLE on a deal-hope rebound), and I fade up-mornings. The down-morning counter-path (hot ADP/ISM flips the tape → long SPY; deal breaks oil → long XLE; AMD/NVO gap-fade longs) is real but secondary, and the single-name longs carry the INTC caveat. Multi-instrument entry permits mixed legs (e.g., short SPY + long AMD simultaneously — partially offsetting); the 2.0% session budget is the correlation control (v5 finding: same-session instruments are directionally correlated).
Sizing vs event risk: standard — 0.75% per instrument, 2.0% session budget, all instruments above T=0.80 enter together. HIGH event risk is not a hard-skip class (FOMC/CPI/NFP/VIX≥30/holiday only); my 15:30 entry is ≥ 5h after the last scheduled print (EIA 10:30) and I exit hard at 15:58 — zero exposure to any after-close earnings. The one live event at entry time: the MRNA PDUFA (anytime intraday) — the 1.5×ATR stop is the control. Single-name legs log as instrument_class: new_candidate (first-session close-window spread observation applies — do not re-enter until spread confirmed ≤ 0.10%); SPY/XLE/XBI are backtested.
Invalidation: first-half-hour |signal| < 0.25% on every candidate (the 7/31 fade pattern — plausible if ADP whipsaws the open); direction changed ≥ 2× in the first half-hour on index/sector candidates; vol_ratio < 1.2 at 15:20 on all candidates (XLE/XBI history says this is the likely binding gate again); exhaustion_score < 0.80; XBI rejection day with first-half-hour ≤ −5% (chaos flag, exclusion); AMD/NVO/SHOP gaps that reverse before 10:00 (INTC early-completion — do not reach at 15:30, let the gates decide); VIX prior close ≥ 30 (16.52 — clear).
The call: No trade expected — QQQ mega-cap earnings-reaction session, hard-skip class. Per the desk's settled working reading (carried from the 8/4 plan and reflection, flagged for the TEMPER temporal-scope conversation): the "QQQ (mega-cap) earnings dates" skip applies to the session the reaction trades. AMD (a QQQ mega-cap) reported after close 8/4 and its reaction trades TODAY (gap-down 5–9%); DIS and UBER (QQQ constituents) report before open today and their reactions trade in today's first hour. That is precisely the session class the skip exists for — earnings-reaction sessions break the three-phase structure. Independently of the skip, the structural read is negative on the same axis: TEMPER's 7/24 retrospective has SURGE 0/2 live on catalyst-driven gap sessions vs 4/4 on clean intraday trend sessions, and today is the quintessential catalyst-gap session (marquee gap-downs + BMO mega-cap cluster + data morning). The three-phase structure must not be forced. p_trade 0.05, direction none, conviction low. If TEMPER later overturns the skip-scope interpretation, today's session-character read would still argue no-trade — but the hard skip is the operative gate today, and the desk does not trade through its own settled reading.
No trades taken.