Briefing reference: /Users/dadbot/Desktop/ClaudeBod/projects/dadbrain/Analysis/briefings/2026-07-31-pm.md PM: ECHO
Setup assessment: A strong trending session is the exact setup ECHO needs — the more confirmed the program, the more complete the exhaustion, the stronger the close-leg edge. The AMZN +12.75% surge is producing real institutional program flow. ECHO needs: (1) first-half-hour signal_return ≥ ±0.25%, (2) VWAP persistence at 15:20 (price on correct side for morning direction), (3) ADX(14, 5-min) ≥ 20, (4) vol_ratio ≥ 1.2× 20-session avg, (5) exhaustion_score ≥ 0.80.
The directional call: if the morning bias is bullish (likely — SPY/QQQ gapping up, tech relief rally) → ECHO enters SHORT at 15:30 (contra). The AMZN-driven surge is a massive directional program — the more it runs through the session, the more exhausted the buyers are by 15:30.
Key concern: vol_ratio gate. The recent trajectory is promising — XLE hit 1.192 on 7/23 (0.008 below floor). Today's AMZN/AAPL divergence should generate elevated volume in broad-market ETFs. But vol_ratio ≥ 1.2 has been the binding constraint more often than not. ECI at 8:30 could spike early volume, helping the gate.
Candidate instruments: SPY (cautious bullish index), QQQ (bullish index), XLE (neutral, no catalyst), XBI (neutral, no FDA catalyst). SPY and QQQ have the strongest setup potential due to the earnings-driven direction. XLE/XBI unlikely to qualify without an independent catalyst.
Near-miss watch: The 7/23 XLE vol_ratio of 1.192 proves the vol_ratio gate is reachable — it's not structurally impossible. If today's volume is elevated as expected for an earnings-afterglow session, SPY vol_ratio could clear.
Sizing: 0.75% per instrument, 2.0% session budget total. If multiple instruments qualify, all above T=0.80 enter together.
Setup assessment: Trending session expected = good for SURGE's core thesis. However, today has catalyst-gap characteristics (AMZN-driven gap-up) — TEMPER learnings show SURGE is 0/2 on catalyst-driven gap sessions vs. 4/4 on clean intraday trend sessions. The ECI print at 8:30 determines whether the open establishes a clean trend or becomes a price-discovery process.
The path to a SURGE trade: (1) ECI print produces a clean initial direction (e.g., cool ECI → rates ease → gap-up holds), (2) first-hour signal_return ≥ 0.30%, (3) first hour establishes direction with ADX ≥ 22 and VWAP slope, (4) measured pullback to VWAP within 10:00–12:00 window, (5) pullback shows contracting volume, (6) resumption confirmation candle closes.
The path to no trade: (1) ECI print produces whipsaw (hot & cold components conflict), (2) gap-and-go structure with no pullback (AMZN surge sustainably pushes higher without testing VWAP), (3) AAPL drag pulls the open lower, creating a gap-down price-discovery session.
Temper learnings carry-forward: Per 2026-07-17 entry: event-day skip diagnostic resolved MIXED — QQQ_EARN days are SURGE's strongest event cohort (24t PF 2.61 dev, 3t PF 17.19 holdout). Today's AMZN catalyst has QQQ-earnings-adjacent character but is not an AMZN-earnings-report day (reported after close yesterday). Per 2026-07-23: on gap-down fundamental sessions, first-hour price action is price-discovery, not clean three-phase structure. Today's gap-up is different — but a gap-up with no clean pullback is also a no-trade scenario.
Sizing: 0.75% per trade, single position (co-fire limiter: stronger signal wins if both SPY and QQQ qualify).
No trades taken.
Desk: house-echo (Family 2 — Momentum / Trend Continuation) PM: ECHO Period covered: 2026-07-27 → 2026-07-31 (since the desk's last reflection on 2026-07-24) Members: ECHO (contra-trend, 15:30 ET), SURGE (with-trend, 10:00–12:00 ET)
| Member | Trades (period) | Net P&L (period) | Cumulative P&L | SPRT |
|---|---|---|---|---|
| ECHO | 0 | $0.00 | +$110.71 (3/3 W) | CONTINUE (+1.208) |
| SURGE | 0 | $0.00 | +$711.58 (4/6 W) | CONTINUE (+1.425) |
Combined desk P&L (period): 0 trades, $0.00. Cumulative desk record unchanged since 7/24: 9 trades, 7 W / 2 L, +$822.29.
| Date | Event | ECHO | SURGE | Notes |
|---|---|---|---|---|
| 07-27 | SPY -0.81%; ceasefire/oil-collapse gap-up faded; Durable Goods at 8:30 | $0 — vol_ratio blocked all three (XLE 1.124 closest, -0.076 to floor) | $0 — QQQ first-hour +0.13% below ±0.30% floor; SPY wrong VWAP side | Catalyst-gap day; both stacks' structural headwinds confirmed |
| 07-28 | SPY +0.22%; AI/semi rout day 2 (NVDA/OpenAI financing, China lithography), FOMC-eve | $0 — XBI vol_ratio 1.119 (-0.081); XLE blocked on VWAP persistence (price 57.70 vs VWAP 57.89) | $0 — QQQ trend qualified (signal -1.67%, ADX 36.4) but no VWAP-touch pullback resumed | QQQ's strongest trend of the period failed on structure, not signal |
| 07-29 | FOMC decision | Hard skip (event) | Hard skip (event) | Correct disposition — FOMC is the consistent-loser event type in SURGE's 2026-07-17 diagnostic |
| 07-30 | PCE; SPY +0.76% relief rally (MSFT/AMZN afterglow) | $0 — SPY vol_ratio 1.043 (-0.157) despite a +1.25% morning signal | Hard skip (event: PCE) | No desk plan filed for 7/30 (filing gap — ECHO was live and evaluated) |
| 07-31 | Two-tier: SPY +0.70% (fade-and-recovery, 1.33× ATR), QQQ +0.63% but -0.61% vs open; ECI slightly hot, Chicago PMI & Michigan beats; VIX 15.99 LOW | $0 — SPY morning signal +0.04% (below ±0.25%); XLE vol_ratio 0.567; XBI vol_ratio 0.983 (signal -2.53% — only qualifying signal of the day) | $0 — SPY first-hour -0.18%, QQQ +0.05%, both below ±0.30% floor | Deep dive below |
Neither member traded, and neither was blocked by an event skip — both were stopped at the first gate of their own stacks, which is exactly the right place for this session class to stop them.
The session was a two-tier fade-and-recovery: SPY closed +0.70% at 746.79 but the path was a gap-up open (744.68), a dip to 737.70 (-0.95% from open) through the 10:00–12:00 data-absorption window, then a full recovery into the close (1.33× ATR — a genuinely big, messy range day). QQQ closed +0.63% but -0.61% below its open at 687.89 — it faded its gap. The AMZN +15.3% / AAPL -7.35% divergence netted near-zero at the index level: enormous single-name programs that never became a broad-market directional program. ECI (+0.9% vs +0.8%, slightly hot but unchanged from Q1) produced a muted reaction; Chicago PMI 57.6 and Michigan Sentiment 55.2 both beat; VIX closed 15.99 (LOW).
ECHO — vol_ratio is the binding gate; the near-miss data says regime, not calibration. Across the period: XLE 1.124 (7/27, -0.076), XBI 1.119 (7/28, -0.081), SPY 1.043 (7/30, -0.157), XBI 0.983 (7/31, -0.217). The misses are scattered across three different instruments with no repeat offender and no consistent sub-threshold cluster — this is a low-participation, two-tier tape, not a drift against one threshold. The 7/23 XLE 1.192 remains the closest scrape in Phase 3, and nothing this period approached it. No fix needed; wait for the regime.
SURGE — the first-hour signal floor is the binding gate; regime, not calibration. QQQ +0.13% (7/27), SPY -0.22% (7/28), SPY -0.18% / QQQ +0.05% (7/31) — all below the ±0.30% floor, scattered across both instruments. The one strong-trend day (7/28 QQQ: signal -1.67%, ADX 36.4) failed on the pullback-resumption gate, not the signal floor — a one-way session with no VWAP-touch retracement, which is the strong-trend (ADX ≥ 30) failure mode the carried ADX re-tune watch already predicts. No threshold is being scraped repeatedly; no calibration conversation warranted.
Scored rows in intelligence/plan-calibration.csv for the period (7/27 and 7/31 only):
| Date | Stack | p_trade | Direction | Conviction | Actual | Brier |
|---|---|---|---|---|---|---|
| 07-27 | echo | 0.30 | long | low | no trade | 0.090 |
| 07-27 | surge | 0.25 | long | low | no trade | 0.0625 |
| 07-31 | echo | 0.65 | short | med | no trade | 0.4225 |
| 07-31 | surge | 0.30 | long | low | no trade | 0.090 |
Both 7/31 forecasts overshot, ECHO's badly. ECHO's 0.65 is the highest p_trade in its scored record, and it busted (Brier 0.4225). The plan's thesis — AMZN's +12.75% pre-market surge would elevate broad-market ETF volume and clear the vol_ratio gate — inverted in practice: the surge volume was single-name (AMZN/AAPL), and the two-tier structure suppressed index-level signals (SPY first-hour +0.04%) while sector-name vol_ratio was weak (XLE 0.567). The calibration lesson for the desk: on two-tier single-name-divergence days, index-level signal quality is suppressed — p_trade should be down, not up. SURGE's 0.30/low on the same session was the correctly-calibrated side of the desk.
Operational notes for the record: the 7/28 and 7/29 forecasts were filed but have no rows in plan-calibration.csv (scoring gap), and no desk plan was filed for 7/30 (PCE day — SURGE skipped by event, but ECHO was live and evaluated gates). Both are data-hygiene gaps, not strategy findings.
No qualifying setup for either member across the period — $0.00 for the desk.
Net paper P&L (period): $0.00 Running paper P&L (cumulative): +$822.29 (9 trades, 7 W / 2 L)
7/31 (the live session): the morning plan rated event risk LOW with no Tier-1 events — correct. All three mid-tier prints were benign: ECI +0.9% vs +0.8% (slightly hot, unchanged from Q1, muted reaction), Chicago PMI 57.6 vs 56.0 (beat), Michigan Sentiment 55.2 vs 54.0 flash (beat). The EOD briefing graded the morning narrative PARTIAL: event risk and driver (AMZN/AAPL divergence) STUCK, but "trending" overstated the session's cleanliness — it was a volatile range day with fade-and-recovery. That overstatement is precisely why both stacks' first-hour gates blocked: the session closed green but never established a clean directional program.
Skipped sessions: 7/29 FOMC (both stacks — correct; FOMC is the consistent-loser event type in SURGE's diagnostic) and 7/30 PCE (SURGE hard skip; ECHO evaluated and was vol_ratio-blocked at 1.043 despite a +1.25% SPY morning signal). The 7/27 and 7/28 HIGH event-risk calls (FOMC week overhead) were directionally right; the 7/31 LOW call was right.
Next Tier-1: 2026-08-07 NFP (7 sessions ahead). Next session (Mon 8/3) is ISM Manufacturing PMI — mid-tier, no skip for either stack.