Reversal Desk — 2026-09-15

Desk Plan — Reversal — 2026-09-15

Briefing reference: /Users/dadbot/Desktop/ClaudeBod/projects/dadbrain/Analysis/briefings/2026-09-15-pm.md PM: Wicker (persona; own stack retired 2026-07-14 — this plan covers active members Null + Meridian only)

Shared Market Read

• Event risk today: HIGH — the FOMC decision + SEP + dot plot + Warsh presser is TOMORROW (Wed Sep 16, 2:00 PM ET), at a 1-day lead. The two-day meeting begins today but has no release today — so neither active member's hard event gate (FOMC/CPI/NFP/PCE) fires today, the same distinction that reopened both windows on 09-14. Firm-wide status is GO. But the Step-4 rule flags Tier-1 within ~2 days, and a decision+presser at a 1-day lead is literal-HIGH. The binding constraint is the FOMC: direction into Wednesday is capped — this is a positioning day, not a repricing-to-conclusion day.
• Session character expected: two-tier de-risk, fundamental-repricing gap. Overnight the AI trade unwinds on two blades at once: the weekend AI-safety repricing (Amodei's "We Must Pace the Frontier," Altman concurring) now compounding with a fresh semiconductor catalyst (Korea memory sell-off spilling into US names — SK Hynix −14%, Samsung −13% — plus The Information reporting a Chinese state-backed firm began mass-producing a key chipmaking tool, hitting ASML directly). Premarket laggards: GLW −9.3%, LRCX −9.0%, HPE −8.9%, MRVL −7.8%, ASML −7.3%, AMD −7%, MU −5.3%, SNDK −13%. Meanwhile energy gaps up (Brent $108+, Hormuz bypass shut, talks postponed) and cyber/software rips on the AI-safety rotation (CRWD +6.1%, PANW +5.1%, ZS +4.6%). The 10-yr broke above 5% (highest since Jul 2007) — the discount-rate blade is live. This is the same fundamental-repricing family both Family 1 mechanisms exist to exclude, now with a named-catalyst semi leg. Futures point modestly lower, Dow-led (S&P −0.45%, Nasdaq −0.42%, Dow −0.61%).
• VIX regime: 17.24 (LOW, <18) — up +0.14, still under the line. LOW-VIX is both members' strongest historical cohort, but it does not rescue the setup: the tape is the excluded fundamental-repricing population, and LOW VIX keeps ADX-gated trend edges suppressed so neither member gets a clean directional impulse to fade. The vol move is orderly, not a panic.
• Key levels (authoritative = pm briefing, Alpaca/IEX):
• QQQ: prior close 709.24 (Null's fill target), prior-day range 702.81–712.88, 20-day SMA 714.61 (below SMA20 — third straight sub-SMA session), ATR(14) 7.82. Null's implied gap band: 0.1× ≈ ±0.78, 0.7× ≈ ±5.47 pts. Meridian's sweep level (prior high) 712.88.
• SPY: prior close 760.75, prior-day range 757.97–763.48, 20-day SMA 766.06 (below SMA20), ATR(14) 5.76. Null band ±0.58 to ±4.03 pts. Meridian's sweep level (prior high) 763.48.
• Pre-market implied QQQ gap is source-conflicted (Investing ~−1.8% ≈ −12.8 pts ≈ 1.63× ATR vs futures ~−0.4% ≈ −3 pts ≈ 0.4× ATR). The briefing's reconciliation says direction is what matters, the exact magnitude is confirmed at the cash open. Either way the direction is DOWN — which is what gates both members.
• Macro backdrop: a stagflation-tilted, FOMC-gated week — the AI-capex-return question hardened from "will they spend?" to "can they profit, and at what discount rate?", with the 10-yr breaking 5% supplying the rate blade while the AI-safety + fresh semi catalyst attack the cash-flow blade. One sentence: a GO session on the calendar whose tape is the fundamental-repricing two-tier gap both reversal mechanisms exist to avoid, capped by a 1-day-away FOMC binary — the honest desk call is near-zero expected trades.

Null — QQQ (gap fade / FVG; MNQ primary)

Setup present: NO expected trade — the trend-alignment gate is the binding block, and today's gap-down tech leg is the structurally wrong way. p_trade 0.15, direction none, conviction med. The event gate is CLEAR (FOMC decision is tomorrow, not today), so gap measurement reopens. But the expected tape is a gap-DOWN tech leg on a sub-SMA tape, and Null's HTF filter excludes exactly that:

1. Trend alignment (the binding gate). QQQ is below its 20-day SMA (709.24 vs 714.61) → the rule allows gap-UP fades (short) only, no gap-down fades. The expected tape is a gap-DOWN tech leg. A gap-down on a sub-SMA tape is structurally untradeable for Null regardless of size — the counter-trend long is excluded by the HTF filter. For Null to fire, QQQ would have to gap up in-band, which a named-catalyst semi-unwind morning makes unlikely.
2. Gap-size ceiling. Even under the friendlier futures read (~0.4× ATR), the direction is the problem — and under the Investing read (~1.63× ATR) it blows the 0.7× ceiling outright. 09-01 (0.933×), 09-10 (1.149×), 09-14 (1.492×) confirm this tape prints oversize gaps, not the sub-0.7× manufactured-imbalance shape the edge needs.
3. Fundamental-repricing exclusion. The v2 thesis is built on rebalancing thin-overnight manufactured imbalances. Today's gap is a named-catalyst fundamental repricing (Korea memory sell-off + Chinese DUV mass-production + AI-safety) layered on a 5% 10-yr — the breakaway population the thesis explicitly excludes, the same class as a print gap, just unscheduled. The 2026-06-18 TEMPER finding applies: a gap that passes size/vol can still be fundamental repricing whose fill direction is not guaranteed. The briefing's conditional is explicit: fade fuel exists only if the semi complex stabilizes — and it is re-cascading on a fresh catalyst, so that condition is closed.

I forecast 0.15 (not 0.0) because the event gate is not closed today and the two-tier surface (cyber/software ripping, energy up) carries a genuine tail where QQQ gaps up modestly in-band on software-led relative strength and the trend filter admits a short fade. That tail is small; the base case is no-trade, $0.

Sizing vs event risk: if the setup were to appear it would be standard — 1 MNQ, 0.5% risk, one position, 11:00 hard exit. No size increase into a FOMC-eve week, no position held into Wednesday. Expected $0.00.

Key levels: QQQ prior close 709.24 (fill target), prior high 712.88, prior low 702.81, SMA20 714.61. In-band gap ±0.78–5.47 pts. None reached in the base case.

Invalidation / stand-down: the session never produces an in-band gap-up fade on a sub-SMA tape. If QQQ gaps down (the expectation), or the gap is oversize, or the semi complex extends rather than stabilizes, Null stands down by rule. No trade condition: no in-band gap-up fade forms by the 10:30 entry cutoff.

Mechanism health: SPRT CONTINUE — n=14 W13/L1, LLR +2.249 (WR 92.9%, cumulative +$1,881.59 through the 09-14 no_trade; the single 09-03 loss un-clustered). This tape keeps producing oversize gaps, not the small in-band shape the edge needs — weight the ceiling side evenly with the sub-floor side, per the 09-11 corrected narrative.

Meridian — QQQ/SPY/IWM (kill-zone prior-high sweep + RSI(2) ≥ 88, bearish only)

Setup present: NO expected trade — the sweep-completion gate is the block, and a down-gapping repricing tape is the wrong structural backdrop for an upside sweep. p_trade 0.20, direction short (bearish-only, structural), conviction med. The event gate is CLEAR (FOMC tomorrow, not today), so the Kill Zone (9:30–12:00, hard close 11:30) reopens for signal work — genuinely a live session. But the mechanism needs an exhausted UP-push into a prior-session high (RSI(2) ≥ 88 inside an 18–32 ADX band), and the expected tape points away from those levels.

• Sweep completion is the binding gate (the 09-10 reflection's confirmed finding: in-band ADX produced no_sweep because the tape drifted away from the prior highs; 09-14 same shape — QQQ +7.05, SPY +4.17 gap-to-level, no_sweep). Today's character is a fundamental-repricing gap-DOWN tech leg — QQQ/SPY open below their prior highs (712.88 / 763.48) and are more likely to bleed away than rally up into them.
• The 2026-06-18 lesson, strongest form: reachable-in-principle is not likely-to-be-reached. The prior highs sit above the open on a gapping-down tape. Meridian's bearish fade needs price to first push UP into the level and spike-reject; a down-drifting, repricing tape is the wrong structural backdrop regardless of how close the level looks.
• The two-tier tail: the cyber/software rotation and energy bid could produce an intraday QQQ/SPY bounce. If the AI complex stabilizes after the initial overshoot and software-led strength lifts tech, a rebound toward the prior highs with RSI(2) exhaustion is the only path to a qualifying setup. That is why p_trade is 0.20, not 0.0 — the two-tier surface carries a real (if conditional) bounce-to-prior-high tail, the exact exhaustible-push shape this mechanism trades. But the briefing warns the fade is conditional on stabilization vs. extension — a durable semi re-cascade excludes even the fade.
• ADX band: QQQ's recent ADX has straddled the band (09-09 30.0 in-band; SPY 45.6 / IWM 38.7 over the ceiling). A gap-down volatility spike is a classic way to blow the 32 ceiling. Either the band blocks, or the tape gives no exhaustion push. Both outcomes are no-trade.
• Volatility gate: prior-day ranges (QQQ 10.07, SPY 5.51) are checked against 1.25× their 20-day averages at open — a possible early per-instrument block on a volatile prior session, another path to no-trade. Not flagged on 09-14 (no_sweep, not volatility), so likely clears, but worth noting.

I forecast 0.20 (not 0.0) because the Kill Zone is genuinely open and the two-tier surface carries a real intraday-bounce-to-prior-high tail. The base case is the drift-away no_sweep, $0.

Sizing vs event risk: if the setup fires it would be standard — $250/leg, cap 10 MNQ / 5 MES / 5 M2K, 2-loss daily cap, hard close 11:30 ET. No size change into the FOMC eve; all positions flat by 11:30/12:30, none carried into Wednesday. iwm_leg n=0 (thin; anchored near ~56%, not the 73.3% headline) — not a today driver. Expected $0.00.

Key levels: QQQ prior high 712.88, SPY prior high 763.48 (sweep thresholds); IWM per the prior leg log. ADX band 18–32. Carried live watch: a tape that pushes QQQ into the prior-high zone with exhaustion — but on a gap-down repricing day that push must form intraday, which the base case does not deliver.

Invalidation / stand-down: no qualifying sweep — the tape either drifts down/away from the prior highs (base case) or blows the ADX ceiling on the gap-down spike, or the volatility gate blocks a leg. No trade condition: no leg completes an RSI(2) ≥ 88 spike-and-reject into its prior high with a bearish FVG within the Kill Zone.

Plan Filed

• Filed: 2026-09-15 (pre-market, ~04:33 ET)
• Frontmatter forecasts complete for active members only (Null, Meridian): yes. Wicker (PM): strategy retired 2026-07-14 (WS2c Tier-2 NO-GO) — no plan, no forecast, no per-member file; persona continues desk oversight only.
• Firm-wide GO (no same-day Tier-1), event risk HIGH (FOMC decision+SEP+presser Wed Sep 16, 1 day out). Neither active member's hard event gate fires today — both Kill Zone/gap measurement windows reopen. But the tape is the fundamental-repricing two-tier gap both Family 1 mechanisms exist to avoid, capped by the FOMC one day out: Null p_trade 0.15 / direction none / conviction med (binding gate = trend alignment: sub-SMA tape admits only gap-UP fades, and the expected tech leg gaps DOWN on a named semi catalyst — also over the 0.7× ceiling on the deeper read, and excluded as breakaway fundamental repricing); Meridian p_trade 0.20 / direction short / conviction med (binding gate = sweep completion: a down-gapping, repricing tape drifts away from the prior highs, so the RSI(2)≥88 up-push-into-prior-high the mechanism requires does not form — a two-tier intraday bounce into prior highs is the only, conditional, tail). This is a conditional low-probability call, not a deterministic stand-down — both are forecast low-but-nonzero because the event gate is open and the two-tier surface carries genuine tails. Direction-refusal discipline holds for Null (none); Meridian's bearish-only lean is structural, not a call.
• Reconfirmation ledger: no open item gates today. Meridian IWM expansion CLOSED 2026-08-20 (strategy_v7); standing ~6-month decay re-backtest next due ~2027-02-20. Wicker's feed/session re-run remains the desk's highest-priority open reconfirmation (unchanged, not a today blocker, and not this desk's to act on).

No trades taken.