Reversal Desk — 2026-09-16

Desk Plan — Reversal — 2026-09-16

Briefing reference: /Users/dadbot/Desktop/ClaudeBod/projects/dadbrain/Analysis/briefings/2026-09-16-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 — and this time it is realized in-session, not tomorrow. The FOMC decision + SEP/dot plot + Warsh presser lands TODAY, Wed Sep 16, 2:00 PM ET (decision/SEP) and 2:30 PM ET (presser) — the first rate move of 2026 and the first under Chair Warsh, ~88–92% priced for a 25bp hike to 3.75–4.00%. Firm-wide status is NO-GO (FOMC), and for this desk that is not an advisory: both active members' Tier-1 hard-skip gates fire deterministically today (Null's is_event_day(trade_date, tier_filter=2), code-enforced 2026-07-14; Meridian's is_event_day() return-before-fetch, the 2026-07-14 fix for the gate that had been log-only). Both members log no_trade:event_day:FOMC_tier1 before any gap or level is even measured. This is a zero-trade session for the desk by rule, not by judgment.
• Session character expected: repricing-to-conclusion held in suspension — a contained, low-conviction morning and mid-day, then a volatility expansion on the 2:00 PM statement and a second on the 2:30 presser. The briefing is explicit: expect most desks to get nothing tradeable before 2:00 PM, and do not expect structure to form cleanly until the dot plot is on the tape. Direction is undefined until 2:00 PM — both the hawkish-dot-plot tail (renewed tech/rates downside, with the 10-yr at ~4.99–5.00%, a 19-year high) and the dovish-relative surprise tail (a squeeze that relieves the semi complex) are live and roughly symmetric. I refuse to pre-assign a side; that is the briefing's own instruction and the sixth consecutive session's discipline.
• VIX regime: 16.97 (LOW, <18) — down −0.23 from the prior close (17.20). Vol compressed into a same-day binary. LOW-VIX is historically both members' strongest cohort, and today it is irrelevant to the forecast: the deterministic event gate closes the session before any cohort logic runs. The vol move itself is orderly, not panicked.
• Key levels:
• QQQ: prior close 704.60 (Null's gap-fill target), prior-day range 703.63–709.52, 20-day SMA 713.34 (below SMA20 — QQQ has closed sub-SMA for a second straight session), ATR(14) 7.70. Null's in-band gap band: ±0.77 to ±5.39 pts. Meridian's sweep level (prior high) 709.52.
• SPY: prior close 757.42, prior-day range 756.18–760.32, 20-day SMA 765.30 (below), ATR(14) 5.82. Meridian's sweep level (prior high) 760.32.
• Pre-market: no QQQ/SPY print is available (thin IEX at 07:00 ET) — the implied futures read is modestly higher, tech-led (Dow ~+0.1%, S&P ~+0.2%, NDX ~+0.4%), semis tentatively bid. The briefing flags the direction, not the magnitude, as the reliable read; confirm the actual gap at the cash open. Moot for both members today.
• Macro backdrop: a stagflation-tilted, rates-gated week into a hawkish regime shift — oil above $100 (Hormuz bypass still shut, Yanbu loadings suspended), the 10-yr parked at a 19-year high, a BOJ tightening expected and a failed Senate Clarity Act cloture vote taking one marginal risk-on buyer off the board. The market's question has hardened to "can the AI complex profit, at what discount rate?" — and the Fed raises that discount rate on purpose at 2:00 PM. One sentence: a firm-wide NO-GO whose binding constraint is a same-day binary that lands hours after both of our windows close — the honest desk call is zero trades, deterministic, and both mechanisms being structurally event-clean is the reason it costs us nothing.

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

Setup present: NO TRADE — deterministic, code-enforced Tier-1 FOMC hard skip. p_trade 0.0, direction none, conviction high. Null's event gate is checked before the daily-bar fetch and before any gap is measured: is_event_day(trade_date, tier_filter=2) returns FOMC Tier-1, the reason string logs event_day:FOMC_tier1, and the script returns. There is no gap to argue about, no band to graze, and no tail to price — the sequencing rule (2026-06-18 TEMPER observation: the event filter must run ahead of gap measurement, because a small gap that passes size and vol filters can still be fundamental repricing) is enforced in code as of the 2026-07-14 v2-track resolution.

This is exactly why the skip is cheap for this stack and not a missed opportunity:

1. The tape today is the population v2 exists to exclude. A same-day FOMC decision with a live dot-plot fork is print-driven fundamental repricing — the same category as FOMC/CPI/NFP/PCE mega-cap earnings, which is the whole point of the v2 filter. Layered on a 19-year-high 10-yr and $100+ oil, the day is structurally outside the "thin-overnight manufactured imbalance" thesis by construction, not by size.
2. The gap, whatever it prints, is not a fade candidate. Even absent the gate: an event-gated gap before the event is a coin-flip, not an edge (the briefing's own Reversal line — fading an event-gated gap before the event is a coin-flip). A gap that is oversize fails the 0.70× ceiling; a sub-floor open fails the 0.10× floor; and if the gap lands in-band the direction resolves only after 2:00 PM, hours past Null's 10:30 entry cutoff and 11:00 hard exit. The 09-11 CPI-day measurement is the precedent: the cash-open gap printed ~0.87× ATR — outside the band — so the deterministic skip cost that session nothing.
3. Shape context (informational, carried): the recent stretch shows this tape printing both extremes — oversize gaps at the ceiling (09-10 1.149×, 09-14 1.492× — both clean regime blocks, 64% and 113% over the line) and a sub-floor flat open (09-15 0.053×). Per the 09-11 corrected narrative I weight the ceiling side evenly with the sub-floor side; neither extreme is a grazed threshold, and neither would have been tradeable today either way.

Sizing vs event risk: not applicable — no setup can be admitted. Standard if it ever were: 1 MNQ, 0.5% risk, one position at a time, 11:00 AM ET hard exit. No position is carried into the 2:00 PM print under any circumstance (a 3.5-hour buffer between the 11:00 exit and the decision).

Key levels: QQQ prior close 704.60 (fill target), prior high 709.52, prior low 703.63, SMA20 713.34, ATR(14) 7.70; in-band gap ±0.77 to ±5.39 pts. Listed for the record only — none of them is reached today.

Invalidation / stand-down: the entire session. No trade condition: no_trade:event_day:FOMC_tier1.

Mechanism health: SPRT CONTINUE — n=14 W13/L1, LLR +2.249, cumulative +$1,881.59, WR 92.9%. First-loss watch stays N=1 (the 09-03 trend-day fade remains un-clustered). Today adds a fourth consecutive no_trade row without touching the SPRT population — a skip day is not an observation.

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

Setup present: NO TRADE — deterministic, code-enforced Tier-1 FOMC hard skip. p_trade 0.0, direction none, conviction high. Meridian's is_event_day(trade_date) check returns before the per-symbol data fetch (the 2026-07-14 fix that converted the firm-wide FOMC/CPI/NFP/PCE skip from log-only to binding), so today no leg is even evaluated — no volatility check, no ADX precheck, no prior-high mark. The log row is no_trade:event_day:FOMC_tier1.

Note the distinction the Scanner watchlist draws and that I carry into this plan: the 2:00 PM SEP/dot plot is not itself a Tier-1 data release — the skip fires because FOMC decision day is a hard skip regardless. I do not read that as license; the day's tape is a binary repricing into a 2:00 PM print, and the rule exists precisely so that the question is never argued at 09:30.

Why the skip costs this stack nothing, structurally:

1. The mechanism needs an exhausted UP-push into a prior high, and the day's character points away from those levels. From the futures read the tape is modestly higher, tech-led — but the day's actual direction is undefined until 2:00 PM, and per the 2026-06-18 lesson in its strongest form, reachable-in-principle is not likely-to-be-reached: a bearish fade requires price to push up into the prior high (QQQ 709.52 / SPY 760.32) and spike-reject, and a coin-flip pre-decision tape gives no directional context for that push to form and hold before the event.
2. The sweep-completion gate, not the ADX band, is what binds in this regime (confirmed 09-09/09-10/09-14, and the 09-15 session added a third distinct blocker: the volatility gate, prior_range=10.21 vs 1.25x_avg ceiling 6.51 on QQQ and 5.59 vs 4.30 on SPY, IWM on an ADX precheck 15.8 below the 18 floor). Three different binding gates across three sessions is regime texture, not a threshold claim. Even if the Kill Zone were open, the 2:00 PM binary would be a classic way to blow the ADX 32 ceiling or produce a one-way drift away from the level.
3. Event-clean by construction, like the whole desk. Meridian's 11:30 AM ET hard close (12:00 Kill Zone end, 12:30 absolute flat) sits two and a half hours before the statement. Even a hypothetical position taken this morning would be flat long before the binary — the skip is a rule about quality of tape, and the flat-before-the-print property is a separate, structural protection that holds regardless.

Sizing vs event risk: not applicable — no leg is evaluated. Standard if it ever were: $250/leg, caps 10 MNQ / 5 MES / 5 M2K, 2-loss daily cap, hard close 11:30 ET. iwm_leg remains n=0 (no qualifying setup since the 2026-08-20 expansion; honest live anchor near ~56%, not the 73.3% headline) — not a today driver.

Key levels: QQQ prior high 709.52, SPY prior high 760.32 (sweep thresholds, unmarked today); ADX band 18–32; prior ranges QQQ 5.89 / SPY 4.14 against their 1.25× ceilings. Listed for the record only.

Invalidation / stand-down: the entire session. No trade condition: no_trade:event_day:FOMC_tier1.

Mechanism health: SPRT CONTINUE — n=3 W2/L1, LLR +0.222, cumulative +$385.22, WR 66.7%; spy_leg n=1 WR 100%; iwm_leg n=0. No boundary risk.

Plan Filed

• Filed: 2026-09-16 (pre-market, ~04:35 PT / 07:35 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 NO-GO (FOMC); event risk HIGH with the decision + SEP + presser inside today's session (2:00 / 2:30 PM ET). Both active members' hard event gates fire deterministically — Null no_trade:event_day:FOMC_tier1 (Tier-2 filter, tier_filter=2), Meridian no_trade:event_day:FOMC_tier1 (return-before-fetch). Null p_trade 0.0 / direction none / conviction high; Meridian p_trade 0.0 / direction none / conviction high. This is a deterministic stand-down, not a low-conviction lean: no gate is grazed, no threshold is in question, and no tail is being priced — the event rule closes both windows before measurement. Both stacks are structurally event-clean (Null's 11:00 hard exit and Meridian's 11:30 hard close both fall 2.5–3.5 hours before the print), so today's skip removes no tradeable exposure that the mechanisms would have wanted.
• 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).
• Carried watch (nothing fires today): Meridian needs a tape that pushes a leg into its prior high with RSI(2) exhaustion and a bearish FVG — a drift-away or event-gated session produces structural absence, as the last three sessions each did via a different gate. Null's suppression is currently two-sided (oversize gaps at the ceiling, sub-floor flat opens); both are regime, neither is calibration.

No trades taken.