House Mneme — 2026-09-15

desk: house-mneme date: 2026-09-15 forecasts: mneme: p_trade: 0.15 direction: long conviction: low trace: p_trade: 0.10 direction: long conviction: low

Desk Plan — House Mneme — 2026-09-15

Briefing reference: /Users/dadbot/Desktop/ClaudeBod/projects/dadbrain/Analysis/briefings/2026-09-15-pm.md PM: MNEME (desk-blind: this desk's own members only; GHOST stays frozen, no plan)

Shared Market Read

• Event risk today: HIGH — but it is the literal-HIGH next-day rule, not my own calendar. There is no Tier-1 release today (Empire State + ADP weekly are LOW/MEDIUM; the FOMC two-day meeting merely begins today with no release). The FOMC decision + SEP + dot plot + Warsh presser is TOMORROW, Wed Sep 16, 2:00 PM ET, ~90–92% priced for a 25bp hike (first since Jul 2023) — the dot plot and presser are genuinely two-sided (one-and-done vs. cycle-start). The 1-day lead on a decision (not minutes) is what earns the literal-HIGH rating. Neither of this desk's stacks hard-skips today — the hard-skip set is FOMC/CPI/NFP/PCE, and FOMC is tomorrow, not today. Firm-wide status: GO, but direction is capped.
• Session character expected: Two-tier de-risk / gap-down tech-led, direction NOT pre-assignable. The overnight delivers three live legs: (1) tech/semis gapping down on named catalysts — memory/storage complex down (SK Hynix −14%, Samsung −13%; US: SNDK −13%, GLW −9.3%, LRCX −9%, HPE −8.9%, MRVL −7.8%, ASML −7.3%, AMD −7%, MU −5.3%) on the deepening AI-safety shock + The Information's report of Chinese mass-production of DUV litho hitting ASML + a fresh Korea memory sell-off spilling in; (2) energy gapping up on the Hormuz/Salalah supply shock (Brent $108.05 +2.35%, WTI $104.04 +2.61%, second consecutive up session, Saudi East-West pipeline still shut); (3) cyber/software ripping on the AI-safety rotation (CRWD +6.1%, PANW +5.1%, ZS +4.6%). The 10-yr broke above 5.00% (highest since Jul 2007). This is the same two-tier shape the last two sessions traded — a fundamental-repricing, FOMC-gated tape, not a clean trend impulse. The briefing's directive: trade the gap's character, not a one-day direction.
• VIX regime: 17.24 — LOW (<18) (+0.14 vs prior close 17.10). LOW VIX is each stack's best-backtested regime (MNEME VIX<20 PF 2.39; TRACE VIX<20 PF 2.46), but it is not a reason to expect a fire — it is the regime where both long-only KNN continuations have been correctly declining on contained/two-tier character. Watch a spike through ~18 (one-way-tape alarm). Neither stack gates on VIX (it is not an input); the binding constraint today is session character, not vol.
• Key levels: SPY prior close 760.75, prior range 757.97–763.48, SMA20 766.06 (SPY closed a fifth session at/under its 20-day — sub-SMA20 confirmed), ATR(14) 5.76. QQQ prior close 709.24, prior range 702.81–712.88, SMA20 714.61 (sub-SMA20; semis in bear territory, >20% off highs), ATR(14) 7.82. DIA is not in the briefing price block — confirm the gap at the open for framing only; TRACE's own data path and KNN gates govern. A >1.5% open gap either direction is both stacks' standing "pattern-library-thin-at-index-extremes" invalidation.
• Macro backdrop: Stagflation-tilted week at a new intensity — oil above $107 with the Hormuz bypass shut and Iran/Gulf talks postponed, the 10-yr at a 19-year high above 5%, and the AI capex boom under simultaneous safety/regulatory + profitability + discount-rate scrutiny ("can they profit, at what discount rate?"). The FOMC one day out caps directional conviction — today is a positioning day, not a repricing-to-conclusion day.
• Desk-wide stance — LIVE but character-cautious; no hard skip, no forced fire; both stacks long-only continuation. This is the family's structurally challenging class: an FOMC-eve two-tier de-risk with the tech leg gapping down on named catalysts. For a long-only KNN broad-index continuation, a gap-DOWN tech-led open means SPY/DIA open lower and must stabilize and form a clean up-pattern to give my library something to continue — and on a repricing gap (Catalyst-Gap Containment, not a trend impulse) with direction capped, that is precisely the class that has been declining both stacks. The honest base case is another stand-down on sub-bar votes, mirroring yesterday (09-14: MNEME A 47%/B 53%/C 53%; TRACE A 50%/B 20%/C 60% — all sub-bar) and 09-10. Forecasts are modest and honest (MNEME 0.15, TRACE 0.10, both long, low conviction). GHOST stays frozen (Phase 2, Decision D2) — no plan, no forecast.

MNEME — SPY KNN (p_trade 0.15, long, low)

Setup present: possible, not likely — a live day on the two-tier de-risk character that keeps declining me, compounded by a gap-down tech-led open and my best window being compromised. No hard skip (no Tier-1 today; FOMC is tomorrow). My mechanism is healthy — SPRT CONSISTENT-WITH-BACKTEST (23 trades, 47.8% WR, +$1,041.36, LLR +3.793). But the character is the class I structurally struggle with, and today it is the sharpest version of it:

• The gap-down tech-led open is the core handicap. My most frequent firing window is Window A (observe 9:30–10:00, trade 10:00–12:00 — 08-10, 08-13, 08-17, 08-21, 08-24, 08-31, 09-08 were all A). A gap-DOWN tech-led open means the first 30 minutes are the repricing-noise window: the semis are gapping down on named catalysts, and SPY opens modestly lower (futures S&P ~−0.45%, below the ±0.5% broad-market threshold). A long-only continuation needs SPY to stabilize off the prior-range low (757.97) and build an up-pattern through the morning. On an FOMC-eve de-risk day, the honest read is that a clean, library-recognizable continuation is unlikely — the briefing is explicit that a repricing gap is not a clean single-directional impulse.
• The realistic (still low-probability) fire path is the afternoon, conditional on stabilization. If the semis' gap stabilizes intraday (the briefing's stated relief condition is "the semis stabilize") and SPY reclaims structure, a Window B (observe 10:00–12:00, trade 14:00–16:00) or Window C continuation becomes possible. My flagged Window B two-tier laggard-leg whipsaw stays at N=2 (07-24, 08-27) — a third same-shaped Window B stop-out on this two-tier tape clears N=3 and re-tests the session-character framing. Window C is flagged (holdout PF 1.25); rolling-15 PF <1.2 monitor open. Vote gate (≥62%) and max_sim (≥0.50) remain the binding arbiter; my read is a two-tier FOMC-eve day lands sub-bar.
• Sizing vs. event risk: Standard — 0.75% per trade ($187.50 on $25K), max $75K notional if it fires. Event risk HIGH does not size me down: the hard-skip set is calendar-anchored and FOMC is tomorrow, not today — a live-day, not a skip. The KNN gates are not sizing inputs. No defensive sizing, no half-size.
• Session-character fit: poor. Two-tier de-risk gap-resolution is the wrong-character class for a long-only broad-index continuation, and today it carries the added handicap that Window A — my best window — observes the gap-down repricing noise, so the window most likely to fire me is the one with the poorest library-match quality today.
• Key levels: SPY prior close 760.75, prior range 757.97–763.48, SMA20 766.06 (fifth session under — the value-led break is confirmed). Whether SPY holds the prior-range low 757.97 off the gap and works back toward the 20-day, or slices through on a one-way risk-off break, tells me whether any library-recognizable continuation can form. A break below 757.97 toward a one-way value-led/risk-off move = no clean long continuation for my library to find. ATR(14) 5.76.
• Invalidation — absolute stand-down if: (1) SPY gaps >1.5% at the open either direction — pattern-library-thin-at-extremes; (2) the cost stack accelerates into a one-way risk-off break — VIX spiking through ~18 and SPY slicing below 757.97 with no reclaim; on such a tape a long-only KNN has nothing clean to continue and any match is spurious. Futures are modestly lower (~−0.45%), so (1) is not expected, but the open is the first read of whether the semi gap holds or fades.
• What I'm watching: (1) Do the semis stabilize intraday? The briefing names this as the relief condition for the tech leg — a stabilization re-opens the possibility of a broad-index continuation; a durable re-cascade excludes even a fade fuel and kills the long-continuation path. (2) Whether SPY holds 757.97 off the gap and works back toward the 20-day — continuation-versus-break governs my only realistic fire path (afternoon). (3) Whether Window A prints a sub-bar vote on the gap-down repricing noise (expected) and whether B/C build clean structure after the semis' stabilization-or-cascade resolves.

TRACE — DIA KNN (p_trade 0.10, long, low)

Setup present: possible but unlikely — live with full data flow for the first time in weeks, but on the instrument (DIA = the Dow = the two-tier day's fragile value/consumer leg) today's character is against me, with the 10-yr at a 19-year high. No hard skip (no Tier-1 today; GO). The 09-05 fetch_today_sip_embargo fix is now 3-for-3 live (full three-window reads on 09-08, 09-09, 09-10 with real past samples — past_n=1728), so the observability constraint that capped me for seven weeks is cleared and my gates can actually be reached. But the character is the read that matters:

• The Dow is the two-tier day's fragile leg, and today the 10-yr is the sharpest it has been in 19 years. DIA = 30 blue chips with a consumer/industrial tilt and only XOM/CVX as the energy counterweight — but today energy is actually up (Brent $108, XOM/CVX bid), which is the one leg that helps the Dow. The offsetting problem is the 10-yr above 5% — the yield-sensitives in the Dow (homebuilders, banks, consumer) get hit on a 19-year-high discount rate, and the semis' gap-down is a tech/duration move that pressures the whole tape. On 09-09 my gates fired at high strength (Window C, vote 70.0%, max_sim 0.7222) and the trade still stopped out (−$12.78, hard-flat @ 524.76) on the fragile-leg two-tier tape. Today's character is the same class, one day closer to FOMC, with the 10-yr above 5% — a long-only Dow continuation is structurally challenged.
• The DEGRADED clock is the live watch and is NOT a sizing or force/suppress input. LLR is −2.057, now only −0.888 from the −2.944 DEGRADED boundary after the 09-09 qualifying C-fire stop-out; the pre-registered desk-PM-review-within-1-session / TEMPER-within-5 protocol is now near-activated. A legitimate fire is neither forced (to test the mechanism) nor suppressed (to avoid the DEGRADED outcome) — I forecast honestly and let the gates decide. Standard 0.75% sizing if it fires; event risk HIGH is a live-day, not a skip.
• Session-character fit: poor. The Dow on an FOMC-eve two-tier de-risk with the 10-yr above 5% is the structurally unfavorable read for a long-only Dow continuation (confirmed on 09-09: a 70%/0.7222 C-fire still lost). A second consecutive C-fire into the same fragile-leg character would be the desk's live question, not a signal-quality one.
• Key levels: DIA is not in the briefing price block — confirm the gap at the open for framing only. My recent Phase 3 band is ~$515–535 (last two real C-fires: 08-27, $535.71 scratch; 09-09, $524.85 loss). Whether the Dow holds its opening dip or fades further off the 10-yr/5% pressure tells me whether today would have been a library-recognizable continuation day.
• Invalidation: (1) All-window no_data on a normal (non-starvation) session — a definitive escalation datapoint that the 09-05 fix did not hold; after 3 clean sessions this is now the low-probability branch, but still the strongest stand-down reason if seen. (2) DIA gaps >1.5% either direction — pattern-library-thin-at-extremes. (3) A one-way risk-off break on the Dow's fragile leg into the close.
• What I'm watching: (1) Whether the Dow's energy members (XOM/CVX) offset the fragile consumer/industrial leg under a 5% 10-yr — DIA's two-sided internal character on a day when one leg (energy) is bid and the other (yield-sensitives) is pressed. (2) Whether any Window-C vote/sim reading lands on clean continuation bars or on 10-yr/5%-gap and semi-cascade noise — a news-driven session's matches are diagnostic of the library, not evidence for the 62%-cluster thesis. (3) A fourth consecutive full three-window read — observability strengthening toward formally closing the ASSAY hand-off.

Pooled note (Family 5)

Both stacks are LIVE today (no hard skip; GO; HIGH event risk is a next-day-FOMC framing, not a today skip) — so a co-fire is structurally possible (both long-only). Per pooled validation, a same-day trade across MNEME and TRACE counts as one clustered observation for significance. Both stacks' binding gates are the KNN vote threshold on whatever data arrives; neither stack's forecast here touches execution (plans are scored, never binding). SPRT monitors stay as filed (MNEME CONSISTENT-WITH-BACKTEST LLR +3.793; TRACE CONTINUE LLR −2.057, −0.888 from DEGRADED) and only move if a trade logs. The two members are separately reasoned above: MNEME's handicap is the gap-down tech-led open compromising its best window (A); TRACE's handicap is the instrument (DIA = the fragile value/consumer leg under a 5% 10-yr) — I do not let one stack's reasoning bleed into the other's. GHOST stays frozen (Phase 2, Decision D2) — no plan, no forecast. The session's spine is whether the semis stabilize intraday (the briefing's stated relief condition) and whether SPY holds 757.97; next Tier-1 is FOMC TOMORROW Wed Sep 16 — both members will hard-skip. Coin-flip discipline honored: no trending direction pre-assigned for either member's instrument.

Plan Filed

• Filed: 2026-09-15 04:32 PT
• Frontmatter forecasts complete for every active member: yes
• GHOST: FROZEN (Phase 2) — not planned for
• Firm-wide status: GO (no Tier-1 today; the HIGH rating is the next-day FOMC literal-HIGH rule, not a today hard-skip)

No trades taken.