Weekly money-loop review — rules, omissions, and evidence gaps
Weekly money-loop review — rules, omissions, and evidence gaps
Internal accountability note, written September 7 against September 4 settled tape. The operational EOD/EOW acceptance remains in progress. This note records completed judgments and names the event evidence still pending; it is not a completed-run stamp or a new set of desk calls.
Score and attribution
The decision CLI scored 1,808 rows, with 2,981 complete horizon windows, 23 no-data rows and 40 voided rows skipped. No completed window had null excess. Backfill added three legacy Pelosi decision records. This pass appended a recommendation readout and regenerated the ladder after attribution.
Independent review supported 39 required attributions: 35 within-plan, three discipline and one execution. All 39 writes succeeded through the decision CLI; readback leaves ten required evidence gaps and 15 optional rows. Within-plan means the recorded action followed the contemporaneous rule. It does not mean the action made money, the original research was correct, or the rule deserves promotion. A later rebound cannot by itself establish an improper exit.
The three discipline cases are historical: BE's June 23 momentum allocation exceeded its then-adopted maximum; KLAC's June 23 trader entry exceeded the position box and had no written rationale; VEEV's June 23 bench exit invoked discretionary rotation instead of its mechanical exit rules. NET's July 15 momentum exit applied an RSI-recovery tactic to a breakout entry that did not meet that tactic's entry prerequisite. That retired book's mechanism is not fully established by the attribution. No active policy is changed by this note.
Ten required rows stay unassigned: ALAB momentum June 26 original setup; RIOT yolo June 26 thesis; LASR yolo July 11 sizing authority; LASR momentum July 11 entry authority and missing session; GOOGL yolo June 27 sizing tier; and conviction's MXL July 10, VST July 24, ARM July 23, MU July 23 and LEU August 4 multi-condition entries. Missing historical proof is not replaced by September evidence or by selecting the most plausible cause.
The readout reports d30 mean/median excess of -1.32%/-1.25% for desk:board (135 complete windows) and -2.72%/-1.53% for desk:read (129 complete windows). External random-entry and momentum baselines are unavailable because historical same-date summary universes are missing. These descriptive forward-price scores cannot establish strategy profitability or the causal value of the analyst. The seven-day conversion readout reports 42 armed rungs, four fired, two converted and zero declined. These are the readout's own windows and population, not a count of this run's new trades.
That readout preceded the later September 7 operation: an explicit MU pass was subsequently recorded after the lane-capacity refusal. The earlier zero-declined count is historical to the quoted readout, not the final run state.
Omission dispositions — 17 reviewed
These are September 7 dispositions against the September 4 summary rows. Historical tags stay historical until their qualitative conditions are checked. A close relative to a level does not prove the intervening reclaim or held base.
| Symbol / source date | Numerical evidence | Disposition and missing proof |
|---|---|---|
| $DELL / Aug 12 | 524.14 exceeds old 418–423 band and 418.72 invalidation. | Historical intact-but-priced stands; old pullback entry absent. Updated company evidence and a current setup required. |
| $FCX / Jul 18 | 72.73 exceeds old 61.63 reconfirmation and 55.77 invalidation; outside 56–63 zone. | Price-side condition met; historical intact remains provisional. Reclaim/hold chronology and current primary thesis evidence are missing. |
| FUBO / Jul 17 | No admitted current summary. | Historical broken is neither reaffirmed nor overturned. Current source identity, market data and primary evidence remain missing. |
| $GDX / Jul 19 | 99.26 exceeds old 83 recovery reference and is not below 72. | Numerical recovery leg met; no-view remains pending instrument underwriting and admission conditions. |
| $ILMN / Aug 12 | 218.22 exceeds old 188–194 band and 178.88 structure threshold. | Historical intact-but-priced stands; old entry absent. Fresh company proof and setup required. |
| $ITB / Jul 17 | 93.91 exceeds old 88–92 zone and 80 stop, below current SMA200 98.51. | No-view stands; long-trend recovery/base and current instrument case remain unreviewed. |
| $MU / Jul 19 | 1016.59 and RSI 60.1 fail old price <1000 AND RSI <45. | The reviewed Sep 6 conditional admission supersedes the old entry instruction: historical breakout must hold 989.96; loss returns entry to WAIT. This does not refresh industry evidence. |
| $NBIS / Aug 13 | 226.39 lies inside old 220–235 band, above 216 and approximately 196. | Band leg met; held retest/base unproven. Historical intact-but-priced stands pending chronology and post-source company evidence. |
| $NTRA / Jun 29 | 328.18 exceeds approximately 229 reset and 214 invalidation. | Historical intact-but-priced/WATCH stands. Reset absent; alternative quality/beat-rate improvement unreviewed. |
| $RVMD / Jun 29 | 210.02 exceeds approximately 161/152 pullbacks and 99 unwind marker. | Historical intact-but-priced stands. Old deep-pullback entry absent; current company case and setup required. |
| $SCCO / Jul 18 | 198.76 exceeds SMA50 188.64 and old support 126.33; below SMA20 201.89, RSI 49.6 fails >50. | Historical intact is not completed entry permission. Joint reclaim/RSI condition and the specified DEF14A review remain missing. |
| $XHB / Jul 17 | 103.25 exceeds old 98–102 zone and 90 stop, below SMA200 106.97. | No-view stands; current zone entry/long-trend recovery and instrument proof absent. |
| $ABCL / Jun 4 | 11.43 exceeds old 4.80–5.40 band and 4.20/3.50 markers. | Historical intact-but-priced stands pending current underwriting/setup. The old expressly uncited AI-health explanation is not reused. |
| $LASR / Jul 17 | 40.06 is below old 50 structure and current SMA200 57.31. | Historical setup failed; reopen. Intact is only a historical business tag. Demonstration delay/failure and contract-cancellation conditions remain unreviewed. |
| $RGEN / Jun 4 | 167.60 exceeds old 115–130/108/105 levels. | Historical intact-but-priced stands. Price cannot establish margin recovery or demand reacceleration; reconcile filing growth and adjusted-margin evidence. |
| $POWI / Jul 17 | 50.45 is below old 58.65 stop and current SMA200 56.37. | Historical setup failed; reopen. Require company evidence plus demonstrated stabilization/reclaim, without relabeling business failure from price alone. |
| $WDC / Jun 29 | 467.46 is below old approximately 511/602 references, above current SMA20 464.23 and SMA200 372.81, below SMA50 506.98. | Historical medium-term setup failed while the long uptrend remains supported. Current operating earnings excluding spin/investment gains need review; SMA20 position does not satisfy the old 602 reclaim. |
New origination dispositions — 16 dated skip/watch notes
These are queue dispositions, not newly issued calls. Every row needs reviewed issuer/instrument underwriting before admission; no proposal follows from a screen label alone.
| Symbol | September 7 disposition | Next proof |
|---|---|---|
| $IREN | Watch recovery: 44.68 above shorter averages, below SMA200 45.66. | Held 200-day reclaim/base; reviewed revenue, delivery and financing case. |
| $SEI | Watch recovery: 55 below all three averages. | Base/reclaim at updated SMA200 59.52 and SMA50 59.87; operating economics. Queue 7D 10.24 differs from the later summary's 10.41; the summary governs this read. |
| $WYFI | Skip call, watch only: 19.54 below SMA200 21.54 and shorter averages. | Held recovery and reviewed issuer identity/business/financing case. |
| $BTDR | Watch base: 12.38 above all averages, annual return 3.08% with negative annual alpha. | Sustained 200-day support, completed flag/retest and operating/cash-flow economics. |
| $KEEL | Watch pullback: 3.47 above SMA200 3.37, below SMA50 4.01. | Held base/50-day recovery, issuer economics and financing obligations. |
| $WULF | Watch recovery: 16.51 above SMA20 16.08, below SMA200 17.87 and SMA50 18.43. | Sustained longer-average recovery/base and delivery/revenue/financing proof. |
| $WGMI | Watch recovery: 47.48 above SMA20 45.86, below SMA200 48.56 and SMA50 50.65. | Held recovery and reviewed mandate, mechanism and underlying economics. |
| $HIVE | Watch base: 3.11 above all averages, annual alpha negative. | Held base/retest, durable relative strength and operating cash generation. |
| $MOD | Watch recovery: 194.66 above SMA20 192.53, below SMA200 210.46 and SMA50 213.11. | Sustained longer-average reclaim/base and demand, margins, cash conversion. |
| $FCX | Watch retest in long uptrend: 72.73 just above SMA20 72.63, above SMA50 66.44. | Demonstrated hold/base and current primary underwriting; old zone is not present entry. |
| $SCCO | Skip new call; retain watch. | SMA20 reclaim, RSI strictly >50, maintained SMA50 recovery and specified DEF14A review. |
| $INTC | Watch pullback: 95.80 above SMA20/SMA200, below SMA50 100.61. | Completed flag/reclaim and current margin, cash-flow and execution proof. |
| $NTAP | Watch retest: 185.59 below SMA20 193.29, above SMA50 178.70 and SMA200 130.62. | Held support/base and demand, margins and cash-conversion proof. |
| $CAMT | Watch recovery: 145.72 below all averages despite positive annual return. | Held base/reclaim including SMA200 152.03; orders/revenue and operating economics. |
| $RGEN | Watch retest; no new call: 167.60 below SMA20 173.20, above SMA50 155.63 and SMA200 141.97. | Held base plus reconciled margin-rebuild/demand-reacceleration evidence. |
| $DRAM | Skip long-trend admission; observe: 59.69 above SMA20/SMA50, no 1Y/2Y/alpha1Y/SMA200. | Reviewed mandate/mechanism and sufficient history for the proposed test; no proxy-filled history. |
Standing questions — event review still pending
Memory bear case. The vulnerable claim is that customer commitments and disciplined supply preserve realized pricing and margins as capacity arrives. A bear would test whether new supply outruns consumption, customer inventories accumulate, or committed orders shrink. The campaign falsifier requires any two verified conditions: accelerated greenfield supply, a negative contract-price quarter, customer inventory build, or major HBM order reductions. DRAM/HBM and NAND need separate evidence. MU's conditional price admission does not settle these tests. The earlier source gap is now narrowed by the completed September 7 pricing-evidence review. TrendForce's new DRAM forecast describes affordability pressure and shifts toward smaller server modules alongside low supplier inventory and disciplined wafer additions. This sharpens the consumption risk, but none of the four exact failure legs is established: supplier inventory is not customer inventory, slower positive pricing growth is not a negative quarter, node migration is not accelerated greenfield supply, and smaller conventional modules do not establish major HBM cancellations. Broader industry trigger coverage remains incomplete. The active campaign and September 30 review remain; no conviction upgrade follows.
Concentration. The September 6 adopted primaries are Memory and Electricity. Their AI-linked branches share exposure to customer infrastructure spending, financing and delivery execution. No adopted capital weights or enumerated primary-ticker denominator were supplied; exposure share and change since the last pass are unmeasured. A verified major-customer spending guide-down is a common stress test. Memory order/pricing deterioration and power-equipment or project deferrals would establish transmission, rather than assuming all supplier earnings fail together. AI-capex-digestion is a risk lens, not a third primary. The old three-of-four-primary concentration estimate is not current.
Operational follow-through still open
MU's conditional in-zone flag was confirmed and SNDK's equality-based flag
retracted through the owning CLI. MU then received an explicit pass after the
existing lane-capacity guard refused conversion; SNDK remained retracted.
The trip bookmark was advanced only after all five changes were triaged.
No independent new signal was earned. The complete 75-name news pull has
74 ok receipts and RAM degraded with raw search retained. Macro fulfilled three
queries but admitted zero recent results. Quote/index pages and inferred dates
do not close the event checks above. No forced call, trade, policy promotion or
desk-wide completion claim is made here. The bounded memory review is recorded
as 2026-09-07-memory-pricing-evidence-review; unchecked campaign events and
shipping conditions still prevent an overall not-triggered verdict.
Sources
TrendForce DRAM research, September 7 and enterprise-SSD research, September 1: attributed industry forecasts, with reviewed source observations in
research/inputs/2026-09-07-primary-event-checks-memory-software-and-shipping.md.research/perspectives/ACTIVE.json, canonical review2026-09-07-memory-pricing-evidence-review, written throughresearch:campaign reviewafter the fresh market-only judgment.research/trading/decisions/LEDGER.json,READOUTS.jsonl,LADDER.jsonandORIGINATION-QUEUE.json, read/written through the recorded decision commands where applicable./tmp/ticker-repair-r3-decision-score.log,/tmp/ticker-repair-r3-decision-readout.log,/tmp/ticker-repair-r3-attributions-approved.json,/tmp/ticker-repair-r3-attributions-apply.log,/tmp/ticker-repair-r3-attribution-after.json./tmp/ticker-repair-r3-decision-backfill.log,/tmp/ticker-repair-r3-attributions-result.json(all 39 command results),/tmp/ticker-repair-r3-ladder-after-attribute.stderrand the post-writeLADDER.jsongenerated timestamp.research/workflow/RUNS.md, run2026-09-07-repair-closure-eod; the two September 7 orchestration audits retain the implementation/operational distinction.Reviewed omission source identities and complete long-horizon numerical receipt table follow below. These are dated historical conditions, not current recommendations.
DELL historical 2026-08-12:
research/deep-dives/2026-08-12-dell-deep-dive.md.FCX historical 2026-07-18:
research/deep-dives/2026-07-18-fcx-deep-dive-refresh-zone-reentry.md.FUBO historical 2026-07-17:
research/company-profiles/_archive/2026-07-17-FUBO.md.GDX historical 2026-07-19:
research/investigations/2026-07-19-thesis-tag-omission-loop.md.ILMN historical 2026-08-12:
research/deep-dives/2026-08-12-ilmn-deep-dive.md.ITB historical 2026-07-17:
research/company-profiles/_archive/2026-07-17-ITB-XHB.md.MU historical 2026-07-19:
research/investigations/2026-07-19-thesis-tag-omission-loop.md.NBIS historical 2026-08-13:
research/deep-dives/2026-08-13-nbis-deep-dive.md.NTRA historical 2026-06-29:
research/deep-dives/2026-06-29-ntra-deep-dive.md.RVMD historical 2026-06-29:
research/deep-dives/2026-06-29-rvmd-deep-dive.md.SCCO historical 2026-07-18:
research/deep-dives/2026-07-18-scco-deep-dive-cost-curve-governance-discount.md.XHB historical 2026-07-17:
research/company-profiles/_archive/2026-07-17-ITB-XHB.md.ABCL historical 2026-06-04:
research/deep-dives/2026-06-04-abcl-deep-dive.md.LASR historical 2026-07-17:
research/company-profiles/_archive/2026-07-17-LASR.md.RGEN historical 2026-06-04:
research/deep-dives/2026-06-04-rgen-deep-dive.md.POWI historical 2026-07-17:
research/company-profiles/_archive/2026-07-17-POWI.md.WDC historical 2026-06-29:
research/deep-dives/2026-06-29-wdc-deep-dive.md.research/deep-dives/2026-09-06-mu-campaign-01-entry-proof.md(superseding MU admission).research/perspectives/ACTIVE.jsoncanonical campaign case objects viaresearch:campaign show;.brief/operating-book.mdSeptember 6 adopted scope.
Long-horizon receipt table
All rows below are supplied summary fields dated September 4. Null is shown as unavailable. Annual alpha is the supplied excess field; no new return math is performed here. The inspected 7D window is August 28–September 4 and 30D is August 5–September 4. FUBO has no admitted row.
| Symbol | 1Y % | 2Y % | Annual alpha | vs SMA200 % | Supplied thesis state |
|---|---|---|---|---|---|
| $DELL | 313.78 | 380.6 | 295.13 | 106.8 | intact-but-priced |
| $FCX | 57.87 | 74.79 | 39.22 | 19.4 | intact |
| $GDX | 53.72 | 168.92 | 35.07 | 10.5 | no-view |
| $ILMN | 123.4 | 69.7 | 104.75 | 43.7 | intact-but-priced |
| $ITB | -17.75 | -20.14 | -36.4 | -4.7 | no-view |
| $MU | 718.44 | 1038.53 | 699.79 | 67.6 | intact-but-priced |
| $NBIS | 248.78 | unavailable | 230.12 | 46.3 | intact-but-priced |
| $NTRA | 97.03 | 189.81 | 78.38 | 39.3 | intact-but-priced |
| $RVMD | 439.2 | 395.21 | 420.55 | 56.7 | intact-but-priced |
| $SCCO | 97.14 | 106.29 | 78.49 | 11.3 | intact |
| $XHB | -11.91 | -8.6 | -30.56 | -3.5 | no-view |
| $ABCL | 174.76 | 353.57 | 156.11 | 122.5 | intact-but-priced |
| $LASR | 35.66 | 258.96 | 17.01 | -30.1 | intact |
| $RGEN | 38.99 | 16 | 20.34 | 18 | intact-but-priced |
| $POWI | 12.79 | -17.27 | -5.86 | -10.5 | intact |
| $WDC | 416.59 | 882.47 | 397.94 | 25.4 | intact |
| $IREN | 70.99 | 564.88 | 52.34 | -2.1 | unrated |
| $SEI | 107.63 | 367.29 | 88.97 | -7.6 | unrated |
| $WYFI | 20.17 | unavailable | 1.52 | -9.3 | unrated |
| $BTDR | 3.08 | 111.26 | -15.57 | 3.2 | unrated |
| $KEEL | 177.6 | 78.87 | 158.95 | 2.8 | unrated |
| $WULF | 86.13 | 322.25 | 67.48 | -7.6 | unrated |
| $WGMI | 62.77 | 201.27 | 44.12 | -2.2 | unrated |
| $HIVE | 14.76 | 11.07 | -3.89 | 7.2 | unrated |
| $MOD | 43.14 | 83.12 | 24.49 | -7.5 | unrated |
| $INTC | 289.27 | 393.05 | 270.62 | 29.2 | unrated |
| $NTAP | 56.18 | 55.85 | 37.53 | 42.1 | unrated |
| $CAMT | 86.34 | 74.39 | 67.69 | -4.2 | unrated |
| $DRAM | unavailable | unavailable | unavailable | unavailable | unrated |
Numerical source files:
research/market-engine/data/summaries/ai-infrastructure.json.research/market-engine/data/summaries/ai-power.json.research/market-engine/data/summaries/biotech-capital-cycle.json.research/market-engine/data/summaries/drone-defense.json.research/market-engine/data/summaries/focus.json.research/market-engine/data/summaries/hard-assets.json.research/market-engine/data/summaries/ibb.json.research/market-engine/data/summaries/market-pulse.json.research/market-engine/data/summaries/memory.json.research/market-engine/data/summaries/neoclouds.json.research/market-engine/data/summaries/thebuildout-ai-data-center-operators-neoclouds.json.research/market-engine/data/summaries/thebuildout-ai-storage-data-pipeline.json.research/market-engine/data/summaries/wfe-test-metrology.json.
Source links
Tweets, videos, filings, articles, and source pages tied to this read.