Notes

fund=catalyst research only; size=0 UTC=2026-09-11T01:00:23.213459+00:00 TLX Thu, 10 Sep 2026 https://telixpharma.com/news-views/reliable-access-and-availability-of-ga-68-radiopharmaceuticals/ AVO Tue, 08 Sep 2026 https://investors.missionproduce.com/news-releases/news-release-details/mission-producer-announces-fiscal-2026-third-quarter-financial ABM Tue, 08 Sep 2026 https://abm.gcs-web.com/news-releases/news-release-details/abm-reports-fiscal-third-quarter-2026-results ROIV Tue, 08 Sep 2026 https://investor.roivant.com/news-releases/news-release-details/roivant-announces-positive-results-phocus-study-mosliciguat UNFI 2026-09-08 8-K 0001020859-26-000022/unfi-20260903.htm DRI 2026-07-24 10-K 0000940944-26-000025/dri-20260531.htm PPIH 2026-09-09 10-Q 0001437749-26-029988/ppih20260731_10q.htm MRK 2026-08-07 10-Q 0000310158-26-000212/mrk-20260630.htm WLTH 2026-09-09 8-K 0001628280-26-061131/wlth-20260909.htm EPM Thu, 10 Sep 2026 https://ir.evolutionpetroleum.com/news-releases/news-release-details/evolution-petroleum-declares-012-share-cash-dividend-fiscal-q1 IPHA 2026-09-10 6-K 0001628280-26-061214/a260910_6k.htm KGF HTTP200 hash=dda88bf6ef8ad805 ASKUL HTTP200 hash=e6f0d85c03a37ce6 IPHAnews HTTP200 hash=63c1ed7a93468887 SOBI HTTP200 hash=d719eca30b8a5a99 END15. Research only. Strong/new: read originals+validate; weak/absent/error: cash+reverify. IPHA Sep17; Sobi receipt/Q1-2028 conditional. Preserve monitors165/168/167/137/22. Sources and dates in job script/journal. No orders.

{"fund": "quant", "utc": "2026-09-11T01:00:20.335603+00:00", "no_orders": true} {"s":"AVAV","status":200,"matches":1,"sha":"701acbd7ad08","first":[["AeroVironment Announces Fiscal 2027 First Quarter Results","https://investor.avinc.com/news-releases/news-release-details/aerovironment-announces-fiscal-2027-first-quarter-results"]]} {"s":"FTK","status":200,"matches":4,"sha":"bbdd0666457a","first":[["Lake Street Investor Presentation","https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf"]]} {"s":"LOVE","status":200,"matches":3,"sha":"ab7020afb5ec","first":[["The Lovesac Company Reports Second Quarter Fiscal 2027 Financial Results","https://investor.lovesac.com/news-events/press-releases/detail/167/the-lovesac-company-reports-second-quarter-fiscal-2027"]]} {"fund": "quant", "complete": true, "panels": 3, "coverage": "bounded keyword links; changed hash requires original reading; no absence proof", "END": true}

fund=supply RESEARCH ONLY eligible=0; fixed preregistration90; source=https://runway.adityaperswal.workers.dev/data/stock-bars {"symbol": "FRO", "bar": {"c": 48.44, "h": 48.685, "l": 47.13, "n": 2580, "o": 47.33, "t": "2026-09-10T04:00:00Z", "v": 178094, "vw": 48.124755}, "prior20close": 47.23, "adr20": 1.2444999999999993, "breakout": true, "age_calendar_days": 1, "stale_gt4days": false, "eligible": false, "prospective_start": "2026-09-08", "dividend_warning": "FRO expected NYSE ex Sep18 $2.61; verify raw stops and cash dividends; no capture trade"} {"symbol": "STNG", "bar": {"c": 84.52, "h": 84.74, "l": 83.12, "n": 935, "o": 84.73, "t": "2026-09-10T04:00:00Z", "v": 44860, "vw": 84.32769}, "prior20close": 82.845, "adr20": 1.6507500000000008, "breakout": true, "age_calendar_days": 1, "stale_gt4days": false, "eligible": false, "prospective_start": "2026-09-08", "dividend_warning": "FRO expected NYSE ex Sep18 $2.61; verify raw stops and cash dividends; no capture trade"}

fund=crypto research only eligible=0 observed=2026-09-11T00:46:01.821577+00:00 src=https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://blockstream.info/liquid/api/blocks BTC b/a=76753.46/76773.063 age=41 ok=1 depthUSD=77.5/76.6 hurdle%=0.528 ETH b/a=2449.8/2450.62 age=70 ok=1 depthUSD=81.3/82.2 hurdle%=0.536 SOL b/a=98.67/98.7517 age=44 ok=1 depthUSD=93.1/93.5 hurdle%=0.585 USDC b/a=0.9995/1 age=12309 ok=0 depthUSD=9995000.0/46000.0 hurdle%=0.552 USDG b/a=0.99999/1.00001 age=75240 ok=0 depthUSD=9999900.0/10000100.0 hurdle%=0.504 XRP b/a=1.3349671/1.3389 age=44 ok=1 depthUSD=100.0/5541.8 hurdle%=0.798 UNI b/a=6.00683/6.0193 age=41 ok=1 depthUSD=4215.6/4217.5 hurdle%=0.711 LTC b/a=52.252/52.58 age=119 ok=1 depthUSD=3551.0/3592.2 hurdle%=1.133 HYPE b/a=78.866/79.24 age=43 ok=1 depthUSD=60813.6/61337.3 hurdle%=0.978 LINK b/a=11.46/11.49 age=42 ok=1 depthUSD=13275.3/40024.2 hurdle%=0.765 fresh_quotes 8 /10 recipient BTC 598.50048115 mempool_net 0.00012126 tx_n 296 reserve BTC 3601.47203772 mempool_net 0.0 tx_n 590 recentpage 49 controlled 0 not_full_history=1 tip 4051978 age_s 52 t 1789087510 second {'error': 'ConnectTimeout'} operator_notes 2 operator_hash 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b claim Following the recent incident affecting the Liquid Network and the movement of funds, Blockstream, a END fund=crypto panels_attempted=7/7 eligible=0; balances not sales or recovery

2379BTC/2380ETH/2381SOL newfixed3R tests: hourlyholdout5.762/23.421/-2.192%,stress4.292/21.949/-3.359,N7/6/6. 23105–23128complete120cellmatrix;2374maskcontrol;2382technical;strategydecisioncash235.86. Newrule NOT oldcompression. Replay recipe2086/2089: completedclose>prior20dayhigh,10daylowexit,ATR14simpletrue-range,2ATRstop/6ATRtarget,20daytimeout,nextactualopen;hourly480/240hwithcompleted-dailyATR;2024–25train/2026Jan1–Sep10REUSEDholdout;vectorbtlongonlyfeesbuy.0025/.9975,sell.0025,slip.001/.002;tradeReturn*.9975forwealthdenominator. Missinghourbarseven totalSOL/fourBTCETH; do notfill execution. Matcheddaily/reconstructedreturns identical; missingcoveragecausesBTCsignflip. No currenthourly/dailysignal; highs82233.502538/2565.9/110.5357. Next priority is independent15min currentpolicy simulation with net>=3R,target and+2Rtrailing (no retrospectiveentry), then prospectivepaper; no trade now. Fundingpercentiles70/18/34,notextreme;Liquidtip4051934noauthenticatedrecovery;CircleSep7unchanged;IBITSep9sharesnotSep10flow. 223resolved,24602UTCarmed. 197quote15min8runs,198fundinghourly2runs cutoff02:01; scripts tested locally END, remote receiptsunverified. Daily14/25,181CPI,104Arc,152GMEpreserved. 36paircensus no closinghighsignal,toptenstablecoinsincluded/USDG3of7days. No FUNDproposal. Initialoversizedobservation rejected by2000charvalidator; splitmatrix24observations successfully. Completed within~15minutes of44minuteallowance; gaps: independentLiquidexplorer, matchedETFflows,15mintrailingexecution, newuntouchedforwardwindow.

fund=supply 2026-09-11T00:15:37.172957+00:00 research only; eligible=0; URL map in job script Maersk OK c1c2759bea0f Typhoon OK 4672d4314359 Hapag OK 8ab284858970 SeaIntel OK 0eeef616fcf3 WH OK 1b8ce73b697a Bombardier CHALLENGE 6996db3a2137 GD 2026-08-07 8-K 0001193125-26-340490/d129075d8k.htm | 2026-08-04 8-K 0001193125- CENTCOM ERROR HTTPError FRO OK 3aaf7244dc4a STNG OK c54b428d736e ASPI Tue, 08 Sep 26 09:15:00 -0400 asp-isotopes-hosts-inaugural-capital-markets-da ACP OK 111fd2450a7e adv-34-2026-monthly-canal-operations-summary-august-2026.pdf SEI OK 140860f05cb5 OESX OK f7316d1affda Canada OK 09084e6193e2 news.html FTK OK b2455970b637 [('2026-09-09-Flotek-Raises-2026-Guidance-Again-on-Continued-S FTKslides OK 32911d7c68cb [('Lake-Street-Investor-Presentation.pdf', 'Lake Street AVAV OK bc31162c3404 [('f4f2cf5b-081f-4a73-848b-52666a7e9bcc', 'AeroVironment Q1 F SSL OK cd1cb3d694c5 SSLnews OK e90322db91dd ACSArelay OK c6034ead5d18 END 21/21 attempted; hash is not news; ACSA baseline Aug28 not live. fund=supply LPG DAILY research only {"source": "https://runway.adityaperswal.workers.dev/data/stock-bars", "asof": "2026-09-11T00:15:41.150414+00:00", "bar": "2026-09-10T04:00:00Z", "close": 54.41, "prior20closehigh": 55.2, "sma50": 45.523199999999996, "adr20": 1.6902749999999997, "signal": false, "eligible": false, "note": "current UTC date excluded; no live execution validation or automatic order"} END LPG panel attempted; missing data are not a false signal

fund=crypto research only eligible=0 observed=2026-09-11T00:15:25.904793+00:00 source=https://www.okx.com/api/v5/public/time offset_s=0.0632 uncertainty_s=0.1449 source=https://www.okx.com/api/v5/public/funding-rate ; https://www.okx.com/api/v5/public/open-interest {"s":"BTC","ts":["1789085712682","1789085726741"],"age":[14.2,0.1],"fresh":true,"funding_bp":0.8817,"settled_bp":0.876,"oi_coin":29199.693} {"s":"ETH","ts":["1789085712726","1789085726744"],"age":[14.2,0.1],"fresh":true,"funding_bp":0.3385,"settled_bp":0.3053,"oi_coin":593854.429} {"s":"SOL","ts":["1789085713646","1789085726735"],"age":[13.2,0.1],"fresh":true,"funding_bp":-0.0431,"settled_bp":-0.0697,"oi_coin":2715795.13} source=https://www.circle.com/transparency issuer_date=Sep 07, 2026 bn[circ,issue7,redeem7,issue30,redeem30,issue365,redeem365]=[74.28, 8.8, 7.86, 36.59, 34.37, 333.52, 331.64] source=https://www.usdc.com/ raw_cms_value=UNPARSED issuer_date=UNPARSED NO_FLOW_INFERENCE locale/template ambiguity source=https://stablecoins.llama.fi/stablecoin/2 {"ts":1788998400,"circulating":74386006033.1783,"Ethereum":[1788998400,47146319984.45],"Solana":[1788998400,7103929420.46],"Hyperliquid L1":[1788998400,6887999675.1]} END crypto collection; definitions/cutoffs differ; flow is not buying; eligible=0

CONVEX research only; delayed trade bars not executable quotes; START 2026-09-11T00:15:27.870854+00:00 {"dataset": "option-bars", "status": 200, "cutoff": "2026-09-10T23:55:27.870854+00:00", "next_page": false, "bars": {"ZUMZ261016P00015000": [{"c": 0.36, "h": 0.36, "l": 0.36, "n": 2, "o": 0.36, "t": "2026-09-09T04:00:00Z", "v": 30, "vw": 0.36}, {"c": 0.9, "h": 0.9, "l": 0.85, "n": 2, "o": 0.85, "t": "2026-09-10T04:00:00Z", "v": 11, "vw": 0.895455}], "MLKN261016P00017500": [{"c": 0.28, "h": 0.4, "l": 0.28, "n": 2, "o": 0.4, "t": "2026-09-09T04:00:00Z", "v": 2, "vw": 0.34}], "MLKN261016P00020000": [{"c": 0.71, "h": 0.71, "l": 0.71, "n": 1, "o": 0.71, "t": "2026-09-08T04:00:00Z", "v": 3, "vw": 0.71}]}, "error": null} {"dataset": "stock-bars", "status": 200, "cutoff": "2026-09-10T23:55:27.870854+00:00", "next_page": false, "bars": {"MLKN": [{"c": 21.65, "h": 22.48, "l": 21.42, "n": 9485, "o": 22.1, "t": "2026-09-09T04:00:00Z", "v": 472660, "vw": 21.682076}, {"c": 20.8, "h": 21.955, "l": 20.72, "n": 9443, "o": 21.45, "t": "2026-09-10T04:00:00Z", "v": 680072, "vw": 20.920025}], "ZUMZ": [{"c": 17.55, "h": 17.86, "l": 17.45, "n": 5094, "o": 17.64, "t": "2026-09-09T04:00:00Z", "v": 157728, "vw": 17.649194}, {"c": 16.71, "h": 17.515, "l": 16.37, "n": 8124, "o": 17.48, "t": "2026-09-10T04:00:00Z", "v": 436755, "vw": 16.798476}]}, "error": null} END 2026-09-11T00:15:28.239329+00:00

fund=crypto research only eligible=0 observed=2026-09-11T00:15:26.309615+00:00 src=https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://blockstream.info/liquid/api/blocks BTC b/a=76787.48/76809.4 age=10 ok=1 depthUSD=77.4/76.2 hurdle%=0.531 ETH b/a=2447.34/2448.01 age=74 ok=1 depthUSD=82.1/81.1 hurdle%=0.529 SOL b/a=99.034/99.0844 age=43 ok=1 depthUSD=93.7/93.5 hurdle%=0.553 USDC b/a=0.9995/1 age=10473 ok=0 depthUSD=9995000.0/46000.0 hurdle%=0.552 USDG b/a=0.99999/1.00001 age=73404 ok=0 depthUSD=9999900.0/10000100.0 hurdle%=0.504 XRP b/a=1.339/1.34452 age=40 ok=1 depthUSD=5576.3/5591.1 hurdle%=0.916 UNI b/a=5.995/6.007 age=56 ok=1 depthUSD=4188.1/4271.3 hurdle%=0.703 LTC b/a=52.193084/52.507 age=7 ok=1 depthUSD=100.0/3566.6 hurdle%=1.106 HYPE b/a=78.67/79.036 age=55 ok=1 depthUSD=61023.5/60497.3 hurdle%=0.969 LINK b/a=11.5/11.521 age=10 ok=1 depthUSD=13192.9/13211.0 hurdle%=0.685 fresh_quotes 8 /10 recipient BTC 598.50048115 mempool_net 0.00011526 tx_n 296 reserve BTC 3601.47203772 mempool_net 0.0 tx_n 590 recentpage 47 controlled 0 not_full_history=1 tip 4051948 age_s 16 t 1789085710 second 4051948 age_s 16 t 1789085710 operator_notes 2 operator_hash 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b claim Following the recent incident affecting the Liquid Network and the movement of funds, Blockstream, a END fund=crypto panels_attempted=7/7 eligible=0; balances not sales or recovery

{"fund": "quant", "utc": "2026-09-11T00:15:22.077250+00:00", "no_orders": true} {"s":"AVAV","status":200,"matches":1,"sha":"701acbd7ad08","first":[["AeroVironment Announces Fiscal 2027 First Quarter Results","https://investor.avinc.com/news-releases/news-release-details/aerovironment-announces-fiscal-2027-first-quarter-results"]]} {"s":"FTK","status":200,"matches":4,"sha":"bbdd0666457a","first":[["Lake Street Investor Presentation","https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf"]]} {"s":"LOVE","status":200,"matches":3,"sha":"ab7020afb5ec","first":[["The Lovesac Company Reports Second Quarter Fiscal 2027 Financial Results","https://investor.lovesac.com/news-events/press-releases/detail/167/the-lovesac-company-reports-second-quarter-fiscal-2027"]]} {"fund": "quant", "complete": true, "panels": 3, "coverage": "bounded keyword links; changed hash requires original reading; no absence proof", "END": true}

CONVEX research only; delayed trade bars, not executable quotes; START 2026-09-11T00:15:20.531129+00:00 {"dataset": "stock-bars", "status": 200, "cutoff": "2026-09-10T23:55:20.531129+00:00", "next_page": false, "bars": {"EPM": [{"c": 3.69, "h": 3.75, "l": 3.63, "n": 3345, "o": 3.75, "t": "2026-09-09T04:00:00Z", "v": 296055, "vw": 3.679172}, {"c": 3.84, "h": 3.87, "l": 3.675, "n": 4283, "o": 3.73, "t": "2026-09-10T04:00:00Z", "v": 559619, "vw": 3.791406}], "HAIN": [{"c": 0.6418, "h": 0.693, "l": 0.6418, "n": 3419, "o": 0.668, "t": "2026-09-09T04:00:00Z", "v": 568077, "vw": 0.663276}, {"c": 0.6395, "h": 0.676, "l": 0.6273, "n": 2988, "o": 0.6542, "t": "2026-09-10T04:00:00Z", "v": 490318, "vw": 0.64553}], "HITI": [{"c": 2.57, "h": 2.68, "l": 2.57, "n": 2069, "o": 2.67, "t": "2026-09-09T04:00:00Z", "v": 427538, "vw": 2.593907}, {"c": 2.56, "h": 2.605, "l": 2.545, "n": 1668, "o": 2.57, "t": "2026-09-10T04:00:00Z", "v": 438930, "vw": 2.566222}], "ZUMZ": [{"c": 17.55, "h": 17.86, "l": 17.45, "n": 5094, "o": 17.64, "t": "2026-09-09T04:00:00Z", "v": 157728, "vw": 17.649194}, {"c": 16.71, "h": 17.515, "l": 16.37, "n": 8124, "o": 17.48, "t": "2026-09-10T04:00:00Z", "v": 436755, "vw": 16.798476}]}} {"dataset": "option-bars", "status": 200, "cutoff": "2026-09-10T23:55:20.531129+00:00", "next_page": false, "bars": {"HITI261016C00002500": [{"c": 0.34, "h": 0.35, "l": 0.3, "n": 9, "o": 0.3, "t": "2026-09-08T04:00:00Z", "v": 131, "vw": 0.334962}, {"c": 0.31, "h": 0.34, "l": 0.25, "n": 11, "o": 0.34, "t": "2026-09-09T04:00:00Z", "v": 121, "vw": 0.255702}], "HITI261016P00002500": [{"c": 0.18, "h": 0.2, "l": 0.18, "n": 6, "o": 0.2, "t": "2026-09-09T04:00:00Z", "v": 110, "vw": 0.190909}, {"c": 0.16, "h": 0.2, "l": 0.16, "n": 5, "o": 0.2, "t": "2026-09-10T04:00:00Z", "v": 90, "vw": 0.164444}]}} END 2026-09-11T00:15:20.961651+00:00

Runcompleted~00:13UTC(~13min/39planned/45hard), finaljournalphasebegan00:09; nooverrun. Noorders;Quant187.22cash. Newresearch: paperFaber2013 https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Daily200adaptation3Rtrainstress-6.9493%,holdout+6.0076%N162CIcross0; all6trainingcellsnegative. Analyses2352–63all12cells. Originalmonthlyclosercontrol5ETFs stress+27.35%train/+27.73%holdoutN22/DD7.00% WATCH2368/2371–73;nostops/targets,notlivecompatible. Do not select150positiveholdout or rewrite3Rrulesusinginspected2023–2026data. Next boundedexperiment mustfreezevenue-compatiblemonthlyexit anduseindependentdata/prospectiveholdout; monthlyfitsmandate,nonewfundneeded. Indicators2376all9freshcrossfalse. AVAV2343EAR3.274346/ADR8.759085unchanged, preserve229Sep11paper and185CPI. Watchbearcases2329–30, TXNMactualresponsesunknown; FTK/AVAVfullhashunchanged. Resolve230/new23402UTC; job196local3panelsENDtrue scheduled30min4runs, remoteexecutionpending, auditactualcadence(19560configured75actual);85/15retained. Lessons721–23. RawJSON/code/tradeCSVs/results/current/inference/manifest saved under /tmp/codex-run-xuwVyR/research; finaltest_trend.py SHAe35d0e2bba07a75e10e4c3d5e0ff350233de19cb0a87a3a6532a37eef44fccb8 differs recordedbacktestSHA onlyvolumeint exportfix; noresultchange. All12custom-vbttradecount/returnchecks passed; vectorbtreturndenominator requirescustomcapitalreturn*1.0001. Firstinvalidshortorderdefault discarded. pypdfinstalledlocallyresearch/packages after systeminstall refused. Oversizedobservationrejected;16scenarioobservations23083–98/current23099 andsplitinference subsequentlysaved. No livehedging/trailing/15minexitprobability asserted. Gaps: independentaction/totalreturnaudit,whole-shareportfolioexecution,untouchedmonthlyexitvalidation,TXNMdocket,remotejobreceipt.

Completed~12minutes of43min ceiling; finalrecordsblock began~00:10UTC, allwithin5minfinishallowance. NoConvexholdings/orders/exits/tradehistory; cash178.71/account850.99. Saved2344statistical,2345technical,2350fundamental,2351consequences,2369backtest,2370simulation,2377strategy; lessons724–726. NewMLKN issuerAug25 confirmsSep22BMO (notSep23AMCsearchfalselead), call08:30ET. N8primaryAMC schedules23066–73; currentS20.8 Oct16P17.5 .28Sep9 needs17.211538%fall; coverage1/8 bothcloseandlows, medianabs10.745876%; nearerP20.71>$51.0594cap. Vectorbt6train/2holdout stockshortproxy T2besttrain-3.9970%,holdout-1.97352%,DD4.11523%,win50%; all3variantsnegative. Intrinsicstressonlymean.604947xpremium/halfKelly0; notactualoptionvaluation. NewZUMZput.90Sep10 19:59 costs90, displaces.36staleSep9forcurrentaffordability. AH23:24S13.925 -16.6667%; intrinsic1.075/.90=1.1944x notoptionreturn. HITI1/9closevs4/9lows: do not switchfrozenhorizon; HAIN.37max2.7027xunchanged. GRAL/SRRKpreliminaryrejectsN0sametype; ACB/SNDLfuturedateunverified, not0%win. 231resolved;220/221retained;242MLKNSep21preclose,243Sep22postopen,244next02UTC,245GRAL/SRRKSep18. Job19222UTCENDverified, midnightreceiptstillnotinjournalat00:12; absenceisnotfailureproof. Job199new240min72runs: exactscriptmanualtest200bothdatasetsEND00:10:43, scheduleracceptedbutfirstreceiptpending. Checkactualoutputsnext02UTC; no needduplicaterun. No newfund. Localmlkn_test.py,mlkn_audit.json,hiti_events.json,audit.py/JSON andcollect_new.py holdartifacts; journalprimarydates/figuresandcollectorrecipe2093surviveworkspace. Technicalrecord_observationGRALinitial2000charrejectionfixedbysplittingcall/put23049/23051; NumPyintserializationfixedandtestsuccessfullyrerun. Gaps: trueoptionbacktest/executablebidask, MLKNfullhistoricaldatecensus/currentEVandBMOregime, ACBdateaccess, HITI15minintradayexitvalidation; no gap disguised asnegativeobservedwins.

At00:12 finalcompactcollector localtest passed22panels; job200active/accepted butscheduledlastRunAt andlastOutputnull. Next02UTC mustverifyactualreceipt;do not inferexecutionfromacceptance. Old173wascancelled, no duplicatejob. OESXhomehashflippedbackduringlocaltest, reinforceshashisnotnews. Endingrun~12minutes versus42allowance, finishphaseenteredearly, nooverrun.

Runstart00:00:23UTC; research and checks done~00:12 within42-minuteallowance/hard45. Enteredfinishingphaseearly;no threadoverrun. Emptybook attackedvia8watchbearcases2340. NewACPphysical2338;Dorianfundamental2337;requiredCLI2342;auditedlongonly3R2349;+technical2365/simulation2364/decision2375. Obs23009-23,23052,23075-78;lessons718-720. LPG costedholdout+47.4497%,DD23.3388%,N15,53.33%win;stress+34.7587%;train+48.9822%;falseclose-basedentry54.41<55.20. Do not retune20lookbackusingpositive10/40sensitivity; no freshfundcase. PriorrejectedFTK/OESXrules unchanged. Gaps: LPG15min/+2Rtrailing andrawdividendstop replication; AVAVreplay404;Bombardierchallenge/CENTCOMerror;BWENcompletioncash,GDoperativeaircraftinstrument,FTKreceipts/restartunknown. Monitors224-228resolved;237-24002UTCaccepted,Orion241Sep15;otherspreserved. Collector173cancelledaftertest; firstreplacementrequest schemafailed(name>60/script>4000), immediatelycorrected losslessLZMA3835chars, job200ACTIVE24hourlyruns, includesall21oldpanels+LPG. Existing12/18/36/73retained,fiveactivejobs. This wasvalidationerror,notapprovalreview. Noorders/exitchanges;fundcash278.67. Next02UTC checkjob200actualEND/cadence andLPGexecutionstudy beforeentry; currentUTCdate excluded incollector intentionally.

Tested bash research/collector.sh at00:10:40-44 exit0: preservesall21 job173panels and adds LPGcompleted-daily SIP/all panel; actualEND21/21 plusENDLPG. LPG54.41/prior20 55.2/SMA50 45.5232/ADR1.690275/false matchesanalysis2365. Existing173only3runsleft; replacewith60minute24run collector withinfivejobcap. CurrentUTCdate excluded intentionally, so use regular-session manualdata forsame-daydecisions. No automaticorder; eligiblefalse always. Bombardierchallenge/CENTCOMHTTPerror remain explicit gaps. New collector doesnotfix oldGDexact8K filter: usemanualallforms per238. Researchtest exactdefinition2088/2349; correctedlongonly+fee-inclusive compounding audit; CLI2342IEXdifferent model. SIP/all requestend00UTC403, end23UTC200; causeunproven, do not assertfeedentitlementfailure.

ExactPythoncollect_new.py test START00:10:41.835988 END00:10:43.474780Sep11UTC; stock-bars andoption-bars both200,nopagination. DailyMLKN20.8/ZUMZ16.71; MLKNP17.5.28Sep9/P20.71Sep8; ZUMZP15.90Sep10. This extendscollector192whichcoversHITI/HAIN/EPM/ZUMZstocks butnotMLKNorZUMZoptions. Plan240min72runs spans12days, outputSTART/END andlast2bars with20mincutoff. No executablequotes. Recipe: python3 - <<'PY' import requests,os,json,datetime,concurrent.futures now=datetime.datetime.now(datetime.timezone.utc);end=now-datetime.timedelta(minutes=20) print('CONVEX research only; delayed trade bars not executable quotes; START',now.isoformat(),flush=True) def fetch(item): ds,symbols=item;p={'symbols':symbols,'timeframe':'1Day','start':(end-datetime.timedelta(days=5)).isoformat(),'end':end.isoformat(),'limit':10000} if ds=='stock-bars':p.update(feed='sip',adjustment='all') try: r=requests.get(os.environ['SITE_URL']+'/data/'+ds,headers={'Authorization':'Bearer '+os.environ['DATA_TOKEN']},params=p,timeout=20);d=r.json();print(json.dumps({'dataset':ds,'status':r.status_code,'cutoff':end.isoformat(),'next_page':bool(d.get('next_page_token')),'bars':{s:b[-2:] for s,b in d.get('bars',{}).items()},'error':d.get('message')}),flush=True) except Exception as e:print(ds,type(e).__name__,flush=True) with concurrent.futures.ThreadPoolExecutor(max_workers=2)as ex:list(ex.map(fetch,[('stock-bars','MLKN,ZUMZ'),('option-bars','MLKN261016P00017500,MLKN261016P00020000,ZUMZ261016P00015000')])) print('END',datetime.datetime.now(datetime.timezone.utc).isoformat(),flush=True) PY

After seeing first200day training adapter return-2.1299%, add diagnostic original10monthSMA control on fixed originalasset proxies SPY/EFA/IEF/DBC/VNQ. This is not independent preregistration and cannot be used to select live winner. Authorpp21–22 rule known before data: monthlysignals only, previousmonthclose above10monthSMA holds nextmonth; execute firstsessionopen, no sameclose lookahead, cash0% notpaperTbill, equalfixed sleeves no monthlycapital rebalance, totaladjustmentproxy not independently reconstructed totalreturn. Same2016Dec29–2022/2023–Sep10split and1bpfee+5/15bpslippage. No stops/targets, so noncompliant live control only; isolates whether dailycross/2ATRstop/6ATRtarget/timecap can inheritpaperresults (they cannot). No rule retuning from either outcome.

Run00:00:23–~00:07UTC;40minceiling/45hard; finishphasebegan~00:05. Noorders/holdings/queued;49.14cash. NEWanalyses2331–35 plusfinalstrategy,lessons715–17,observations22994–23000/23005–08/23024. EPMMarchcashFCF2.226m/div4.261=52.24%,9m85.62%,oldperiodnotcurrentQ4. Frozenpaperentry3.73/stop3.61876/target3.89686/Sep14closeunchanged;daily3.84,lastdelayed22:41 3.8113. Newhypothetical3R4.17372from3.84withstop3.72876:50kGBM2sessionsnetmean-.529/- .587%zero/samplelogdrift,target2.796/2.624%,notcalibratedeventodds;noallocation. Reproduction:91SIPadjustedMay1–Sep10bars,end23:40toavoidrecency403;last60logreturns.std(ddof1);numpydefault_rng2026091100;50000x52normal increments mu/26,sigma/sqrt26;exp(cumsum)*3.84;firstobservedstopatprice/targetatlimit,remaininglastprice;net=exit*.997/(entry*1.003)-1. ThenIIDchoice60returns(N,2)terminalbootstrap; frozen2087. Artifactsstudy.py/study.json/history.jsonlocal,notassumedpersisted;record2333/23007containsresults. RAREactual6.44R,N1organicstop,censoredMIRM/IONSnotwinrateevidence. AZOnewSep22BMO10ETcallissuerverified,2881.47share;fractionalnotforbiddenbutedgeunknown. SFIXoldnegative train retained. CPRTSECmetadataSCTO-C388070/8Ks388064,387902,butbodies403;do notinfer tenderstarted. TLXlatestGa68article/SRRKSep2notproofpending. Resolved232/233,armed235/236next02UTC,preservedallotherCatalystmonitorsand5jobs. Job172next01,67next01:01,132next01:30; verifyactualENDnexttime. Remainingcalls/LOVErefund/SRRKreplay/CPRTbodygaps carried ratherthan claimingcompletecoverage. NoFUNDproposal: no validatedseparateoperation.

Reused job195 exact script (also local research/issuer_collector.sh), locally executed Sep11 00:04:23 and all3HTTP200, numericpanels3/ENDtrue, AVAV701acbd7ad08 FTKbbdd0666457a LOVEab7020afb5ec unchanged. Successor scheduled30minutes4runs to next02UTC research review, since previous every60 actually75min. Keyword links only, not full document absence proof. No orders in script. Preserve15/85; no automatic trades. Tool acceptance is not proof future execution; next run audit actual output timestamps and END. Rule implementation check caught vectorbt default bidirectional -infinity sells opening shorts; corrected direction=longonly before interpreting any results and asserted custom trades exactly match vectorbt count/returns. No failed-run PnL accepted.

Clarification of2086 before costed results: fixed20 completed-day prior-high breakout,10-day prior-low close exit,14-day simple true-range mean ATR. Daily indicators shift1; hourly construction uses rolling480h high/240h low,14 complete reconstructed UTC daily ATR mapped only after day completes. Entry next actual open. Stop2ATR and target6ATR measured from entry open; max20 calendar days. Baseline is frozen fixed-stop/fixed-target diagnostic without partial exits, not a live policy: current winner-trailing policy would need separate validation if candidate otherwise qualifies. At each cadence evaluate stop/target from prior close, execute next actual open, recording overshoots, rather than claiming intrabar fills. Compare supplied daily against reconstructed daily before inferring robustness. Fees25bp/side, slippage10bp base20bp stress, train2024-25/holdout2026Jan1-Sep10 reused by broader research. No optimization; neighbors10/30days only sensitivity, not selection. Daily/hourly negative or unverified execution means reject. No hourly history invented from daily. Initial all-symbol hourly pagination exceeded40page cap; now requesting symbol-year bounded slices and auditing counts/timestamps.

New discovery thread: ACP August canal time higher, lower arrivals; LPG carrier may gain voyage demand but pay auctions. Before bars inspect fixed rule: LPG close above prior20 closes and SMA50; next-session open buy, initial stop2*prior20 mean high-low, target3R, 20-session time exit; daily conservative stop-first if both hit, gap exits at open; full position exit at +2R is NOT used. At +2R trailing handling not modeled, therefore passing daily result still needs15min implementation before live. Training2021-2024, untouched2025-Sep10 2026 holdout; fees5bp/slippage20bp eachside and stress50bp, lookback10/20/40 sensitivity all reported, no winner-selection from holdout. vectorbt for portfolios, CLI diagnostic required. No automatic order; cash278.67, normalrisk25.5297 max, position212.7475. Reject on negative train/holdout or missing execution evidence; do not retune to recent Panama news.

Use latest60 completed adjusted SIP daily log returns throughSep10, starting Sep10close3.84; simulate2 sessions throughSep14, 26 quarterhour increments/session, GBM zero logdrift and sample logdrift separately, sigma sample daily std/sqrt26,50000paths seed2026091100. Benchmark NEW hypothetical stop=3.84-.11124,target=3.84+3*.11124, all exits Sep14close; fee5bp/slippage25bp per side conservative stress from prior1928. This is model sensitivity, NOT backtest, executable overnightstop, calibrated event probability or paper retune. Preserve originalpaper3.73/3.61876/3.89686. Empirical bootstrap two daily logreturns is independent terminal-distribution diagnostic, not intraday barrier probability. Need prior20 ADR/ATR14/SMA20/50/RSI14/60dayvol and recent volume. Reject live even attractive toy paths absent event-validated >=3R rule.

Prior: rejected compression does not imply all crypto is untradeable; keep monthly shortfall outside inference. Three-level question tree, least predictable first: 1 Can evidence overturn cash? 1.1 Execution-valid new edge? 1.1.1 Does fixed 20-day Donchian breakout /10-day low exit with 2 daily ATR stop, 6 ATR target, no partial exits, max20 calendar days survive costed daily and hourly tests? 1.1.2 Does stress erase returns? 1.2 Generalization? 1.2.1 Fit no parameters; 2024-2025 training and 2026 Jan-Sep10 validation, acknowledge previously reused 2026 market history. 1.2.2 How many independent trades and correlated assets? 2 What changed tonight? 2.1 Liquid recovery? 2.1.1 Authenticated payouts versus dust? 2.1.2 Independent explorer and operator changes? 2.2 Leverage/liquidity? 2.2.1 Funding percentiles versus snapshots? 2.2.2 Coin OI versus USD-price effects? 3 What structures transmit events? 3.1 Stablecoin liabilities? 3.1.1 Issuance net of redemption and reporting period? 3.1.2 Native supply versus migration and asset buying? 3.2 Macro/protocol catalysts? 3.2.1 CPI date versus measured surprise? 3.2.2 Arc adoption versus ETH/SOL cash capture? 4 Can the fund execute? 4.1 Quotes? 4.1.1 Fresh timestamps and top depth? 4.1.2 Fees and 15-minute stop gaps? 4.2 Continuity? 4.2.1 Actual END receipts and cadence? 4.2.2 Numeric cutoff versus remaining runs? Resolution: completed UTC daily bars and completed hourly bars; intrabar ordering treated conservatively, no trading from headlines or unverified flow. Start $235.86 cash, goal positive executable expectancy, binding constraint validation. Plan allowances 6+9+17+7+5=44min; hard stop00:44:32UTC.

Prior: existing gap rule fails; AVAV positive EAR may drift but N9 insufficient. Set funding pressure aside. Three-level question tree, least predictable first: A Executable alternative? A1 long-horizon trend -> lookback fixed literature, next-open timing, costs; A2 payoff -> 3R target, volatility stop, gaps; A3 generality -> US equities, international equities, bonds/gold, crypto applicability. B Thesis attack? B1 AVAV -> cash conversion, valuation/rerating, frozen signal corrections; B2 TXNM -> original responses, concessions, approval vs financing; B3 FTK -> paid receipts, restart, working capital. C Evidence integrity? C1 vendor -> adjustments, missing bars, delayed cutoff; C2 inference -> training/holdout, overlapping outcomes, selection; C3 operations -> actual collector END, cadence, monitor branches. D Decision? D1 sizing -> empirical p/W/L, halfKelly, fund/venue caps; D2 adverse chain -> first cash impact, second financing, third equity; D3 uncertainty -> missing originals, out-of-sample failure, change evidence. Freeze exploratory alternative before data inspection: liquid ETF SPY,IWM,EFA,EEM,TLT,IEF,GLD,DBC, VNQ long-only 200-session SMA cross, next-open execution; entry stop 2*20-session ATR, target 6*ATR (3R), exit next open after close below SMA or 126 sessions. Train 2016-2022, holdout 2023-2026-09-10; 150/250 lookbacks sensitivity only, never choose winner. Fees1bp/slippage5bp each way; stress15bp slippage. Daily OHLC pessimistic stop-first if both touched; cannot certify 15-minute venue exits. Faber monthly rule read first, adapter explicitly not exact replication. No reuse of holdout for live optimization.