Stats

Numbered premises: 1. Supply is flat with278.67cash, no open/pending lots; its two operator-liquidated gains contain no natural loss sample and do not establish an edge. 2. Exchange-directory screen requested13,155 non-test listings and obtained12,576snapshots,12,285 price pairs,3,012 fresh liquid nonETF rows. Foreign37selected names and8crypto requests/5returns broadened discovery; incomplete nonstandard/OTC/global coverage is explicit. 3. Primary sources show genuine delivered scarcity economics for AXT,power suppliers and Maersk, alongside export permissions, supplier concentration, cost pass-through, financing and commissioning risks. Generic MLCC quotes do not prove accelerating shortage; IREN contracted ARR and Render cumulative burn are not realized profits or net token deflation. 4. Fixed net range tests83-88 reject AXTI,AEHR,VSH,ETN rules; POWL WATCH on weak training, PSIX WATCH with4trainingbars and selection-sensitive confidence bounds. All six current entry signals false. 5. No validated foreign/crypto mandate-specific entry or venue-compatible whole-share/15minute/+1R/dividend exit distribution; conservative quarter-Kelly authorization0. Conclusion no PROPOSE/EXIT/ADJUST/FUND; retaincash. Deduction valid under numerical-edge requirement; soundness bounded by selected historical data, primary-source incentives, lack of current distributor quote series and actual execution tests. Strongest counterargument: PSIX laterperiod +159.63%,29trades,p55.17%,avgwin10.715%,avgloss4.820%; data gap, selection and absent signal prevent deployment, rather than proving future losses. Existing tanker rejection/watch unchanged, no retuning. New daily research collector stores six candidates, eligible0; no pending event requires exposure before next04UTC run. ExistingFROmonitor11 retained. Sources and premises in observations313-335,467/469,516/517; analyses83-88,94-97,103-105,110. Primary URLs: https://www.sec.gov/Archives/edgar/data/1051627/000143774926025061/ex_974537.htm ; https://investors.psiengines.com/node/13891 ; https://www.maersk.com/news/articles/2026/08/13/maersk-delivers-strong-q2-and-raises-full-year-guidance ; https://www.sec.gov/Archives/edgar/data/1878848/000187884826000051/irenreportsfy26results.htm ; marketdata https://runway.adityaperswal.workers.dev/data/stock-bars . All measured screen rows and code archived373parts with exact read-back verification. Gaps named, sources no longer change cash decision; final five-minute phase reserved and used early, total run well below45minutes.

Required CLI executed: node /app/agent/dist/backtest.js --symbols AXTI,AEHR,PSIX,VSH,POWL,ETN --strategy breakout --lookback 20 --days 730 --stop-pct 3 --target-pct 6. AXTI return9.493%,DD24.977%,win35.849%,N53; AEHR21.806%,DD18.129%,win40.909%,N44; PSIX-22.307%,DD30.486%,win26.829%,N41; VSH35.552%,DD8.935%,win44.737%,N38; POWL-14.101%,DD23.071%,win30.952%,N42; ETN17.585%,DD12.792%,win47.222%,N36. Baseline same-close fills, exact percentage barriers, no fees/slippage or holdout, IEX source no explicit corporate-action adjustment; cannot trust as executable validation. Different rule from fixed next-open trend/range tests83-88, so differences do not isolate any one bias. Price ranking over all returned listings is a discovery tool, not a historical strategy with point-in-time constituents. No adopted portfolio screener or FUND proposal. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; cli.json, engine source and per-symbol observations. Positive frictionless outputs cannot override rejected or unvalidated cost-adjusted results.

Start state278.67cash, no open or queued holdings; goalrepeatable net profit. Binding constraint is independently validated executable edge, not access to trading hours. Premise1: scarce powered racks can benefit IREN, but contracted4bn ARR versus1bn operating ARR is a commissioning gap, not4bn current annual recognized revenue. FY26netloss702.6m includes638.8m impairments; equipment suppliers can raise prices without corresponding customer pass-through and delayed deployment can reduce payments. First order: GPU/power/cooling costs and9%financing consume future rental revenue; second: delivery slippage or customerdefault stretches cash and converts contracts into penalties/credits; third: new capacity or fasterhardware obsolescence can erode returns before debt retires. Strong counterargument: prepayments and6%Microsoft-backedfinancing cover most GPUcapital; still not all site costs or utilization. Illustrative fully drawn interest3.6bn*6%+2.4bn*9%=432m/year, not reportedforecast. Source https://www.sec.gov/Archives/edgar/data/1878848/000187884826000051/irenreportsfy26results.htm . Premise2: Render fiat-priced jobs burn tokens while scheduled emissions reward operators. First: users buy compute; second: operators' reward/cost balance changes supply; third: token returns depend on matched burns/emissions and valuation rather than GPUscarcity alone. Cumulative1.532m burn andYear2emissions5.906m coverdifferentperiods; reject false deflation ratio. Sources https://know.rendernetwork.com/basics/burn-mint-equilibrium ; https://stats.renderfoundation.com/ . Premise3: Maersk realized higherfreight rates benefits scarce vesselspace but shippers pay; then inventory workingcapital and productmargins suffer; then rerouting/newvessels/demanddestruction can normalize rent. Supplier ETN/PSIX similarly may both benefit downstream and pay upstream constraints. Countercase sustained demand and successful commissioning can sustain margins, but no stockpayoff follows logically without valuation and tested timing. Source https://www.maersk.com/news/articles/2026/08/13/maersk-delivers-strong-q2-and-raises-full-year-guidance . Facts are datedissuerdisclosures; chains are hypotheses, not forecasts. No distinct newfund justified: physicalcompute fitsSupply and majors belongexistingCrypto.

1. DigiKey general-purpose MLCC control GRM033R71A472JA01D shows17-week factory lead and15,177units on shelf at same time. A second nonstocked1.8pFpart also shows17weeks. Factory replenishment, current inventory and customer promised delivery are different measures; cached dates differ and no time series permits shortage acceleration inference. Generic controls do not disprove AI high-cap MLCC shortage. Sources https://www.digikey.com/en/products/detail/murata-electronics/GRM033R71A472JA01D/10705945 ; https://www.digikey.com/en/help-support/place-an-order/product-lead-time . 2. HPS explicitly cites previous pricing, custommix and operations, withGM31.5%,backlog+96.9%; adjustedEPS2.76 vsGAAP.79 requires reconciliation. Siemens Energy Q3 orders17.9bnEUR/backlog162bn, customer advances support cash; MarchsoldoutFY28 statement is older than Q3, not new slot count. Sources https://emea.hammondpowersolutions.com/news/2026/july/Quarter-2-2026-Financial-Results ; https://www.siemensenergy.com/global/en/home/investor-relations.html . 3. Maersk realized loaded freight rate2746USD/FFE,+22%YoY, much stronger evidence than an announced GRI.44%longterm rateproductshare and annualEBIT sensitivity700m per100USD/FFE imply3.64%rate change equals12.73%FYmidpoint EBIT, all else equal; not a predicted shock. LA portJuly960464TEU and capacity readiness do not contradict carrier congestion in other regions. Source https://www.maersk.com/news/articles/2026/08/13/maersk-delivers-strong-q2-and-raises-full-year-guidance ; https://portoflosangeles.org/references/2026-news-releases/news_081826_july_cargo . 4. MediaTek is sole foreign close-breakout among37selected;Julyrevenue+12.16% butYTD+.84%; issuer2bnAIASICtarget not realized companywide scarcity rent. Source https://www.mediatek.com/investor-relations/financial-information . Conclusion watchMaersk/HPS/ENR economics, reject blanketMLCCshortage claim, noforeignorder without venue and netexecution validation. No fair-value estimate or comprehensive non-US census claimed; missing part-level current quotes and actual vessel wait distributions explicitly retained.

1. AXT Q2 margins44.9% and revenue47.6m demonstrate delivered economics, but 10-K flags InP export permits and limited quartz/polishing suppliers. Manufacturing in China plus raw-material stakes does not remove export risk. At61.62, price/(4*.17Q2EPS)=90.62x is a peak-quarter annualization, not normalized P/E; net interest accounts for31.25% pretax income and diluted weighted-average shares grew45.22%YoY. Counterargument: genuine scale gains may outrun dilution; require shipment permissions, recurring margin and capacity evidence. Sources https://www.sec.gov/Archives/edgar/data/1051627/000143774926025061/ex_974537.htm ; https://www.sec.gov/Archives/edgar/data/1051627/000143774926008612/axti20251231_10k.htm . 2. Aehr's FY27 revenue130-150m guidance/backlogabout100m offers growth, yet top5 customers70%,topthree26/14/11%; pilot evaluations not production awards. Source https://www.sec.gov/Archives/edgar/data/1040470/000165495426006919/aehr_10k.htm . 3. PSIX data-center demand coexists with sales-21%YoY and persistent Wisconsin ramp costs; margin improved sequentially but not YoY. Price/(4*Q2EPS)=13.88x does not normalize cyclicality. Source https://investors.psiengines.com/node/13891 ; supplier risk https://www.sec.gov/Archives/edgar/data/1137091/000162828026013207/psix-20251231.htm . 4. Vishay passive bookbill1.40/backlog6.1months indicates demand, but30m tariff refunds lower revenue denominator without raising gross profit. POWL bookbill2.994 includesoneDCorder>42.83%bookings; EPSsplitadjusted,price/quarterannualEPS31.90x not normalized. ETN price/annualadjustedguide midpoint30.43x versus39.27xGAAP, single-source inputs and pass-through uncertainty remain. Sources https://ir.vishay.com/news-releases/news-release-details/vishay-intertechnology-reports-second-quarter-2026-results ; https://powellindustriesinc.gcs-web.com/news-releases/news-release-details/powell-industries-announces-third-quarter-fiscal-2026-results ; https://www.eaton.com/us/en-us/company/news-insights/news-releases/2026/eaton-reports-record-second-quarter-2026-results.html ; https://www.sec.gov/Archives/edgar/data/1551182/000155118226000007/etn-20251231.htm . Conclusion sound as a risk map, not a fair-value or surprise forecast. Issuers interested; no independent customer-level delivery data or normalized DCF established. Watch fundamentals while tested rules/no signals prohibit entries.

Universe first: Nasdaq Trader nasdaqlisted.txt plus otherlisted.txt, file creation Sep4 21:31, 13,155 non-test unique securities including ETFs. Requested every listing; nonstandard symbols caused whole-batch HTTP400 failures. Retrying missing plain-letter symbols raised coverage to12,576 snapshot entries,12,285 usable positive price pairs. 540 nonstandard symbols omitted from retry and579 total without snapshots; exchange-listing coverage is95.6%, not complete all-US/OTC coverage. Filters price>=5, IEX day dollar volume>=100,000, ETF=N, exactly Sep4 daily/Sep3 previous timestamps leave3,012 rows. One-day return p5/median/p95=-3.132/.0489/4.140%,1,543 positive. Original1,392 subset superseded; stalePDEX comparison removed. Read highest positive/negative names across the whole returned universe, not just holdings: NX already-reacted earnings/pricing recovery, AOUT/biotechs outside demonstrated bottleneck thesis; AXTI/AEHR/PSIX chosen for primary supply exposure, VSH/POWL/ETN added as fundamental comparators. No causal inference from ranking. Foreign37 selected physical supply names across Japan,Korea,Taiwan,Europe,Canada,Australia,HongKong; only2454.TW prior20-close breakout, although several series endSep3 and require freshness check. This is targeted foreign coverage, not all foreign listings; AUR.AX speculative penny row not adopted. Crypto8 requested,5 returned (BTC,ETH,SOL,FIL,RENDER), none above prior20-close high. MissingFET/AR/AKT not proof unavailable for trading; no complete crypto census claimed. Marketdata is partial venue, Yahoo indicative foreign history, not primary exchange quotes. Screening rule exploratory, no validated portfolio alpha or foreign execution eligibility. Sources https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt and otherlisted.txt ; https://runway.adityaperswal.workers.dev/data/stock-snapshots ; https://query1.finance.yahoo.com/v8/finance/chart/ ; https://runway.adityaperswal.workers.dev/data/crypto-bars . Every measured screen row and source code preserved in373 numbered compressed observation parts, hash in note; primary findings individually recorded.

NumPy10000paths,seed905,circular3-trade blocks,10%ofchangingequitypertrade,eachsymbol'snaturalexitcount40/39/29/38/34/32,compounded aftermodeledcosts. PSIX29-trade terminalp5/median/p95+2.827/+11.277/+20.801%,lossfreq1.21%,p95trade-stepDD3.869%;POWL34trades-2.060/+2.596/+7.829%,loss18.22%;AXTI40trades-8.948/-.356/+9.197%,loss52.57%;AEHR39tradesloss53.15%;ETN32tradesloss93.78%. These are empirical resampling frequencies conditional on selectedpasttrades, not nexttrade/monthend probabilities; no intratrade gap tails, waiting time or crossasset dependence. Quarter-Kellycapital plugin .25*(p/L-q/W) givesPSIX181.55%,POWL101.28%,VSH45.53%,negativesAXTI/AEHR/ETN; unstable means cause absurd caps. Venue25%cap would bind but eligibility binds first: no positive signals, PSIXno training, missing15minute/+1Ractualexits, multiple-selection uncertainty. Actualrisk0. At referenceADR, nominalnotionalriskceilingsAXTI73.45/AEHR79.83/PSIX212.77/VSH198.19/POWL212.77/ETN212.77 beforewhole-sharerounding; stops cannot guarantee1%realizedloss. Startcash278.67,goalrepeatablenetreturns,bindingconstraintindependentimplementableedge. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; exact stats for allsix archive. Monthly348.92shortfall does notaltermodel.

Premise1:419complete common daily return rows2025-Sep4 from all-adjustedIEX; OLS versusSPY beta AXTI2.511,AEHR3.447,PSIX2.892,VSH2.387,POWL1.899,ETN1.539. POWL/ETNcorr.6593,AEHR/VSH.5628; distinct products do not eliminate shared market exposure. Intercepts descriptive, not causalalpha; no annualized alpha claim. Premise2: 10000 circular3-trade block mean bootstrapseed905 on naturalexit samples40/39/29/38/34/32; five means have95%interval crossingzero. PSIXmean3.7507%,CI[.4832,7.1587]%, but only4trainingbars; this is selected postlisting history. Additional20000draws conservative six-test tail sensitivity[.025/6,1-.025/6] gives[-.6271,8.3548]%; cannot correct all3012-name screening/unknown prior research and is not a formal population guarantee. RemovebestPSIXtrade mean2.5700%; not entirely one outcome, still smallsampleandregime risk. Premise3: contemporary survivorship, selected highmovers, assumed costs, conditionalbarrierfills and missing natural ledger losses undermine transportability. Conclusion PSIX merits prospective collection, not size; others rejected/watch perbacktests83-88. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; stats.json,tests.json,derived.json/trades archived. Prior on PSIX shifts fromunknown tointerestingdiagnostic, never adopted; genuinely independent netexecution outcomes would change verdict.

1. Completed Sep4 IEX closes AXTI61.62,AEHR86.245,PSIX40.54,VSH31.78,POWL181.22,ETN410.805; all below prior20-close maxima96.025,145.555,43.05,35.53,218.9,459.85. PSIX alone above SMA50 among these six. 2. AXTI/AEHR one-day rises9.59%/13.13% coexist with20-session losses30.41%/16.27%; rebound is not the registered close-breakout+SMA50 conjunction. Indicators share price input, not independent confirmation. 3. ADR20 respective7.14025,9.19425,1.50575,1.36475,6.8655,10.81175 supplies bounded reference moves, not validated support. PSIX reference target43.5515 is2ADR from40.54; prior20high43.05 constrains upside; actual entry absent. Whole-share affordability under212.77cap3,2,5,6,1,0 respectively BEFORE risklimit; one AXTIshare risks7.14 at nominalADR, oneAEHRshare9.19 exceeds8.5108equityrisk. ETNrequiresfractional, restricting extendedhours exits. No real positions to adjust. Conclusion noentry signal; negative/unstable/missing-training backtests independently prohibit risk. Price/volume partialIEX and dividend/15minute enforcement gaps remain. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; technical.json and per-symbolobservations archived.

Premises: (1) protocol note127 fixed prior20-close breakout AND SMA50, next-open entry, frozen signal-day ADR20 stop1/target2, timeout after5sessions at nextopen, longonly. (2) vectorbt explicit two event rows per daily bar represent open and later barrier/close, including entry-day exits; synthetic timestamps do NOT claim intraday observations. Adverse gaps fill open, favorable target gaps conservatively target, dual touches stopfirst. Fees5bp/slippage10bp eachside; stress doublesboth. Independent cash/share values reconcile maxerror 3.637978807091713e-12. (3) training2022-24, diagnostic chronological laterperiod2025-Sep4; all-adjusted IEX, paginated. REJECT: training and later-period net returns negative. Training sweep targets1.5/2 stops1/1.25: {"(1.5, 1)":-12.752220910464256,"(1.5, 1.25)":-6.8086762427363094,"(2, 1)":-10.708581864755178,"(2, 1.25)":-12.052126108536413}. All contemporary selected names have survivorship/selection bias; no pristine prospective evidence claimed. Terminal open marks excluded from payoff sample; currentholdoutopen0,trainingopen0. Initial entry-day omission detected and corrected before journal metrics; discarded output not used. No cash-dividend,15minute monitoring,partialfills,+1Rreduce/trail or whole-share execution model. Daily target touches may not execute under venue timing. Thus not live eligible, regardless of favorable historical result. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; executable test.py and full trades stored measured-dataset archive. Current signal false.

Premises: (1) protocol note127 fixed prior20-close breakout AND SMA50, next-open entry, frozen signal-day ADR20 stop1/target2, timeout after5sessions at nextopen, longonly. (2) vectorbt explicit two event rows per daily bar represent open and later barrier/close, including entry-day exits; synthetic timestamps do NOT claim intraday observations. Adverse gaps fill open, favorable target gaps conservatively target, dual touches stopfirst. Fees5bp/slippage10bp eachside; stress doublesboth. Independent cash/share values reconcile maxerror 3.637978807091713e-12. (3) training2022-24, diagnostic chronological laterperiod2025-Sep4; all-adjusted IEX, paginated. WATCH: positive selected later period, all four training variants lose and mean interval crosses zero. Training sweep targets1.5/2 stops1/1.25: {"(1.5, 1)":-6.741325252297993,"(1.5, 1.25)":-18.69843938469283,"(2, 1)":-3.852527187305641,"(2, 1.25)":-21.788176286351028}. All contemporary selected names have survivorship/selection bias; no pristine prospective evidence claimed. Terminal open marks excluded from payoff sample; currentholdoutopen0,trainingopen0. Initial entry-day omission detected and corrected before journal metrics; discarded output not used. No cash-dividend,15minute monitoring,partialfills,+1Rreduce/trail or whole-share execution model. Daily target touches may not execute under venue timing. Thus not live eligible, regardless of favorable historical result. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; executable test.py and full trades stored measured-dataset archive. Current signal false.

Premises: (1) protocol note127 fixed prior20-close breakout AND SMA50, next-open entry, frozen signal-day ADR20 stop1/target2, timeout after5sessions at nextopen, longonly. (2) vectorbt explicit two event rows per daily bar represent open and later barrier/close, including entry-day exits; synthetic timestamps do NOT claim intraday observations. Adverse gaps fill open, favorable target gaps conservatively target, dual touches stopfirst. Fees5bp/slippage10bp eachside; stress doublesboth. Independent cash/share values reconcile maxerror 3.637978807091713e-12. (3) training2022-24, diagnostic chronological laterperiod2025-Sep4; all-adjusted IEX, paginated. REJECT: all four training variants lose; double costs turn holdout negative. Training sweep targets1.5/2 stops1/1.25: {"(1.5, 1)":-13.635595520075913,"(1.5, 1.25)":-19.17348293114387,"(2, 1)":-3.52923149049484,"(2, 1.25)":-8.73000753401997}. All contemporary selected names have survivorship/selection bias; no pristine prospective evidence claimed. Terminal open marks excluded from payoff sample; currentholdoutopen0,trainingopen0. Initial entry-day omission detected and corrected before journal metrics; discarded output not used. No cash-dividend,15minute monitoring,partialfills,+1Rreduce/trail or whole-share execution model. Daily target touches may not execute under venue timing. Thus not live eligible, regardless of favorable historical result. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; executable test.py and full trades stored measured-dataset archive. Current signal false.

Premises: (1) protocol note127 fixed prior20-close breakout AND SMA50, next-open entry, frozen signal-day ADR20 stop1/target2, timeout after5sessions at nextopen, longonly. (2) vectorbt explicit two event rows per daily bar represent open and later barrier/close, including entry-day exits; synthetic timestamps do NOT claim intraday observations. Adverse gaps fill open, favorable target gaps conservatively target, dual touches stopfirst. Fees5bp/slippage10bp eachside; stress doublesboth. Independent cash/share values reconcile maxerror 7.275957614183426e-12. (3) training2022-24, diagnostic chronological laterperiod2025-Sep4; all-adjusted IEX, paginated. WATCH: only four pre-2025 IEX bars and identical SIP coverage, hence no training sample. The nominal holdout is a selected post-listing diagnostic, not independent OOS validation. Training sweep targets1.5/2 stops1/1.25: {"(1.5, 1)":0,"(1.5, 1.25)":0,"(2, 1)":0,"(2, 1.25)":0}. All contemporary selected names have survivorship/selection bias; no pristine prospective evidence claimed. Terminal open marks excluded from payoff sample; currentholdoutopen0,trainingopen0. Initial entry-day omission detected and corrected before journal metrics; discarded output not used. No cash-dividend,15minute monitoring,partialfills,+1Rreduce/trail or whole-share execution model. Daily target touches may not execute under venue timing. Thus not live eligible, regardless of favorable historical result. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; executable test.py and full trades stored measured-dataset archive. Current signal false.

Premises: (1) protocol note127 fixed prior20-close breakout AND SMA50, next-open entry, frozen signal-day ADR20 stop1/target2, timeout after5sessions at nextopen, longonly. (2) vectorbt explicit two event rows per daily bar represent open and later barrier/close, including entry-day exits; synthetic timestamps do NOT claim intraday observations. Adverse gaps fill open, favorable target gaps conservatively target, dual touches stopfirst. Fees5bp/slippage10bp eachside; stress doublesboth. Independent cash/share values reconcile maxerror 3.637978807091713e-12. (3) training2022-24, diagnostic chronological laterperiod2025-Sep4; all-adjusted IEX, paginated. REJECT: training and later-period net returns negative. Training sweep targets1.5/2 stops1/1.25: {"(1.5, 1)":-38.643851939335995,"(1.5, 1.25)":-57.67302561057384,"(2, 1)":-32.86874570103676,"(2, 1.25)":-46.090402422060095}. All contemporary selected names have survivorship/selection bias; no pristine prospective evidence claimed. Terminal open marks excluded from payoff sample; currentholdoutopen0,trainingopen1. Initial entry-day omission detected and corrected before journal metrics; discarded output not used. No cash-dividend,15minute monitoring,partialfills,+1Rreduce/trail or whole-share execution model. Daily target touches may not execute under venue timing. Thus not live eligible, regardless of favorable historical result. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; executable test.py and full trades stored measured-dataset archive. Current signal false.

Premises: (1) protocol note127 fixed prior20-close breakout AND SMA50, next-open entry, frozen signal-day ADR20 stop1/target2, timeout after5sessions at nextopen, longonly. (2) vectorbt explicit two event rows per daily bar represent open and later barrier/close, including entry-day exits; synthetic timestamps do NOT claim intraday observations. Adverse gaps fill open, favorable target gaps conservatively target, dual touches stopfirst. Fees5bp/slippage10bp eachside; stress doublesboth. Independent cash/share values reconcile maxerror 3.637978807091713e-12. (3) training2022-24, diagnostic chronological laterperiod2025-Sep4; all-adjusted IEX, paginated. REJECT: training and later-period net returns negative. Training sweep targets1.5/2 stops1/1.25: {"(1.5, 1)":-31.442725836626114,"(1.5, 1.25)":-35.08498901772668,"(2, 1)":-34.23281713352532,"(2, 1.25)":-38.59042367158891}. All contemporary selected names have survivorship/selection bias; no pristine prospective evidence claimed. Terminal open marks excluded from payoff sample; currentholdoutopen0,trainingopen0. Initial entry-day omission detected and corrected before journal metrics; discarded output not used. No cash-dividend,15minute monitoring,partialfills,+1Rreduce/trail or whole-share execution model. Daily target touches may not execute under venue timing. Thus not live eligible, regardless of favorable historical result. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; executable test.py and full trades stored measured-dataset archive. Current signal false.

Start state flat Supply book278.67 cash, goal repeatable net gain, binding constraint robust executable edge rather than order placement. Premise1: live ledger has no holdings/queued orders, so no stops/targets require adjustment; two forced profitable exits do not supply natural losses. Premise2: issuer bear review28 and broader fundamental31 identify contractual upside caps, ballast/volume/counterparty risks, and upstream concentration; no fresh primary fact proves an unpriced stock trade. Premise3: preregistered fixed range-exit tests36/37 are now reproducible: FRO positive historical OOS but mean CI includes0 and training sensitivity unstable; STNG loses34.71% training and doubled-cost OOS approximately0 despite positive baseline OOS. Premise4: statistics46, simulation47 and technical48 distinguish genuine Sep4 signals from eligibility; cash-dividend/raw-price exits,15-minute execution, independent prospective data and consolidated coverage remain gaps. Conclusion: no PROPOSE/EXIT/ADJUST/FUND. Validity follows mandated robust edge requirement; soundness bounded by selected histories and data/implementation limits, not proof of negative future returns. Strongest counterargument is profitable OOS plus real issuer pricing power; rejected as sufficient sizing evidence because confidence bounds cross0 and adjusted-price implementation can change actual stops. Quarter-Kelly plug-ins would hit venue cap but are unstable; conservative edge and actual allocation0. No fundamental bearish assertion is used to short. Existing arrangement favors suppliers/owners able to fulfill scarce capacity; contracts transfer upside/volume risk to customers, and normalization can change rents. Rate-to-total example STNG three charter schedules116.7M over3years before offhire/commissions illustrates gross contract revenue, not net earnings. Sources: observations74-81,102-104,110-111,118,198-200,239-240 with original URLs; primary price https://runway.adityaperswal.workers.dev/data/stock-bars and issuer https://www.frontlineplc.cy/fro-second-quarter-and-six-months-2026-results/ . Archives97-99 preserve code; CLI53 is rejected baseline. Job12 daily30runs saves prospective bars, never executes; monitor11 checks before expected Sep18 dividend. Sources stopped changing conclusion; no thread exceeded allowance, original old-artifact recovery closed with named absence. Research finished early within44-minute ceiling and final recording phase. Other funds' monitors/trades left to owners.

Executed node /app/agent/dist/backtest.js --symbols FRO,STNG --strategy breakout --lookback20 --days730 --stop-pct3 --target-pct6 (actual flag uses space: --lookback 20). Fixed diagnostic baseline; not strategy selection or fresh independent holdout. FRO return3.57694%,DD18.75681%,win33.3333%,N51; STNG16.44747%,DD11.37013%,win41.1765%,N34;502 bars each. Mean trade0.11895%/0.50102%,buyhold107.32255%/20.49312%. Source inspection /app/agent/dist/backtest/engine.js and strategies.js proves same-signal-close entry, exact stop/target fills, no fees/slippage; bars.js uses IEX and does not follow next_page_token or specify adjustments. Therefore reject CLI as executable edge. Corrected independent implementation/backtests36/37 use explicit next-open fills and costs, with training/holdout separation. Results differ in rule, period, accounting and costs; cannot attribute all difference to timing alone. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; artifacts/cli.json. Library runtime vectorbt1.1.0,numpy2.4.6,pandas3.0.5. No trade follows CLI.

Sep4 completed IEX daily bars: FRO46.15 above prior20 closes45.43 and SMA20/50/20042.850/39.820/33.599; STNG82.345 above81.46 and78.463/77.164/71.137. Both trend+breakout indicators align, but they are correlated transformations of price, not independent corroboration of fundamentals. Day moves+1.607%/+2.867%,20-session moves+16.188%/+8.185%. ADR20=1.23575/1.67, ATR14=1.55321/1.91107. Diagnostic2-ADR targets at these closing reference entries would be48.6215/85.685;1-ADR stops44.91425/80.675, not orders or support-level claims. Actual next-session opening entry, tick rounding, dividend and15-minute execution need revalidation. FRO planned dividend2.61 exceeds two recent daily ranges, making mechanical raw-price exits particularly unsafe across ex-date; NYSE expectedSep18 per issuer attachment, monitor11 set for prior close review. Flat book means no current stop/target reset. Maximum single position212.77, fund278.67, nominal equity loss budget8.5108; price gap can exceed nominal stops. No validated entry permitted despite signals. Data source https://runway.adityaperswal.workers.dev/data/stock-bars ; dividend https://ml-eu.globenewswire.com/Resource/Download/59431810-8e9c-4a39-b3d0-cc2dd63da402 . IEX partial-market prices, not consolidated official closes; recent SIP request denied403. Research watch12 stores same split-adjusted close/range definitions, no automatic adoption.

Numpy simulation uses naturally completed net all-adjusted holdout returns from tests36/37;10000 paths circular blocks of3 trades, seed501, marks each outcome and compounds10% of changing equity per trade. No fabricated Gaussian drift. FRO41-trade path terminal return p5/median/p95=-2.729/2.062/7.332%;24.17% lose money,95th-percentile trade-step drawdown4.478%. STNG38-trade path=-2.168/1.400/5.000%;25.74% lose,p95DD3.592%. This is a hypothetical number-of-trades horizon, not a forecast through month-end; downtime, intratrade paths, new gaps and regime changes absent. Binary mean-payoff quarter-Kelly capital heuristic .25*(p/Wloss-(1-p)/Wwin) gives81.88% FRO,66.75% STNG; using .25*(p-q/(Wwin/Wloss)) instead measures loss-risk fraction, not notional. Neither unstable plug-in justifies capital, and venue cap25% would bind anyway. Lower confidence mean is negative, so conservative authorized size0. At hypothetical10% account allocation85.108 dollars, median compounded gains are too small/uncertain to justify chasing348.92 monthly shortfall. Premises: sampled history is best available diagnostic; sampled mean uncertainty spans0; model lacks implementation fidelity. Conclusion: zero proposal valid under required robust edge, not proof losses inevitable. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; raw/code artifacts and archive97-99. This is research simulation, no executed portfolio.

Premise1: all-adjusted daily returns Jan2025-Sep42026,420 paired rows, give FRO/STNG Pearson0.7506; OLS stock return on SPY plus intercept slopes0.6291/0.4025. Low market beta does not mean low shared tanker exposure. Premise2: corrected natural-exit samples41/38 trades, FRO mean0.53025%, STNG0.37899%, differ from two operator liquidations which cannot calibrate average loss. Premise3:10000 circular moving-block resamples of3 consecutive trades, seed501, yield95% percentile mean intervals FRO[-0.87436,1.95854]%, STNG[-0.75304,1.45383]%. Resample means positive76.17%/74.69%; these are resampling frequencies, not Bayesian probabilities that current trades win. Conclusion: evidence cannot robustly separate positive mean from zero; original two-ticker diversification is overstated. Valid inference only for selected histories; current two-name survivor selection, earlier historical inspection, small N, overlapping market exposures and possible regime shifts limit soundness. Bootstrap cannot invent missing tail events or audit dividend/15-minute fills. Prior neutral-to-cautious unchanged. Independent prospective net outcomes with a positive lower mean interval and actual execution/corporate-action handling would change eligibility; a lower bound<=0 keeps policy size0. No multiple-test corrected significance claimed. Source https://runway.adityaperswal.workers.dev/data/stock-bars retrieved Sep5; code archived notes97-99; observations110/111 superseded for FRO payoff estimates by198; backtests36/37 contain final figures.

Premise1: preregistered note90 before retrieving outcomes: previous20-close breakout, next open entry, fixed signal-day20-session mean(high-low), stop1x/target2x, exit after5 sessions at following open, no pyramiding. Premise2: vectorbt explicit LONGONLY orders, fees5bp/slippage10bp per side, cash compounding checked against manual returns; opening adverse gaps use open, dual touches stop first. Corrected terminal window marks are excluded from natural payoff samples. Premise3: training2022-2024, historical chronological holdout2025-01-01 through2026-09-04, all-adjusted IEX OHLC. Figures are percent, unlevered full-notional theoretical portfolio, not actual fund return. Training-only target1.5/2,stop1/1.25 sweep did not select parameters. All four all-adjusted training variants lose33.74% to36.69%; positive later holdout does not establish stable edge, and doubled-cost holdout is approximately flat. Source https://runway.adityaperswal.workers.dev/data/stock-bars retrieved Sep5;1172 bars, one missing IEX date2022-03-08 versus SPY. Artifacts/fetch.py,test.py,results.json and bars-split/all.json; code archived notes97-99 with hash. Prior historical reported tests could not be recovered; this is a fresh diagnostic. Previously viewed tanker history makes historical holdout exploratory rather than pristine; prospective collection starts Sep8. Limits: dividends modeled through adjusted-price proxy, not actual cash entitlement/raw exit levels; split-only comparison omits cash dividends. IEX partial coverage, SIP recent data403; no exact15-minute/overnight stop enforcement or whole-share model. Baseline costs assumed, not measured slippage. Mean-payoff intervals cross zero. Validity: insufficient robust implementation evidence implies zero eligible size; does not disprove every tanker strategy. No retuning after holdout. Current signals do not authorize entries.

Premise1: preregistered note90 before retrieving outcomes: previous20-close breakout, next open entry, fixed signal-day20-session mean(high-low), stop1x/target2x, exit after5 sessions at following open, no pyramiding. Premise2: vectorbt explicit LONGONLY orders, fees5bp/slippage10bp per side, cash compounding checked against manual returns; opening adverse gaps use open, dual touches stop first. Corrected terminal window marks are excluded from natural payoff samples. Premise3: training2022-2024, historical chronological holdout2025-01-01 through2026-09-04, all-adjusted IEX OHLC. Figures are percent, unlevered full-notional theoretical portfolio, not actual fund return. Training-only target1.5/2,stop1/1.25 sweep did not select parameters. All-adjusted training sensitivity returns range-22.85% to16.28%; selected default was fixed beforehand. Holdout positive but fragile to costs and statistical uncertainty. Source https://runway.adityaperswal.workers.dev/data/stock-bars retrieved Sep5;1172 bars, one missing IEX date2022-03-08 versus SPY. Artifacts/fetch.py,test.py,results.json and bars-split/all.json; code archived notes97-99 with hash. Prior historical reported tests could not be recovered; this is a fresh diagnostic. Previously viewed tanker history makes historical holdout exploratory rather than pristine; prospective collection starts Sep8. Limits: dividends modeled through adjusted-price proxy, not actual cash entitlement/raw exit levels; split-only comparison omits cash dividends. IEX partial coverage, SIP recent data403; no exact15-minute/overnight stop enforcement or whole-share model. Baseline costs assumed, not measured slippage. Mean-payoff intervals cross zero. Validity: insufficient robust implementation evidence implies zero eligible size; does not disprove every tanker strategy. No retuning after holdout. Current signals do not authorize entries.

Premise1: MU FY26Q3 10-Q explicitly reports allocation under supply constraints, but largest strategic contracts generally cap existing-product prices near calendarQ2; take-or-pay protects volume and floor economics, not unlimited upside. FY25 10-K adds single-source lithography risk and17% revenue from one unnamed customer. Strongest bear: customer capex slows while upstream equipment/capacity spending stays committed; counterexample is binding multi-year demand. Premise2: GEV Q2 electrification orders/revenue6347/3637 million imply1.745 book-to-bill, EBITDA margin18.4%; evidence of realized supplier economics. Its FY25 risk factors disclose constrained critical inputs and delivery penalties. Reject the overly strong bear that all backlog is freely cancelable: RPO excludes orders cancelable without substantive penalty. Premise3: Murata July31 forecast sees capacitor revenue+23.6%, consolidated+15.2%, capex+2.9%; server/distributor demand strength coexists with weakness elsewhere, so broad component-shortage claims lack part-level proof. Premise4: Hapag $600/container Conosur GRI takes effect Sep4, but announcement does not prove realized yield or port congestion. Sources respectively https://www.sec.gov/Archives/edgar/data/723125/000072312526000015/mu-20260528.htm ; https://www.sec.gov/Archives/edgar/data/723125/000072312525000028/mu-20250828.htm ; https://www.sec.gov/Archives/edgar/data/1996810/000199681026000147/gevpressrelease2q26.htm ; https://www.gevernova.com/sites/default/files/gevernova_2025_form10-k.pdf ; https://corporate.murata.com/en-us/ir/financial/forecast ; https://www.hapag-lloyd.com/zh/services-information/news/2026/07/shipping-with-our-conosur-service--con---a-gri-is-coming-up.headless.html . Read Sep5; issuer claims interested, filing/report dates differ, no contradiction from timing. Consequences inferred:1 scarce capacity transfers cash from buyers to qualified suppliers;2 caps/floors, cancellations and upstream shortages determine retained margin;3 capacity investment/substitution can normalize rents before volumes peak. Binding constraint is incremental unpriced evidence and a validated stock rule. Watch only, no new fund warranted; these are already within Supply mandate. Missing: part-level lead times, independent actual shipping flows, normalized valuation, non-US venue eligibility and OOS rules for new names. Legacy coverage not used as corroboration.

No open holdings or queued orders; no stops/targets to adjust. Strongest FRO bear: normalization releases vessel capacity, while prolonged disruption can destroy cargo and raise uninsured/off-hire costs; both directions can hurt the naive more-war-more-profit thesis. (1) Realized load-to-discharge revenue excludes ballast. (2) Lower effective earnings plus delivery commitments can reduce distributable cash. (3) Lower expected distributions and vessel values can compress equity before guidance. Counterexample: stockpile rebuilding/longer Atlantic routes and fixed charters could sustain earnings; this prevents a categorical short thesis. Newly read full attachment shows one charter explicitly stepping110k/70k/45k over three years, limiting that ship's upside without proving spot collapse. Strongest STNG bear: new fixed contracts trade upside for certainty; their two LR2 rates average41344/day versus64900 current spot/pool, a36.3% discount across different durations. (1) Locked revenue cushions spot weakness but caps those ships. (2) Product-volume contraction/counterparty risk can offset rerouting gains. (3) Fleet growth and lower marginal freight can reprice residual spot earnings. Counterexample: these are just2 of25 owned LR2s, not the entire fleet; guaranteed term income can increase value. Original diversification premise weakened by measured0.751 daily FRO/STNG correlation over420 OOS observations. Validity: these channels challenge monotonic pricing-power assumptions; they do not establish near-term negative returns. Soundness: issuer contracts direct, global orderbook estimates secondhand, no independent live cargo/war-insurance series. Sources: https://www.frontlineplc.cy/fro-second-quarter-and-six-months-2026-results/ ; https://ml-eu.globenewswire.com/Resource/Download/59431810-8e9c-4a39-b3d0-cc2dd63da402 ; https://www.sec.gov/Archives/edgar/data/913290/000162828026021774/fro-20251231.htm ; https://www.scorpiotankers.com/scorpio-tankers-inc-announces-updates-on-third-quarter-2026-tce-rates-and-time-charter-out-agreements/ ; https://runway.adityaperswal.workers.dev/data/stock-bars . Issuer dates Aug28/Sep3, filings FY2025; not new Sep4 announcements. What changed Sep4: both IEX closes crossed prior20 closing highs, but price strength does not resolve implementation/uncertainty. Start cash, goal repeatable returns, binding constraint robust tested payoff distribution.

Existing-research trading review. Premise 1: live notes/observations/analyses/lessons show #19 remains latest supply strategy, rejecting re-entry for absent reproducible cost-inclusive chronological OOS entry/exit calibration. No newer validated rule is returned. Premise 2: live trade ledger contains only two closed operator-liquidated lots; monitor #4 done, no pending/open exposure. These selected exogenous exits cannot estimate natural payoff distribution or validate stops/targets. Premise 3: issuer source reopened https://www.frontlineplc.cy/fro-second-quarter-and-six-months-2026-results/ retains strong booked rates and ballast-day qualification; repeat reading adds no independent evidence. Conclusion: retain $278.67 cash; no entry, exit or adjustment. Validity: missing mandated numerical edge implies zero eligible size. Soundness bounded by unreviewed historical raw artifacts and absence of a broad fresh-news scan; this does not prove tanker stocks will fall. Counterexample: strong rates could support further gains, but do not identify expected win/loss or attainable exits. Start flat, goal repeatable positive net returns, binding constraint audited implementation and OOS calibration. Quarter-Kelly undefined; policy notional zero. Qualitative consequence chain: freight scarcity can favor owners over charterers; ballast/utilization alters realized earnings; equity gains further depend on valuation and entry/exit timing. No fresh backtest, statistical fit or Monte Carlo run because no new rule is proposed and assumed payoff inputs would not repair missing evidence. No eligible signal or exposure needs a new monitor/job before September 5 00:00 UTC research. Sources: query_research/query_trades/list_monitors fund=supply September 4 21:18 UTC; supplied account snapshot; issuer URL above. Prior automatic-buy notes remain superseded.