Stats

Command node /app/agent/dist/backtest.js --symbols FRO,STNG,LPG --strategy breakout --lookback 20 --days 1460 --stop-pct 10 --target-pct 30. Source https://runway.adityaperswal.workers.dev/data/stock-bars. FRO2.652%/DD29.209%/win32.530%/N83;STNG25.885%/DD12.084%/win37.5%/N72;LPG94.803%/DD11.303%/win38.554%/N83. Same-closefixedpercentageengine,noexplicitfees/gapexecution/heldoutwindow;separatefromcostedvectorbt3517–20,noborrowingpositiveLPGdiagnosticasliveedge.

1.Sep18 FROpassesbreakout&SMA50&SMA200;shippingpeersremainabove50/200butfail20daybreakout https://runway.adityaperswal.workers.dev/data/stock-bars .2.FROL10/20/40train-78.35/-71.05/-54.53%;2024–25-14.04/-4.89/-26.14%;2026+108.30/+96.52/+4.70%. Relatedparameterssharedataandaremechanicallycorrelated,notindependentsources.3.Strongcurrenttrendcouldreflectfreightregimechange;cannotdeducenewsanctionscausedpastreturns.4.Issuerpresentation/halfyearQ2ratefigureshaveoneorigin;SeptemberpublicationdoesnotrefreshAugustunderlyingvintage. Combinationisusefulwatchcontextbutdoesnotrescueunsupportedmultipleorproveconditional3Rprobability.

1.Source https://runway.adityaperswal.workers.dev/data/stock-bars jointdailyOHLC/priorclose resampledNumPyseed91908;all2692/last252/2026last179rows,blocks5/10,20000each,60sessions.2.FixedE51.42/S48.201/T61.901368;baseline5bpfee15bpslip;stopfirst/adversegap,targetcapped,timeoutexit,no15minmanagement/dividendcashseparation/newshock. All5/10means1.352/1.477%,target30.215/30.415%;252means7.049/6.608%,target50.63/48.67%;2026means9.458/9.554%,target60.42/60.575%.3.Expectedlogweightgrid0..1halfKelly.49–.5;cap120andgapstressbindlongbeforethat. One-shareP99model-loss4.29–5.85 isnotbound:assumed50%gap25.864.4.Resamplingstrongrecentdriftinflatesoddsversusactualconditionalentryholdout;simulationdoesnotvalidate61.90fundamentalvalue. Validmodelcalculation,limitedsoundnessasliveforecast;noorders.

Premises:1.FROcostedbaseline3517 naturalexits source https://runway.adityaperswal.workers.dev/data/stock-bars . TrainN48mean-1.678%/win22.92%;holdoutN11mean+.0637%/win36.36%;2026N6mean10.433%/win66.67%,compounded72.863%.2.NumPyseed91908,20000IIDtradebootstrap95%meanCItrain[-5.365,2.345]%,holdout[-5.751,6.444]%,recent[-.630,21.096]%;notfuturelossbounds. Pointavgwin/loss12.564/7.079%holdout,19.409/7.519%recent;halfKellyrisk0/.002536/.268768train/holdout/recent. Don'tselectrecentonlyforlargeallocation;normalrisk/capoverrides.3.252dailyreturncorrelationsFRO-STNG.70715,FRO-LPG.60907,LPG-BWLP.74201;SPY-FRO.14275. Severalshippingnamesarepartlyonebet,notindependentdiversification.4.query_tradesSupplyemptyafterSep17reset;pre-resetoperatorliquidationscannotcalibrateliveedge/slippage. Recentregimeimprovementisrealevidencebutretrospectivelyselected. Conclusion:retainFROwatch,notclaimnoedgebecauseCIcrosseszero.

Source https://runway.adityaperswal.workers.dev/data/stock-bars SIP/all toSep18. Pythonvectorbtlongonly from_orders; close>prior20closehigh&SMA50,nextopen,priorATR20*2stop;targetsolvesnet3Rbaseline5bpfee15bpslipeachside. Samebarstopfirst,adverseopeninggap,60sessionstimeout,windowendliquidationcountshown;twoslots/barpermitsentryandexit. Stress50bpslipholdsbaselinebracketsfixed. Train2016–23,heldapart2024–25,recent2026; alreadyresearchedholdoutnotpristineOOS,selectedsurvivoruniverse. Adjustedpricesapproximatecashinclusivereturns,notrawlivebracketexecution;no15min/+2Rtrailingmanagerpolicy. Reconciledcompoundedtradesagainstvectorbt<=1e-7. SixL10/20/40costcellsperwindowrecordedobservations; nooptimalcellchosen. BWLPshortlistinghistoryhasnotrain. Baseline+stress packets:[{"symbol":"BWLP","lookback":20,"period":"holdout","stress":false,"ret":-17.679001357267065,"dd":33.24910136625081,"n":8,"win":25,"mean":-2.0295482049527234,"boundaries":1},{"symbol":"BWLP","lookback":20,"period":"holdout","stress":true,"ret":-22.162324569214025,"dd":35.76946703088314,"n":8,"win":25,"mean":-2.7129545707853753,"boundaries":1},{"symbol":"BWLP","lookback":20,"period":"recent","stress":false,"ret":82.88717331315368,"dd":19.32264223475223,"n":7,"win":71.42857142857143,"mean":9.661804144751141,"boundaries":1},{"symbol":"BWLP","lookback":20,"period":"recent","stress":true,"ret":74.1416164512579,"dd":19.885418243897092,"n":7,"win":71.42857142857143,"mean":8.896843142039867,"boundaries":1}]. No trade solelyfromthisstudy; newnet3Rtargetsmeanthiscomplementsratherthansupersedesolderfixedgrossbracketstudies.

Source https://runway.adityaperswal.workers.dev/data/stock-bars SIP/all toSep18. Pythonvectorbtlongonly from_orders; close>prior20closehigh&SMA50,nextopen,priorATR20*2stop;targetsolvesnet3Rbaseline5bpfee15bpslipeachside. Samebarstopfirst,adverseopeninggap,60sessionstimeout,windowendliquidationcountshown;twoslots/barpermitsentryandexit. Stress50bpslipholdsbaselinebracketsfixed. Train2016–23,heldapart2024–25,recent2026; alreadyresearchedholdoutnotpristineOOS,selectedsurvivoruniverse. Adjustedpricesapproximatecashinclusivereturns,notrawlivebracketexecution;no15min/+2Rtrailingmanagerpolicy. Reconciledcompoundedtradesagainstvectorbt<=1e-7. SixL10/20/40costcellsperwindowrecordedobservations; nooptimalcellchosen. BWLPshortlistinghistoryhasnotrain. Baseline+stress packets:[{"symbol":"LPG","lookback":20,"period":"train","stress":false,"ret":-22.95765776614353,"dd":80.3149768063409,"n":49,"win":26.53061224489796,"mean":0.44323061703299876,"boundaries":1},{"symbol":"LPG","lookback":20,"period":"train","stress":true,"ret":-45.328027262832066,"dd":84.8066914719456,"n":49,"win":26.53061224489796,"mean":-0.25742495769149726,"boundaries":1},{"symbol":"LPG","lookback":20,"period":"holdout","stress":false,"ret":-17.335491545148784,"dd":41.265114558798565,"n":9,"win":22.22222222222222,"mean":-1.4094003523902454,"boundaries":0},{"symbol":"LPG","lookback":20,"period":"holdout","stress":true,"ret":-22.38275592267282,"dd":43.28529948815123,"n":9,"win":22.22222222222222,"mean":-2.0971326448919365,"boundaries":0},{"symbol":"LPG","lookback":20,"period":"recent","stress":false,"ret":68.66243227425319,"dd":24.342258427148668,"n":6,"win":66.66666666666666,"mean":9.988893478866286,"boundaries":1},{"symbol":"LPG","lookback":20,"period":"recent","stress":true,"ret":61.725230745080786,"dd":24.870019412915934,"n":6,"win":66.66666666666666,"mean":9.221650819502065,"boundaries":1}]. No trade solelyfromthisstudy; newnet3Rtargetsmeanthiscomplementsratherthansupersedesolderfixedgrossbracketstudies.

Source https://runway.adityaperswal.workers.dev/data/stock-bars SIP/all toSep18. Pythonvectorbtlongonly from_orders; close>prior20closehigh&SMA50,nextopen,priorATR20*2stop;targetsolvesnet3Rbaseline5bpfee15bpslipeachside. Samebarstopfirst,adverseopeninggap,60sessionstimeout,windowendliquidationcountshown;twoslots/barpermitsentryandexit. Stress50bpslipholdsbaselinebracketsfixed. Train2016–23,heldapart2024–25,recent2026; alreadyresearchedholdoutnotpristineOOS,selectedsurvivoruniverse. Adjustedpricesapproximatecashinclusivereturns,notrawlivebracketexecution;no15min/+2Rtrailingmanagerpolicy. Reconciledcompoundedtradesagainstvectorbt<=1e-7. SixL10/20/40costcellsperwindowrecordedobservations; nooptimalcellchosen. BWLPshortlistinghistoryhasnotrain. Baseline+stress packets:[{"symbol":"STNG","lookback":20,"period":"train","stress":false,"ret":-14.133942257699697,"dd":76.46550773825108,"n":40,"win":32.5,"mean":0.8589706035147984,"boundaries":1},{"symbol":"STNG","lookback":20,"period":"train","stress":true,"ret":-35.10404009827884,"dd":79.17868842659757,"n":40,"win":32.5,"mean":0.15541497731471315,"boundaries":1},{"symbol":"STNG","lookback":20,"period":"holdout","stress":false,"ret":3.9166496244140827,"dd":27.261380615000153,"n":9,"win":33.33333333333333,"mean":1.0324955271615563,"boundaries":0},{"symbol":"STNG","lookback":20,"period":"holdout","stress":true,"ret":-2.428211231649845,"dd":28.77298176784935,"n":9,"win":33.33333333333333,"mean":0.32772945398201253,"boundaries":0},{"symbol":"STNG","lookback":20,"period":"recent","stress":false,"ret":-7.208707953907004,"dd":26.786876808747863,"n":7,"win":28.57142857142857,"mean":-0.4794319926970626,"boundaries":1},{"symbol":"STNG","lookback":20,"period":"recent","stress":true,"ret":-11.645932862448557,"dd":29.056364629352384,"n":7,"win":28.57142857142857,"mean":-1.1736514074421731,"boundaries":1}]. No trade solelyfromthisstudy; newnet3Rtargetsmeanthiscomplementsratherthansupersedesolderfixedgrossbracketstudies.

Source https://runway.adityaperswal.workers.dev/data/stock-bars SIP/all toSep18. Pythonvectorbtlongonly from_orders; close>prior20closehigh&SMA50,nextopen,priorATR20*2stop;targetsolvesnet3Rbaseline5bpfee15bpslipeachside. Samebarstopfirst,adverseopeninggap,60sessionstimeout,windowendliquidationcountshown;twoslots/barpermitsentryandexit. Stress50bpslipholdsbaselinebracketsfixed. Train2016–23,heldapart2024–25,recent2026; alreadyresearchedholdoutnotpristineOOS,selectedsurvivoruniverse. Adjustedpricesapproximatecashinclusivereturns,notrawlivebracketexecution;no15min/+2Rtrailingmanagerpolicy. Reconciledcompoundedtradesagainstvectorbt<=1e-7. SixL10/20/40costcellsperwindowrecordedobservations; nooptimalcellchosen. BWLPshortlistinghistoryhasnotrain. Baseline+stress packets:[{"symbol":"FRO","lookback":20,"period":"train","stress":false,"ret":-71.04953424090631,"dd":81.78333450149938,"n":48,"win":22.916666666666664,"mean":-1.677963275004787,"boundaries":0},{"symbol":"FRO","lookback":20,"period":"train","stress":true,"ret":-79.31140876580707,"dd":86.13547111829227,"n":48,"win":22.916666666666664,"mean":-2.3638221698893584,"boundaries":0},{"symbol":"FRO","lookback":20,"period":"holdout","stress":false,"ret":-4.889544523268773,"dd":43.32812023963554,"n":11,"win":36.36363636363637,"mean":0.06372373588194262,"boundaries":0},{"symbol":"FRO","lookback":20,"period":"holdout","stress":true,"ret":-11.93827284716731,"dd":45.277348139715244,"n":11,"win":36.36363636363637,"mean":-0.634284536378359,"boundaries":0},{"symbol":"FRO","lookback":20,"period":"recent","stress":false,"ret":96.51833483413102,"dd":28.040252386079377,"n":7,"win":71.42857142857143,"mean":10.89758767356922,"boundaries":1},{"symbol":"FRO","lookback":20,"period":"recent","stress":true,"ret":87.12094386043951,"dd":29.040681099551556,"n":7,"win":71.42857142857143,"mean":10.124006292826486,"boundaries":1}]. No trade solelyfromthisstudy; newnet3Rtargetsmeanthiscomplementsratherthansupersedesolderfixedgrossbracketstudies.

1.CompletedSep18SIP/all https://runway.adityaperswal.workers.dev/data/stock-bars ;FRO51.42>50.66prior20/39.7668SMA50/31.5307SMA200;ATR1.6095. Other3failprior20entry.2.FROhypothetical48.2012ATRstop;net3R61.90137using5bpfee15bpslipeachside;onecost51.52288/risk3.41824,50%gap25.86428(account2.586%);twogap51.7286breaches30normalbudget. Technicalstopisresearchproxy,notindependentproofbusinessinvalidatesspecificallythere.3.Sep11issuer3.41totalexSep18 https://www.frontlineplc.cy/fro-declares-special-one-time-dividend-after-sale-of-two-vllcs/ ;newbuyerdoesnotearnthatdistribution. Retainadjustedhistoricalsignalversusrawliveexitdistinction. Noownholdings:R/daysheldn/a,noexitstomove. Affordabilitypermits1shareifentry/payoffapproved;actual0.

Premises:1.EnactedSep18 law https://www.whitehouse.gov/briefings-statements/2026/09/congressional-bill-h-r-5334-signed-into-law/ ;enrolledtext/implementationconditionsrecorded29663 https://www.govinfo.gov/content/pkg/BILLS-119hr5334enr/pdf/BILLS-119hr5334enr.pdf .2.OFACSep18GL131Jispermissionforparticularactivities,notnewfleetblocking https://ofac.treasury.gov/recent-actions/20260918 .3.ACPmaintenance29672tentative https://pancanal.com/wp-content/uploads/2026/09/adv-34-2026-monthly-canal-operations-summary-august-2026.pdf . Firstorder:actualvesselrestrictionscouldreduceavailablecapacity,whilelostexportbarrelsreducedemand. Second:replacementAtlanticcargoesandlongerroutescouldbenefitlegalspotcarriers;importers/refinersbearfreightandworkingcapital,ownersbearidle/insurancecost. Third:highrentsencouragefleetinvestmentanddemanddestruction;normalizationreversesrent. Theseareconditionalbusinessinferences,notpolicyforecasts. UraniumexemptionspreventblanketCCJscarcityclaim;crypto-evasionlanguagecreatesnoverifiedtokenvaluecapture. MondayNeocapacity10withinnormal9–11doesnotprovegrowingqueues;TuesdayPanamax23needsclassmix/waitdata. Strongcounterargument:physicaldisruptionmaybepersistentandcurrent2026tankerprofitsalreadyshowrent;bindinggapisincrementaldurationandvaluation,notwhetherfreightprofitsarepossible.

Premises:1.Sep17 issuer deck model usesAug28 assumptions https://www.frontlineplc.cy/wp-content/uploads/2026/09/Company-Presentation-Pareto-Conference-2026.pdf ;6.80/10.35/13.91FCFps implies13.22/20.13/27.05% at51.42,notadvertised24%baseyield.2.H1 cashaftercapitaladditions308.116m excludesassetdisposals; forecastFCFmetricisnotidenticaltohistoricalOCF-capex https://ml-eu.globenewswire.com/Resource/Download/458804c1-1f6b-4b8b-9ac6-900fc165b0fc .3.Assumed5xbaseFCF=51.75;6x=62.10justexceeds61.901net3Rhurdle. No independentmultiple/rate-durationevidence establishes6xasfairvalue.4. Risks read from20F https://www.sec.gov/Archives/edgar/data/913290/000162828026021774/fro-20251231.htm :supplier/yarddependenciesandinsuredlossdeductiblesmatter. Conclusion:realrentcaptured,butincrementalweekendshockandtargetneedabusinessbridge;watch. Countercase:longerAtlanticvoyagescanpersistandupsidecashscenarioislarge;wouldreopenonupdatedpaidfixturesanddefensiblevaluation,notrequirecertainty. NofreshNAVclaimed.

Start:account1000,Supply120cash,noownpositions/queues,query_tradesemptyafterreset. Goal:capturebottleneckprofitsatasymmetricpayoff. Bindingconstraint:currentdefensiblepayoffplusrepeatableentry,notcashavailability. Premises:1.PSIX45.87passesbreakoutbutbelow55.48675SMA200;ETN/POWL/NVTfailentries https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all.2.Fixednet3RPSIX2025+110.022%N4;2026-11.169%includingoneboundarymark, naturalexits-16.139%N4,25%win,avgwin37.851%/avgloss15.261%. No pre2024train; reusedholdoutnotindependent.3.Issuer https://www.sec.gov/Archives/edgar/data/1137091/000162828026054328/psix-20260630.htm supportsoperatingrecoverybutassumed20xQ2annualizedvalue58.464only2.365Rvs61.710hurdle.4.Simulatedtargetodds22.105–34.405%varybywindow/block;notcalibratedliveodds. HalfKellyfromnaturalexitreturns2025.5vs20260;choosingfavorablepoolunjustified. Conclusion:0orders,120cash;notpredictionofdeclineorclaimnoopportunityanywhere. Validity:jointpremisessupportrejectingthissetup; soundnesslimitedbyshortselectedhistory,privatecontractsanddailyexecutionproxy. Countercase:EPS3.5at20x70supports4.62R;furthermargin/throughputgainscouldmakeprobeappropriate. Reopenonnewpaidshipments,credibleEPSbridgeabove3.0855atassumed20x,orindependentlysupportedalternativepayoff/entry;don'tturnthesethresholdsintoissuerpromises. One-share46.122cost/5.083plannedrisk/23.31350%gapstressfitscaps;twosharehistoricalgap32.20breaches30normalriskbudget. Actualsize0, noarbitrarygap-probabilityveto. Existing270PSIXfilings/214signals/274powerwatchand378refund/393NVTfundingmonitorssuffice;next08UTCreadchangedoriginals,preservefrozenfigures. No newFUND;allworkfitsSupply.

Command node /app/agent/dist/backtest.js --symbols PSIX,ETN,POWL,NVT --strategy breakout --lookback 20 --days 1460 --stop-pct 10 --target-pct 30. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all. Engine inspection /app/agent/dist/backtest/engine.js shows same-close signals/entries, fixedpercentagebrackets, noexplicitfees/holdout/adversegapexecution. Therefore diagnostic, not trade authorization; costednextopenvectorbt3479–3482 isdifferentstrategy. PSIXreturn24.8316%,DD18.7723%,win41.9355%,N31;ETN15.4513%,DD16.1143%,win41.7582%,N91;POWL-4.9892%,DD33.0766%,win38.4615%,N78;NVT-14.7825%,DD27.9100%,win31.9588%,N97. Fullrecords/parametersobservedandlocalcli.json; no choosingCLIresultovercostedtest.

Premises:1.NumPyseed91906,20000pathsfor3pools(all432,last252,2026179)*2blocklengths5/10,60sessions,jointOHLC/priorclose https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all .2.Frozen45.87/41.2664/61.710237;5bpfee50bpslip,adversegapstop,cappedtarget,samebarstopfirst,timeoutexit. Allhistory5/10mean+2.8986/+3.2096%,target33.415/34.405%;last252mean-1.7638/-2.9310%,target23.855/22.105%;2026mean+1.1850/-.1533%,target29.510/27.145%. Medianall-11.0204%.3.Expectedlogstockweightgrid0..1thenhalf:all.337/.361,last2520/0,2026.139/0;cap120bindspositivemodelsbutnotapproval. NaturalexittradehalfKelly2025.5/20260also unstable.4.TwoshareP99loss31.64–36.24dollars;quantilesnotbounds/unconditionalresamplingnotliveentryodds. No15minexits,+2Rratchet,newshocks/newsconditioning. Conclusion:don'tselectfavorablewindowtosize;one-shareprobeaffordableifvalue/edgeimprove,actual0.

Premises:1.Baselinecohorts/separatelyrerunSMA200strategy https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all .2.PSIX2025fourentrieslack200-barwarmup:comparisonisunknown,notbelowtrend. ZeroSMA200filteredtradescannotproveprotection.3.2026baselineaboveSMA200N2mean11.3202%,belowN3mean-8.21568%includingoneboundaryliquidation. FilterrerunN3notdirectlycomparable:changesentrysequence. All5baseline2026entriesSPYaboveSMA200,yettotal-11.169%;markettrenddoesnotrescueissuer-risk.4.L10/20/40returns2026-6.52/-11.17/+6.42%;selecting40nowisoverfit. RSI71andSMA50breakoutderivefromsamepricehistory,notindependentsources. Conclusion:retainfrozenrule,prospectiveoutcomesneededforconditionalodds.

Premises:1.SIP/allSep18 https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all .2.PSIX45.87>prior20close45.55/SMA5036.0078,but<SMA20055.48675;ATR202.3018,simpleRSI1471.0399,20dayannvol79.4489%.3.ETN424.77<425.37trigger;POWL183.96<197.67/SMA50203.2032;NVT157.88<162.57trigger. Allthreefailentry. PSIXIEX45.82/SIP45.87feeddifference,notgain.4.HypotheticalE45.87/S41.2664/T61.710237 netrisk5.082862/gain15.248585persharewith5bpfee50bpslip/side. Onecost46.1224/two92.2448fit120fund,buttwohistorical34.187%gaprepeatloss32.2002>30normalbudget,one16.1001. Assumed50%gapone23.3135/two46.627;notprobabilities/bounds. Noorders,holdings,Rordaysheld.

Premises:1.Common432dailyadjustedreturnsDec27,2024–Sep18,2026 https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all .2.PSIXtotal55.5443%;deletebest5=>-40.3676%,deleteworst5=>628.9994%;best5are217.025%ofnetloggain. Symmetricdeletionisdescriptivestress,notfeasibletradingorreasonstododgefat-tailinvesting. Mean+.32275%/day,median-.34725%,P1/P99=-17.5354/+19.0215%.3.OLSintercept+SPY+equalweightETN/POWL/NVT:intercept.065405%/day,HAC5t.2555;marketbeta1.37135,powerbeta.83742,R2.27974. Correlatedfactors/selectedsurvivorsprecludecausalalpha.4.PSIXnaturalexit2025N4mean23.6247%,IIDbootstrap95meanCI[-2.3439,49.5933];2026N4mean-1.9832%,CI[-15.9065,24.5858]. Wideintervals;bootstrapcannotrecovermissingcrises. Conclusion:taildependenceexpected,butregimes/fewtradescannotcalibratethisentry.

Premises:1.FY25Item1A https://www.sec.gov/Archives/edgar/data/1137091/000162828026013207/psix-20251231.htm limitsquicksubstitutionofsophisticatedinputs andwarnscompetitive/contractualcostpassthroughlimits. 2.Aug6Q2 https://www.sec.gov/Archives/edgar/data/1137091/000162828026054328/psix-20260630.htm SupplierA<10%purchases but290.108m2026–29SWIECminimums/selectedengine1.75%MORfees. Broadcollaborationexpirydoesnotcancelthesecontracts. Firstorder:powerdemandcanraiseproductionwhileinput/rampcostsabsorbmargin. Secondorder:vendor/customertermsdividecashrent;tariffcashnotautomaticallyshareholdercash. Thirdorder:expandedcapacity/substitutioncanimprovedeliverywhileerodingscarcityprices. Chainsconditionalinference,notforecasts. Bear:shipmenttiming/suppliertermsmakebacklogunreliablecash. Bull:Q2sequentialmargin/positiveexrefundFCFshowprogress. Bindingunknowns:customerpricing,cancelrights,throughput,mix. Purchaseshareisnotswitchingcost. Noassertednewoutage.

Premises:1.Q2/H1June30 issuer https://www.sec.gov/Archives/edgar/data/1137091/000162828026054328/psix-20260630.htm Q2GM27.119% versusQ122.882%, sales-21%y/y, Wisconsin rampcosts persist. H1OCF75.724m-capex2.736m=72.988m; remove22.687m customerrefundliability=>50.301m; remove12.594mAR/3.992minventoryrelease=>33.715m cash-quality sensitivity, not normalizedforecast. Acquisition11.911m separate. 2.Q2annualizedEPS2.9232 versusH1annualized2.09485, neitherforecast. At45.87,mcap1057.3035m,Q2annualizedPE15.692. 3.2ATRstop41.2664/costed3R61.7102;assumed20xQ2value58.4639 yields2.36484R. TargetrequiresEPS3.08551at20x. BullEPS3.5*20=70 yields4.62R butunconfirmed. Conclusion: recoverycanjustifyupside;20xisassumednotfairvalue, don'traiseittoforce3R. Counterexample:furtherthroughput/mixcouldexceed3.09EPS. Onecorporatesourcegroup; annual/quarterlynotindependentcorroboration.

Premises:1.SIP/all completed daily source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all . Train2016–2023, holdout2024–2025, 2026 separate reused diagnostic. PSIX begins2024Dec26; no training. NVT begins2018May. 2.Frozen L20 prior closing high AND SMA50, next-open entry; prior arithmeticATR20, stopE-2ATR, target solves net3R at5bpfee+50bpslip per side; 60sessions max. Adverse opening stops, capped target, same-bar stop-first. Vectorbt longonly10000 fractional reinvestment; expanded open/close marks allow same-day exit; compounded trade returns reconcile <1e-8. Boundary liquidations counted separately. 3.Coststress100bpslip holds target fixed; L10/40 andSMA200 are sensitivities, not selected rules. PSIX SMA200 unavailable at2025entries; zero filteredtrades is warmup, not proof of protection. Summary [{"period":"train","return":26.578023309322376,"DD":37.10277433475685,"N":29,"win":31.03448275862069,"boundary":1,"without_best":3.3886674743498935},{"period":"holdout","return":87.17086632432587,"DD":22.11321555380127,"N":11,"win":54.54545454545454,"boundary":0,"without_best":33.423351254305175},{"period":"recent","return":-0.6491850271975519,"DD":26.604408244933385,"N":5,"win":20,"boundary":1,"without_best":-23.604175791528448}]. Conclusion: retrospective robustness only, previously viewed periods/current survivors; no primary bottleneck-event labels, actual15min exits or+2Rmanagement modeled. Best-trade deletion diagnoses concentration, not investable strategy. No allocation solely from this test.

Premises:1.SIP/all completed daily source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all . Train2016–2023, holdout2024–2025, 2026 separate reused diagnostic. PSIX begins2024Dec26; no training. NVT begins2018May. 2.Frozen L20 prior closing high AND SMA50, next-open entry; prior arithmeticATR20, stopE-2ATR, target solves net3R at5bpfee+50bpslip per side; 60sessions max. Adverse opening stops, capped target, same-bar stop-first. Vectorbt longonly10000 fractional reinvestment; expanded open/close marks allow same-day exit; compounded trade returns reconcile <1e-8. Boundary liquidations counted separately. 3.Coststress100bpslip holds target fixed; L10/40 andSMA200 are sensitivities, not selected rules. PSIX SMA200 unavailable at2025entries; zero filteredtrades is warmup, not proof of protection. Summary [{"period":"train","return":-68.54884354722476,"DD":73.1680251931859,"N":48,"win":20.833333333333336,"boundary":1,"without_best":-75.54253212651038},{"period":"holdout","return":33.287957806031734,"DD":51.38488012452538,"N":16,"win":37.5,"boundary":0,"without_best":-1.793485624584501},{"period":"recent","return":81.44926171139836,"DD":23.941248783294423,"N":5,"win":60,"boundary":0,"without_best":33.24684369271884}]. Conclusion: retrospective robustness only, previously viewed periods/current survivors; no primary bottleneck-event labels, actual15min exits or+2Rmanagement modeled. Best-trade deletion diagnoses concentration, not investable strategy. No allocation solely from this test.

Premises:1.SIP/all completed daily source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all . Train2016–2023, holdout2024–2025, 2026 separate reused diagnostic. PSIX begins2024Dec26; no training. NVT begins2018May. 2.Frozen L20 prior closing high AND SMA50, next-open entry; prior arithmeticATR20, stopE-2ATR, target solves net3R at5bpfee+50bpslip per side; 60sessions max. Adverse opening stops, capped target, same-bar stop-first. Vectorbt longonly10000 fractional reinvestment; expanded open/close marks allow same-day exit; compounded trade returns reconcile <1e-8. Boundary liquidations counted separately. 3.Coststress100bpslip holds target fixed; L10/40 andSMA200 are sensitivities, not selected rules. PSIX SMA200 unavailable at2025entries; zero filteredtrades is warmup, not proof of protection. Summary [{"period":"train","return":45.87936291118282,"DD":21.63138452995964,"N":40,"win":45,"boundary":1,"without_best":23.870220764929705},{"period":"holdout","return":2.7270691743964237,"DD":27.000140816993767,"N":13,"win":30.76923076923077,"boundary":0,"without_best":-20.56053422537576},{"period":"recent","return":3.4519729420564804,"DD":14.484544522620036,"N":4,"win":50,"boundary":0,"without_best":-10.836246395344984}]. Conclusion: retrospective robustness only, previously viewed periods/current survivors; no primary bottleneck-event labels, actual15min exits or+2Rmanagement modeled. Best-trade deletion diagnoses concentration, not investable strategy. No allocation solely from this test.

Premises:1.SIP/all completed daily source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX,ETN,POWL,NVT,SPY&timeframe=1Day&feed=sip&adjustment=all . Train2016–2023, holdout2024–2025, 2026 separate reused diagnostic. PSIX begins2024Dec26; no training. NVT begins2018May. 2.Frozen L20 prior closing high AND SMA50, next-open entry; prior arithmeticATR20, stopE-2ATR, target solves net3R at5bpfee+50bpslip per side; 60sessions max. Adverse opening stops, capped target, same-bar stop-first. Vectorbt longonly10000 fractional reinvestment; expanded open/close marks allow same-day exit; compounded trade returns reconcile <1e-8. Boundary liquidations counted separately. 3.Coststress100bpslip holds target fixed; L10/40 andSMA200 are sensitivities, not selected rules. PSIX SMA200 unavailable at2025entries; zero filteredtrades is warmup, not proof of protection. Summary [{"period":"holdout","return":110.02245663389418,"DD":28.94803335298578,"N":4,"win":75,"boundary":0,"without_best":33.026122096773115},{"period":"recent","return":-11.168965671430408,"DD":44.555485542890914,"N":5,"win":40,"boundary":1,"without_best":-35.560295110382725}]. Conclusion: retrospective robustness only, previously viewed periods/current survivors; no primary bottleneck-event labels, actual15min exits or+2Rmanagement modeled. Best-trade deletion diagnoses concentration, not investable strategy. No allocation solely from this test.

Start: account1000, Supply120 cash, no own holdings or queued orders; query_trades empty after reset. Goal: buy durable bottleneck beneficiaries at asymmetric payoff. Binding investment constraint: defensible price payoff plus an appropriate entry, not available capital. Premises: (1) Q3 operating benefit and positive quarterly cash are real, but1.580m liability financing exceeds1.506m OCF; annualized margins/cash remain uncertain. https://www.sec.gov/Archives/edgar/data/1000230/000143774926029950/occ20260731_10q.htm https://www.sec.gov/Archives/edgar/data/1000230/000143774926019811/occ20260430_10q.htm (2) Standard terms offer repricing protection, actual realization/allocation unknown. https://www.occfiber.com/terms/ (3) Growth scenario110m revenue/38%GM/30mSG&A/25%tax/22x yields19.815, only2.09R after55bp costs; net3R hurdle21.920. Stronger bull assumptions can yield32.272 but are not guidance. (4)14.82 remains below14.92 inherited breakout; prior costed train-83.82%, holdout+19.16% with51.36%DD does not establish a transferable scarcity edge. https://runway.adityaperswal.workers.dev/data/stock-bars . Conclusion: actual size0, no orders/exits/cancellations, keep120cash for a qualified Supply payoff. Validity: this supports rejection of tonight's setup, not certainty stock falls. Countercase: improved allocation, working-capital release and new mix could make the bull scenario happen quickly; a small probe would then be appropriate without waiting for certainty. Subjective business prior for positive postinvestment cash in next two reported quarters moves40% to55% after terms andQ3 review; not calibrated share-return odds. Two quartersGM>=35% andOCF/NI>=70% without growing liability reliance would move judgment toward70%; GM<32% with renewednegativeFCF orcreditavailability<3m toward25%. These are evidence sensitivities, not mandatory waiting gates. Current modeled binary half-Kelly risk stakes range.515%–6.049% by source regime, not deployment authorization. At14.82 a one-share probe costs14.82, planned netrisk2.299, net3Rgain6.898; explicit50%gap loss7.532. Three-share50%gap stress22.597 includingcosts; four30.129 exceeds normal30 budget. No hard assertion gaps cannot exceed50%. Existing job274 monitors operating filings until next06run; monitor403 covers put/refinancing preparation. Source saturation: final specific allocation/price queries added no new primary cash economics; private contract and SKU price/volume gaps remain. No new FUND: this belongs in Supply.

NumPy seed91904,20,000 paths/model,60sessions,five-day blocks of jointOHLC/priorclose ratios. Two source pools:2016–23 training2011returns and latest252sessions. E14.82/S12.672/T21.919686,55bp each side; gap stops at worse open, target capped, same-bar stop first, timeouts marked at60. Source https://runway.adityaperswal.workers.dev/data/stock-bars . Training/recent modeled target rates14.305/34.13%,stop56.01/65.075%,mean netreturn+.2761/+5.5309%,median both-15.4293%. Mean wins26.8199/45.7187%,mean losses14.7034/15.8906%,profit rates36.075/34.77%. Binary half-Kelly risk-stake approximations.5148%/6.0489% of bankroll; not direct position notional and not validated live edge. Sampling recent AI regime nearly triples target frequency but also raises stop frequency; don't select the pool to justify an order. Daily path model misses monitoring delays, future jumps and conditional fundamental information; simulated maximum loss is not a bound. Four shares59.28 imply planned netstoprisk9.197 and assumed50%gap loss29.64 plus costs, below3%account only approximately; use3 shares if reserving explicit friction under30. Quarter-probe of a feasible4-share ceiling would be1share14.82, stoprisk2.30 and50%gap7.41. All actual sizes0 because current entry/defended payoff remain unqualified, not because an arbitrary tail threshold vetoes risk. If qualified later, recompute half-Kelly with appropriate evidence and current cap; don't transfer these unconditional odds.