Stats

Python uses 2540 complete adjusted OHLC observations Sep19 2016–Sep16 2026; rawfeed has2542timestamps,Sept17–18closesmissing. 2026 common-date log returns withDAX/LFUS N175: corrELG/DAX .352652,ELG/LFUS .418820,LFUS/DAX .290343. OLS ELG=a+b*DAX: b1.082706,a.001876274/day,R2 .124363; descriptive sample, no predictive alpha inference or causal attribution. US and German close times differ; DAX performanceindex and adjustedshares give imperfect comparability. Sep8–9 SKannouncementwindow ELG-.418988%,DAX-1.657904%,LFUS-.229342%; one observation, primarysupplierrelease07ETduringEuropeantrading, no independent event edge. DortmundSep16–17/18returns unavailable, notzero. RV20annualized42.28358%,historicalworstadjustedopeninggap-12.621354% notlossbound. Primary input https://query1.finance.yahoo.com/v8/finance/chart/ELG.DE ; sameendpointGDAXI/LFUS. Dividend events10/no splitsreturned; no independent fullcorporateaction audit. Budgetedresearch doesnot generalizeoneissuer toentire200mmindustry.

Premises: L20 held-apart2023–25 net trades,20bp per side; no own live closedtrades (N0), so no measured live slippage or winrate. IID20000 bootstrap means, selection/autocorrelation uncorrected. ATI N19 win42.105%,avgwin20.151%,avgloss6.471%,95%meanCI[-1.249,11.189]%;SEI N20 win35%,avgwin26.102%,avgloss10.057%,CI[-5.059,11.123]%. HalfKelly risk=.5*(p-(1-p)/(avgwin/avgloss)):LFUS.029794,ATI.117571,SEI.049780,VSH/UMC/PXS0. Divide riskbudget bylossfraction fornotional, thenfund/venue/gapcaps; values are not recommendedequityallocations. 2026 SEIvol80.119%,betaSPY2.694;LFUS–VSHcorr.70093 shows shared factor; PXSbeta.122 does not remove singleissuerloss. WorstobservedSEIgap−16.949%Jan27 2025, notmaximumpossible. Allstatistics observed. Positive CIs crosszero and businessregimechanges limit soundness. Source https://runway.adityaperswal.workers.dev/data/stock-bars . Noorder solelyfromKelly.

Premises: all13221 non-test listings from Nasdaq symbol directories attempted against completed SIP/all Aug1–Sep18 bars.12571 returned,12416 measurable;650 missing includes384 unsupported symbol formats and266 valid-format misses after bounded retry;155 histories too short. Thus broad attempt, not complete all-listing measurement. Filter nonETF,close>=2,20day mean dollarvolume>=1m yields3869,181prior20closebreakouts. 20day distribution P10-15.3606%,median-3.4636%,P90+8.1108%. Preferred/CEF classification can remain; current-listing survivorship and momentum-selection biases. UMC,PXS,SEI affordable breakout candidates; ATI/LFUS/VSH primary capacity candidates currently below frozen breakout. FEAM/PSIX prior research retained; speculative top percentage movers lack demonstrated supply rent. Separate19foreignstock+Brent scope covers Japan/Korea/Taiwan/Europe/India/Brazil, not all foreign listings: DBHiTek+40.5797%20d/−17.2572%60d strongest; prior data with unfinishedSep21 discarded. Eightcrypto completedSep20:AVAX+56.3908%20d,annualvol86.9035%,BTC+3.3058%20d. No demonstrated token claim on paid industrial scarcity, so no mandate-basedcryptoentry. Sources https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt ; https://www.nasdaqtrader.com/dynamic/SymDir/otherlisted.txt ; https://runway.adityaperswal.workers.dev/data/stock-bars ; https://query1.finance.yahoo.com/v8/finance/chart/000990.KS ; https://runway.adityaperswal.workers.dev/data/crypto-bars . Every measured screen row journaled in chunks; missing coverage named, not repaired beyond discovery allowance.

Source SIP/all daily https://runway.adityaperswal.workers.dev/data/stock-bars . 2026N179: LPG betaSPY.4719/corr.1391/annvol44.97%;BWLPbeta.3225/corr.1040/vol41.12%;CSTMbeta2.0005/corr.5441/vol48.73%. LPG-BWLPcorr.7668: low marketbeta does not make two shipowners independent. BaseLPGholdoutN17 win6/17,avgwin20.4379%,avgloss7.9172%,mean2.0905%,20kIIDbootstrap95%mean[-4.2318,8.8474]%; recentN8CI[-.1053,17.8982]. BWLPheldapartN8mean-2.2291%,CI[-6.9539,4.1100]. BinaryaveragepayoffhalfKellyrisk=.5*(p-(1-p)/(avgwin/avgloss)):LPG.0511419,BWLP0. Riskfraction mustbedividedbylossfractiontoconverttonotional,then capped—not51USDnotional. Sample andselectionuncertainty substantial; no liveclosedtrades tocalibrateslippage/probabilities. LPGworstobservedopen gap-22.275%Mar16 2020;BWLP-13.299%Apr4 2025;CSTM-19.625%Oct23 2024;corporateactionauditnotcomplete. These are descriptivebounds ofsample,notfuturelosslimits. Prior neutral; currentpositiveLPGcash/testevidenceupgradeswatch, notcounterfeitprobability; freshpaidQ4fixturesandsignal wouldjustifynewsize, route normalization andcashabsorptionweakenit. Noorder.

1. Aligned SIPadjusteddailyreturns https://runway.adityaperswal.workers.dev/data/stock-bars : 2017–22/2023–25/2026N1510/752/179. CSTMOLSbetas1.849/1.488/2.001;2026vol48.73%,R2.296. CENX2026beta1.346,vol64.02%. AA/CENX2026corr.7471 andCSTM/KALU.7195: multiplealuminumpositionsarenotindependentbets. OLSdescriptive, nopvalueorcausalalphaassertion. 2. CSTMcompletedpullbackholdoutN12,6wins:avgwin17.0718%,avgloss6.13565%,mean5.4681%;20kIIDmeanbootstrap95%[-1.3390,12.5495]%. UniformpriorBeta(7,7)posteriorwin95%[25.1345,74.8655]%, notfrequentistconfidence. CENXN13mean7.7655%CI[-5.0361,20.9290]%. 3. Binaryaverage-payoffhalfKelly CSTMholdoutRISKfraction.160149; dividingbyaveragefractionalloss.0613565 giveswealthallocation2.61014 beforecaps. Do notmislabel16%riskKellyas16%stocknotional. TrainCSTM11.2375%risk/119.645%wealth; CENXholdout11.7722%risk/85.003%wealth. Estimatesignorewithintaildistributionandparameteruncertainty. 4. CSTM2026completedN3mean+2.0802%doesnoterasewindow-endmarkedfullreturn-2.0641%N4; only3naturalexitsarecompletedwinodds. LiveSupplyledgerN0: noactualfill/slippageedge. Conclusion: positiveconditionalpointestimatesmeritwatch; currententryfalseallocates0. Anyfuturetradecappedbyfund120,accountposition250 andgaprisk,notunboundedKelly. Selection/multiplelookbacksandserialdependencemeanbootstrapCIunderstatesmodeluncertainty.

Premises: 1. Adjusted SIPdaily returns https://runway.adityaperswal.workers.dev/data/stock-bars ; 2017–22/2023–25/2026 windows N1510/752/179. 2. OLSagainstSPY CEbetas1.2436/1.4622/.2641 andEMN1.1838/1.1138/.9004; 2026CEcorr.0657, annualvol53.24%. Descriptive, no robust-significance or causalshortage claim. 3. FixedL20 heldapartCE N12 wins25%,avgwin9.4077%,avgloss5.3917%;EMN N15 wins13.333%,avgwin15.1456%,avgloss4.5233%. HalfKelly=max(0,.5*(p-(1-p)/(win/loss)))=0both. 4. 20kIID meanbootstraps95% CIsCE[-5.4125,3.1093]%,EMN[-4.8230,2.0034]%; these are uncertaintyintervals, notdownsidebounds. LiveSupply ledgerN0 means no measured slippage orhistorical fundedge. Conclusion: no calibratedpositiveposition size from thisrule; changedCEbeta iscounterevidence to assumingordinarybroadbeta, notproof ofscarcityalpha. Selectednames/regimes, repeatedhistoricalwindows, smallsamples andserialdependence limittransportability. Preservepositivecashvaluationcountercase instead ofclaimingbusinesscannotwork.

Premises:1.Fixed L20 long-only chronological holdout tests3589–92 use https://runway.adityaperswal.workers.dev/data/stock-bars .2.Winrate/averagewin/averageloss%:CE17.647/21.6997/6.0195 N17;IMOS29.1667/16.6557/6.7616 N24;BTDR33.3333/42.9153/16.8298 N24;AMTX33.3333/41.2324/15.0053 N21. HalfKelly risk fractions max(0,.5*(p-(1-p)/b)) are0/.002054/.035946/.045360; these are stake-at-risk fractions, NOT capital allocation. 3.20000 IID bootstrap mean CIs allcrosszero: CE[-5.833,4.665]%,IMOS[-4.275,4.664],BTDR[-8.060,15.665],AMTX[-7.657,16.539]. Average positive trade returns can coexist with negative compounded returns; no arithmeticmean-as-PnL. Conclusion: uncertain regime-dependent distributions do not establish actual execution edge. No live postreset Supplytrades to estimate slippage; costs are assumptions. Candidate selection uses current winners and historical windows are not untouched prospective evidence. Actual size0 because cashvalue/currentrule gates fail, not because uncertainty itself forbids a convex probe.

Premises:1.Nasdaq directory https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt and https://www.nasdaqtrader.com/dynamic/SymDir/otherlisted.txt yielded13220unique non-testentries.2.Snapshot https://runway.adityaperswal.workers.dev/data/stock-snapshots?feed=iex queried12836after384preferred-$formats excluded;12689returned12420usable pairs,8133Sep18dailybars.3.FreshnonETF IEXdollarvolume>100000 subset3430;medianreturn-.55243%,quartiles[-1.64402,.40347]%. Mean+.59191%/SD50.059% contaminatedbyOIG2917% corporate-action/collisionoutlier, NOTeconomicmarketbreadth. Initialwholebatchfailurefrompreferredsyntax repaired; nofinalrequesterrors. Broad discovery flaggedFEAM,BTDR,SGML,IMOS,FPS/AMTX pluscrypto-beta cohort; allcandidatesrequireSIPactionsandprimaryverification. Names withoutnewbottleneckconnection rejectedatscreenonly,notfundamentalverdict. Dataset iscurrentlisting survivoruniverse inclETFs/warrants/preferreds,notpoint-in-time allUSsecuritymaster/OTC. A fullnamekeywordfiltermissesopaquecompanynames; manuallyaddedCE/EMN/LYB/OLN/MEOH chemicalchain. Sources close tomarket butIEXpartialvolume notconsolidated. Fullrawfiles /tmp/universe.csv,/tmp/snapshots.json,/tmp/screen.csv. No order on1dayreturns.

Premises:1.Costed L20 backtest3555 trades train49/holdout16/recent5 from https://runway.adityaperswal.workers.dev/data/stock-bars .2.NumPy20000IID resamples seed92000:95%meanCI train[-3.723,2.966]%,holdout[-5.924,13.196]%,recent[-4.230,32.607]%. Holdout avgwin27.949%/avgloss11.193%,win37.5%; halfKelly riskfraction .5*(p-(1-p)/b)=.06236,train0/recent.22371. This is fraction at risk under two-point payoff model, NOT6.24% portfolio allocation; gaps andcapitalcaps govern.3.FRO/POWL252dailyreturncorrelation.00122, sample alignment common2693dates; backward correlation does not exclude crisis convergence. No live Supplytrades to calibrate fillslippage. Conclusion:positive regimes meaningful but selected small sample means odds uncertain; no blind recent-only Kelly. Absence currentPOWL entry and unsupportedtarget remain binding,actualsize0. CIcrossingzero alone is not rejection rationale.

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.

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.Align daily adjusted simple returns for2026 throughSep18, N179 from https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=OCC,GLW,SPY&timeframe=1Day&feed=sip&adjustment=all .2.OCC/GLW correlation.481718;OCC/SPY.199248. OLS with intercept andSPY beta2.059435,R2.039700,daily intercept+.869058%.3.OCC daily return percentiles1/5/50/95/99=-13.2705/-10.4709/-.28454/14.8018/28.6742%. Conclusion: positive mean/intercept coexists with negative median and fat positive tail. The tiny marketR2 and survivor-selected hot-stock sample cannot establish supplier-rent alpha or precise future odds. GLW andOCC are partly correlated exposures, not independent positions. No significance or stationarity claim; no regression-based position entered.

Premises: (1) SIP adjusted daily returns aligned on 155 common sessions Feb6-Sep18,2026; the FPS IPO constrains the sample. https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all (2) GLW/MU correlation .5976, GLW/FPS .5541, FPS/BTDR .5052. Intercept-plus-SPY OLS beta GLW3.254, FPS3.387, BTDR4.412; R2 .272,.300,.305. (3) These are noisy sample associations, not proof of common economic causation, permanent betas or positive alpha. Selection after observing AI narratives biases inference. Conclusion: glass, switchgear and computing may be one risk cluster despite different supply bottlenecks. A future combined allocation must share a stress budget; do not sum standalone half-Kelly allocations as independent bets. Counterexample: firm-specific funding or production shocks can overwhelm this average correlation. No current position or trade.

Measured costed fixed-bracket trade distributions, numpy seed91902/20000 IID bootstrap draws. GLW training39 mean+.322%/trade,95%CI[-2.004,2.870],win33.33%,averagewin10.22/averageloss4.63%; holdout16 mean4.557%,CI[.090,9.335],win56.25%,win12.00/loss5.01%. Training halfKellyaveragepayoffriskbank1.574%,holdout18.988%: large regime change, use neither as unboundednotional. All6 adjacent GLW holdouts positive71.86–210.05%,not6independentfindings. OCC trainingmean-2.996%,holdoutmean5.339% withCI[-9.824,21.832]; high variance and training-83.82% defeat scarcity-only selection. BTDRbaselineholdout-42.26% while3ATRcells+161–162%; do not select stopwidth after inspecting holdout. FEAM/FCEL all6holdoutcellsnegative. MU6positivecells but currentpricefailsentry and whole-sharecap. FPSshortIPOtest separately recorded; no long-historyproxy silently borrowed. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all ; actual reset query_trades(fund=supply)N0. Thus no liveexecution winrate/slippage estimate. Assumptions: tradesIID, chosen surviving companies, corporateadjustmentfaithful. Missing delisted peers, clusterdependence and changingAIregime can widenerror. Validity: evidence ranks candidates for investigation; not proof of current edge. Soundness limited by selection on today's narratives and historicalperiods potentially examined in sharedcorpus. A nominal CI excludingzero is not an automatictrade; currententry,contracteconomics and target still required. Strongcountercase: newscale/contracts can improve futurecash despite historiclosses; freeze a newprospectivestudy before using it.

1. Corrected longonly baseline3368–71: NVT training36 mean3.036%/trade, bootstrap95%CI[-.185,6.415]%; holdout12 mean3.772%,CI[-4.197,12.995]%; win44.44%/33.33%. Training avgwin13.558%,avgloss5.382%; holdout avgwin24.078%,avgloss6.380%,consistent with fat-tail return concentration. 20000 IID trade draws seed91900; dependence/nonstationarity can widen uncertainty. 2. All6 adjacent NVT holdout cells positive(+38.0 to+117.0%); not six independent discoveries and do not select best. HUBB baseline holdout-21.13%,2/6cells positive; transfer of same industrial narrative not supported. 3. Supply query_trades yields two operator-liquidated winners FRO/STNG, no stop/target exits; 2/2 win rate cannot estimate a strategy stop probability, average losing trade or execution slippage without reference quotes. Source query_trades(fund=supply), datedSep2–4 after reset. 4. Historical point-estimate halfKelly risk fraction NVT.11196training/.07833holdout, HUBB holdout0. These are average-payoff approximations, not permission to put11% account risk into one trade; normal3% equity risk/caps override. Source https://runway.adityaperswal.workers.dev/data/stock-bars . Conclusion NVT merits a prospective signal/cash monitor; statistical confidence alone neither vetoes nor authorizes entry. Current missing trigger and current target economics control.

1. DirectACP Sep10 AugustA34, fullPDF recovered viaNode afterPython403: https://pancanal.com/wp-content/uploads/2026/09/adv-34-2026-monthly-canal-operations-summary-august-2026.pdf . Monthlytotal1029,transits33.19/day,arrivals33.1/day; no vessel-level distribution. 2. Python derived water-minus-transit23.63h=67.7077%of34.90h, but this includes all non-transit time, not proven congestion wait. Roundedarrivals-minusdepartures-.09/day (~-2.79over31days) doesnot establish actualqueuechange: cohortboundary,smallvessels/definitions androunding unverified. 3. 175/178Neo bookedslots=98.3146%,204/227auction=89.8678%; Neo excludesauctions so addingpercentagesdoublecounts/changesdenominator. Physical33.19/36-38sustainablecombinedcapacity implies87.34-92.19% utilizationrange; cannotaddindividual-lockmaxima. 4. Prior: constraints can raisefreight; evidence weakens inference ofacceleratingqueue from headlinebookingfill. Counterexample concentratedNeo arrivalwaves canproduceVLGCqueues despitebalancedmonthlytotals. Needdailyclass-specificarrivals,waitdistribution,auctionprices and paidTCE beforecalibratingstockedge. No regression/correlation fromN1monthlyaggregate; no inventedprecision. No order.

1. Source https://runway.adityaperswal.workers.dev/data/stock-bars SIP/all Jan2016-Sep17. TEN has2691 daily gaps and none <=-30%; worst gap-9.5963% (Apr29 2020), worst pre2024 close return-17.58995%. That does NOT establish zero future disaster risk. Under deliberately simplistic IID Bernoulli, zero/2691 yields one-sided95% daily upper.111262%; converting that bound across60sessions gives6.46126%. A5% probability PER TRADE cannot be compared directly with a daily frequency, and clustered crisis regimes undermine IID. Thus earlier5%/-30% remains a stress case, not calibrated probability, and absence does not refute it. 2. PSIX May12 2026 actual adjusted gap-34.18735%,2025Nov7-29.99754%; short history misses other regimes but confirms meaningful gap risk. LPG worst pre2024 gap-22.27513%. 3. AA feed Nov1 2016 apparent-74.35646% coincides with corporate separation confirmed by https://investors.alcoa.com/resources/investor-faqs/default.aspx ; remove predecessor boundary by starting current AA historyNov1, not discretionary return clipping. Clean worst gap-13.2465%approximately; exact values saved in observation packets. 4. Last252 TEN/FRO correlation.819413,TEN/STNG.708182,TEN/LPG.600194; holdings would share a shipping factor, not independent bets. Conclusion retire uncalibrated tail veto while retaining explicit gap sizing and regime uncertainty. Closed-trade sample2 operator-liquidated wins cannot estimate edge. Hard account risk and fund101.99 cap remain; no forced entry from removing an invalid inference.

Premises: 1. Prior strategy3287 explicitly used an assumed5% probability of a30% loss; conditional mean1.4467783586% gives mixture-.12556055933%, breakevenq4.60072044%. 2. No measured event series supported5%. Own venue history consists of two operator-liquidated winning lots, with no independently managed stop/target or loss sample (runway:query_trades?fund=supply). 3. Adding an entire tail loss to a historical return distribution can double-count risk already present; whether it does must be inspected rather than assumed. Conclusion: supersede use of that scenario as a factual negative-expectancy hard veto. Retain it as an explicitly labeled sensitivity alongside empirical gap and trade distributions and stronger/weaker stress assumptions. Validity of arithmetic does not establish soundness of probabilities. This is a model correction, not authorization to chase TEN: frozen52.81/45.41/55.36 still yields only.31453R. Bar source https://runway.adityaperswal.workers.dev/data/stock-bars . New panel data fetched2016-Sep17 with held-out calendar windows, results pending.

Own trade query contains only two completed market entries, both operator-liquidated Sep4. Python gross returns3.805162958% and.752256770%, mean2.278709864%,2/2wins with no loss sample and no stop/target exits. Cannot estimate average loss, execution slippage or calibrated Kelly from this censored pair. Source runway:query_trades?fund=supply . Preserve independently recorded TEN3214 chronological test: train22,+19.190992%,DD49.046502%,win36.3636%;knownholdout11closed+1.924385%,stress-.293468%,marked+16.576931% includes open trade. https://runway.adityaperswal.workers.dev/data/stock-bars . No new backtest run or new rule trusted; selected name and previously viewed holdout remain limitations.

Gate3214 train22netreturns mean1.649875%,median-4.372739%,P05/P95 -10.027048/30.464151%;20kIIDmean95CI[-3.684452,7.796336]%,10kcircular3tradeblock[-3.751510,9.431886]%. Knownholdout11mean.487839%,median-4.672473%,IID[-3.925075,5.518712]%,block[-2.680960,2.559161]%. Seed91308; chronologicalholdoutkeptseparatebutalreadyseen, currentTENselectionbias. 4/22trainingtargets,1/11holdouttargets;stopinclBE17/22and8/11, medianholds14/7sessions. Two liveSupplywinners wereoperatorliquidations,0strategyexits and no executionquote baseline; cannot calibrate100%winrate orslippage. Positive arithmeticmean supports furtherboundedpolicywork; counterexampleis negative median and closedstressloss. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; livequery_trades fund=supply IDs3/6.

SIP alladjusted daily logreturns2020–Sep11; matching complete5series TEN,FRO,STNG,XLE,SPY. TrainN1257 TEN/FROcorrelation.672633,TEN/STNG.642277,TEN/XLE.425455;knownholdoutN424 .835756/.754700/.349823;last60 .822488/.752456/.142194. OLS TEN dailyreturn on intercept,FRO,SPY: trainbetaFRO.667135/SPY.268046,R2.460691,intercept-4.788881bp/day;heldout.706038/.211466,R2.705710,intercept2.067818bp/day;last60.698832/-.019706,R2.676522,intercept14.850864bp/day. Descriptive contemporaneous factor attribution; intercept not validated alpha, no significance/causal return forecast. Correlation can rise in stress; do not treat TEN/FRO two positions as diversified supply bets. Cross-sector XLE correlation falling does not identify another paid bottleneck automatically. Current book flat. Source https://runway.adityaperswal.workers.dev/data/stock-bars.

Primary cadence model3183. Training31 closed mean2.112847%,P05/median/P95 -13.885557/-7.312793/30.214313%;20k IID meanCI[-3.115923,7.696204]%,10k circular3tradeblockCI[-2.617690,6.994972]%. Knownholdout11 mean1.906418%,P05/median/P95 -7.107902/-5.564518/15.266447%;IIDCI[-3.505778,7.708049]%,blockCI[-1.634633,6.012300]%. Seed91306. Training outcomes all before2025; no boundary-spanning trades. Selection is current surviving TEN after recent strength, not random universe; holdout already examined in prior runs, so not fresh independent validation. 2/2 live winners operator-liquidated,0 strategy exits: no calibrated live Kelly or quote slippage. Counterexample to strong confidence is negative median and wideCI; positive arithmetic mean and payoff skew still permit investigating a small probe. Do not use certainty threshold or100%livewinrate. Source https://runway.adityaperswal.workers.dev/data/stock-bars.

Managedrawpapertrain29mean+.66741%/trade,IID20kbootstrap95%[-4.02761,5.78851]%;circular3tradeblock10kCI[-3.52885,5.13261]%. Knownholdout10mean6.01385%,IID[-.19308,11.98345]%,block[1.15657,10.98076]%; blockCIpositivewithonly10tradesdoesnotoutweighsample/modelselection. TrainP05/P50/P95=-13.0172/-6.7140/+28.1638%;holdout-8.0535/+8.4027/+18.1984%. Historicalsample iscurrentTENidentityselectedafterseeingstrongrecentperformance; regimesnotIIDandmissingsampledquotesremain. OwntradehistoryFRO/STNGN2bothprofitbutoperator-liquidatedSep4,notstrategyexits; no source contemporaneousquote means no measuredorderslippage. Do not convert100%observedwinrateintoKelly. Models andtradeobservations https://runway.adityaperswal.workers.dev/data/stock-bars ; https://runway.adityaperswal.workers.dev . Evidence needed: exact manager-time/rawquote implementation andprospectiveoutcome, not furtherlookbackselection.

Frozen dailytrainN33 netmean1.0437%,20,000IIDbootstrap95%CI[-3.7876,6.3721]%; holdoutN12mean5.4980%,CI[-.4618,11.4242]%. Training22stops/8targets/3time,holdout5/4/3. N424common2025–Sep11returnsTEN/FROcorr.835826; TEN/CNQ.276531. SPY+XLIOLS R2.0682 isdescriptive, notproofidiosyncraticalpha. ExistingFROwin andTENsetup areonegeopolitical/freightrisk, nottwoindependentbets. https://runway.adityaperswal.workers.dev/data/stock-bars . Positivepointestimates warrant prospectivevalidation, not treating12tradesasstable probability. BootstrapIID omitsregimeclustering; currentlyselectedticker biases screen.

Fixed L20 trainingN27 mean-2.1782%, IIDbootstrap20000 95%CI[-4.8701,+.8161]%; holdoutN14 mean+1.3471%,CI[-4.2810,+7.4645]%. Training20stop/gapstop vs2targets and5time exits; holdout9stop/gapstop vs5target/gaptarget. Paired daily2025–Sep11 N424: KMT/XLIcorr.5665; OLSintercept.5074bp/day, SPYbeta-.2713,XLIbeta1.4806,R2.3248. Descriptive selected sample, no robust alpha significance and no causal tungsten coefficient. Ownledger still twooperatorliquidations, no naturalstop/target exits;100%winN2 cannot calibrate new strategy or slippage. https://runway.adityaperswal.workers.dev/data/stock-bars ; query_tradesfund=supply. Premise1 smallconditionaledge; premise2 trainnegative; conclusion do not trade from holdoutalone. IIDresampling understates serial/event clustering and currentlistings create survivor selection.