Stats

Frozen protocol note213 before price inspection. Chronological later2023-Jan through2026-Sep4, already seen for other rules so NOT pristine independent validation. Endmonth deadlineD=T-(settlementlag+1), lag3/2/1 across SEC changes. Buy nextopen only if preceding5sessionreturn<0, max3open-to-open sessions; longonly full-capital research normalization, not actualfundallocation. Compare fixedhold to stop1laggedATR20/target1.5ATR20. Daily stopfirst when both hit, stopgap atopen, favorabletargetgap capped at target; same-day entries/exits supported through separate synthetic slots. Fees5bp+slippage5bp eachway, cashreturn0. SPY is primaryUSproxy; EFA/EEM transfer diagnostics do not establish underlyinglocalflows. Complete8052adjustedSIPbars,2684each,NYSEcalendar no missing/extra dates, OHLC/volume checks pass. Vectorbtfrom_orders results reconcile to independent compounded cash ledger for every36experiment terminal equity, errors<1e-6 on10000 initialcapital. Drawdown is open/close-marked, not intradaymax. Fixed return -2.359834%,DD9.357740%,N19,p52.6316%,W0.94873%,L1.30116%; range figures below. Range3-calendar-month-block10000bootstrap mean95CI [-0.5488546052653037,0.6886803891309617]%; inactive months retained. All trial metrics and uncertainty separately observed; code/raw inputs/fulltradeandequityledgers local artifacts. Conclusions concern this newly specified proxy, not exact reproduction of institutionalANcerno study. Mandatory15min/exthours/trailing-profitmanagement,rawcashdividends,banksettlementholiday differences and whole-share sizing unvalidated; no execution authorization. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://aaltodoc.aalto.fi/server/api/core/bitstreams/0f01d3c2-3b16-44f9-9135-3f03cdd29334/content ; https://www.sec.gov/exams/educationhelpguidesfaqs/t1-faq .

Frozen protocol note213 before price inspection. Training2016-2022. Endmonth deadlineD=T-(settlementlag+1), lag3/2/1 across SEC changes. Buy nextopen only if preceding5sessionreturn<0, max3open-to-open sessions; longonly full-capital research normalization, not actualfundallocation. Compare fixedhold to stop1laggedATR20/target1.5ATR20. Daily stopfirst when both hit, stopgap atopen, favorabletargetgap capped at target; same-day entries/exits supported through separate synthetic slots. Fees5bp+slippage5bp eachway, cashreturn0. SPY is primaryUSproxy; EFA/EEM transfer diagnostics do not establish underlyinglocalflows. Complete8052adjustedSIPbars,2684each,NYSEcalendar no missing/extra dates, OHLC/volume checks pass. Vectorbtfrom_orders results reconcile to independent compounded cash ledger for every36experiment terminal equity, errors<1e-6 on10000 initialcapital. Drawdown is open/close-marked, not intradaymax. Fixed return 3.260950%,DD11.791381%,N31,p54.8387%,W1.23598%,L1.24099%; range figures below. Range3-calendar-month-block10000bootstrap mean95CI [-0.6321942322129009,0.38874040945891103]%; inactive months retained. All trial metrics and uncertainty separately observed; code/raw inputs/fulltradeandequityledgers local artifacts. Conclusions concern this newly specified proxy, not exact reproduction of institutionalANcerno study. Mandatory15min/exthours/trailing-profitmanagement,rawcashdividends,banksettlementholiday differences and whole-share sizing unvalidated; no execution authorization. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://aaltodoc.aalto.fi/server/api/core/bitstreams/0f01d3c2-3b16-44f9-9135-3f03cdd29334/content ; https://www.sec.gov/exams/educationhelpguidesfaqs/t1-faq .

Frozen protocol note213 before price inspection. Chronological later2023-Jan through2026-Sep4, already seen for other rules so NOT pristine independent validation. Endmonth deadlineD=T-(settlementlag+1), lag3/2/1 across SEC changes. Buy nextopen only if preceding5sessionreturn<0, max3open-to-open sessions; longonly full-capital research normalization, not actualfundallocation. Compare fixedhold to stop1laggedATR20/target1.5ATR20. Daily stopfirst when both hit, stopgap atopen, favorabletargetgap capped at target; same-day entries/exits supported through separate synthetic slots. Fees5bp+slippage5bp eachway, cashreturn0. SPY is primaryUSproxy; EFA/EEM transfer diagnostics do not establish underlyinglocalflows. Complete8052adjustedSIPbars,2684each,NYSEcalendar no missing/extra dates, OHLC/volume checks pass. Vectorbtfrom_orders results reconcile to independent compounded cash ledger for every36experiment terminal equity, errors<1e-6 on10000 initialcapital. Drawdown is open/close-marked, not intradaymax. Fixed return 1.220813%,DD11.784379%,N19,p52.6316%,W1.31237%,L1.29674%; range figures below. Range3-calendar-month-block10000bootstrap mean95CI [-0.9636297394857071,0.5704831950615763]%; inactive months retained. All trial metrics and uncertainty separately observed; code/raw inputs/fulltradeandequityledgers local artifacts. Conclusions concern this newly specified proxy, not exact reproduction of institutionalANcerno study. Mandatory15min/exthours/trailing-profitmanagement,rawcashdividends,banksettlementholiday differences and whole-share sizing unvalidated; no execution authorization. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://aaltodoc.aalto.fi/server/api/core/bitstreams/0f01d3c2-3b16-44f9-9135-3f03cdd29334/content ; https://www.sec.gov/exams/educationhelpguidesfaqs/t1-faq .

Frozen protocol note213 before price inspection. Training2016-2022. Endmonth deadlineD=T-(settlementlag+1), lag3/2/1 across SEC changes. Buy nextopen only if preceding5sessionreturn<0, max3open-to-open sessions; longonly full-capital research normalization, not actualfundallocation. Compare fixedhold to stop1laggedATR20/target1.5ATR20. Daily stopfirst when both hit, stopgap atopen, favorabletargetgap capped at target; same-day entries/exits supported through separate synthetic slots. Fees5bp+slippage5bp eachway, cashreturn0. SPY is primaryUSproxy; EFA/EEM transfer diagnostics do not establish underlyinglocalflows. Complete8052adjustedSIPbars,2684each,NYSEcalendar no missing/extra dates, OHLC/volume checks pass. Vectorbtfrom_orders results reconcile to independent compounded cash ledger for every36experiment terminal equity, errors<1e-6 on10000 initialcapital. Drawdown is open/close-marked, not intradaymax. Fixed return 12.367856%,DD11.143295%,N34,p52.9412%,W1.81719%,L1.27186%; range figures below. Range3-calendar-month-block10000bootstrap mean95CI [-0.6868827262363107,0.40455541742822726]%; inactive months retained. All trial metrics and uncertainty separately observed; code/raw inputs/fulltradeandequityledgers local artifacts. Conclusions concern this newly specified proxy, not exact reproduction of institutionalANcerno study. Mandatory15min/exthours/trailing-profitmanagement,rawcashdividends,banksettlementholiday differences and whole-share sizing unvalidated; no execution authorization. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://aaltodoc.aalto.fi/server/api/core/bitstreams/0f01d3c2-3b16-44f9-9135-3f03cdd29334/content ; https://www.sec.gov/exams/educationhelpguidesfaqs/t1-faq .

Premises: stock signal profitability is necessary only for a directional thesis, never sufficient for option mispricing; next-close execution and costs must be included. Required CLI diagnostic run: node /app/agent/dist/backtest.js --symbols F,SOFI,IBIT --strategy mean-reversion --lookback20 --entry2 --exit0 --days365. CLI implementation inspected: exit0 unused; long only when z<=-entry, flat otherwise. Its unseparated251-bar outputs were F+0.79%,DD5.23%,win66.67%,N3; SOFI+7.07%,DD14.61%,win87.5%,N8; IBIT-25.80%,DD31.54%,win42.86%,N7. These are not adopted. Independent vectorbt replication uses20day population z, one-day delayed close execution, no same-close lookahead; trainJan2023-Dec14 2025 (IBITJan2024 start), held-outJan2026-Sep4. Sweep thresholds1.5,2,2.5 and all-in one-way cost10/30bp split fee/slippage; no parameter selected after viewing holdout. All12 per-symbol cells saved in observation. Figures show predeclared2-sigma10bp holdout; at30bp return -24.461%. Small samples, no option historical bid/ask execution, and profits in underlying cannot establish long-premium profits. Strong counterexample: rebound too small to recover spread/theta. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=IBIT&timeframe=1Day&start=2023-01-01&end=2026-09-05&adjustment=all&feed=iex . Reject using this proxy to size options.

Premises: stock signal profitability is necessary only for a directional thesis, never sufficient for option mispricing; next-close execution and costs must be included. Required CLI diagnostic run: node /app/agent/dist/backtest.js --symbols F,SOFI,IBIT --strategy mean-reversion --lookback20 --entry2 --exit0 --days365. CLI implementation inspected: exit0 unused; long only when z<=-entry, flat otherwise. Its unseparated251-bar outputs were F+0.79%,DD5.23%,win66.67%,N3; SOFI+7.07%,DD14.61%,win87.5%,N8; IBIT-25.80%,DD31.54%,win42.86%,N7. These are not adopted. Independent vectorbt replication uses20day population z, one-day delayed close execution, no same-close lookahead; trainJan2023-Dec14 2025 (IBITJan2024 start), held-outJan2026-Sep4. Sweep thresholds1.5,2,2.5 and all-in one-way cost10/30bp split fee/slippage; no parameter selected after viewing holdout. All12 per-symbol cells saved in observation. Figures show predeclared2-sigma10bp holdout; at30bp return 0.508%. Small samples, no option historical bid/ask execution, and profits in underlying cannot establish long-premium profits. Strong counterexample: rebound too small to recover spread/theta. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SOFI&timeframe=1Day&start=2023-01-01&end=2026-09-05&adjustment=all&feed=iex . Reject using this proxy to size options.

Premises: stock signal profitability is necessary only for a directional thesis, never sufficient for option mispricing; next-close execution and costs must be included. Required CLI diagnostic run: node /app/agent/dist/backtest.js --symbols F,SOFI,IBIT --strategy mean-reversion --lookback20 --entry2 --exit0 --days365. CLI implementation inspected: exit0 unused; long only when z<=-entry, flat otherwise. Its unseparated251-bar outputs were F+0.79%,DD5.23%,win66.67%,N3; SOFI+7.07%,DD14.61%,win87.5%,N8; IBIT-25.80%,DD31.54%,win42.86%,N7. These are not adopted. Independent vectorbt replication uses20day population z, one-day delayed close execution, no same-close lookahead; trainJan2023-Dec14 2025 (IBITJan2024 start), held-outJan2026-Sep4. Sweep thresholds1.5,2,2.5 and all-in one-way cost10/30bp split fee/slippage; no parameter selected after viewing holdout. All12 per-symbol cells saved in observation. Figures show predeclared2-sigma10bp holdout; at30bp return -1.650%. Small samples, no option historical bid/ask execution, and profits in underlying cannot establish long-premium profits. Strong counterexample: rebound too small to recover spread/theta. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=F&timeframe=1Day&start=2023-01-01&end=2026-09-05&adjustment=all&feed=iex . Reject using this proxy to size options.

Preregistered note210 before bars. SIP all-adjusted daily GEV/SPY start2024 toSep4 2026;610GEVbars startApr2 2024. Completed-close SMA10 crossing SMA50; nextopen entries/reverse-cross exits, longonly allcash. Vectorbt fees5bp/slippage10bp EACH side; stress10/20bp. 2024 training marked return82.6037%,DD9.2479%,ZERO closed/one terminalopen trade; cannot estimate training win/loss. Chronological2025+ holdout startsflat:129.1912%,DD27.205%,5naturalexits,40%wins,avgwin66.1459%,avgloss5.5673%; no terminalopen. Stress125.779%. Nine fast5/10/20 slow40/50/60 training combinations archived; fixed10/50 retained, not tuned. Independent cash arithmetic reconciles total return within3e-10percentagepoints; max tradecost difference4.04e-8dollars. CLIcontrol --symbols GEV --strategy sma-cross --fast10 --slow50 --days730 (actual flags separated in savedcli):501bars,134.4719%,DD27.7697%,win33.333%,6trades,buyhold375.2721%; different window/sameclose/noexplicitcosts, diagnostic only. Selected survivor, very small sample, no natural training payoffs; no range stop/target,+1R,15minute or fractional-session replay. Thus no trading authorization despite positive later performance. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; fullinputs/code/results archived observations2958–2997. This is a new GEV diagnostic, not revalidation of prior tankers.

Executed node /app/agent/dist/backtest.js --symbols XLB,XLE,XLF,XLI,XLK,XLP,XLU,XLV,XLY --strategy mean-reversion --lookback20 --entry2 --exit0 --days730 (actualflags spaced). All9complete502bars;per-symbolmetricsobservedandarchivedcli-summary.json. Thisis20daypricez-score<=-2,notcross-sector5dayranking. Sourceauditsamecloseentry/nofees/slippage,iex/noexplicitadjustments. exitparameter0doesnotproduceazscoreexitthreshold:strategyreturnsflatwhenz>-entry,engineexitsonsignal. CLIrawseries visiblychangeslevelnearcorporateactionperiods;unadjustedbuyholddropscannotbeinterpretedascorrecttotalreturns. Actualsplitdatesnotindependentlyissuerverifiedhere. Positivefrictionlesscontrolsareselecteddiagnostics,notnetOOSedge. Customtestexplicitadjustmentall/SIP/calendaralignmentandnextopen. Source https://runway.adityaperswal.workers.dev/data/stock-bars ;/app/agent/dist/backtest/{bars,strategies,engine}.js.

Frozen5dayformationremainsmain;predeclaredtraining-only3/10daylookbacksrecordedobservations1773-1776withN892/897,payoffs/DD;notselectedforholdout. Cost-onlyattributionnote162preservesrule,20totalexperimentmetrics1765-1784. Zero/2bpone-wayfixedholdout72.880/60.678%,range22.797/14.129%; rangezero/2bpmean95%CI[-.0178,.2502]/[-.0561,.2102]%. Theseoptimisticvariantsimproveheadlinebutnotconfidenceoractualexiteligibility. Costsareassumptionsand50bpstress,notassertedvenuefee;testedfees/slipssymmetric. Grossrangeavgtrade.120566% givesroughfirstorderbreak-even6bponewaybeforeparameter/estimationrisk,notan executableedgeestimate. Repeatedselectednamesareliquidated/reboughtbyfrozenrule;nettingmayreducecostbutrequiresnewseparateregistration,noteditedpastresult. Allvariantsvectorbt/independentcashledgererror<1e-6; source https://runway.adityaperswal.workers.dev/data/stock-bars .

RequiredCLI executed node /app/agent/dist/backtest.js --symbols GRAL --strategy breakout --lookback 20 --days 900 --stop-pct 5 --target-pct 10. IEX560bars,44trades,return5.70810%,DD24.29516%,win31.81818%,avgtrade.290269%,buyhold467.31449%. Engineenterssignalclose, usesexactstoppricewithoutgaploss, nofees/slippage,noseparateholdout; pricehistoryincludeswhen-issuedperiod. Comparisonwithcustomtest132notcontrolledsingle-variableexperiment: rule,feed,cost,timingandexitsdiffer. CLIpositiveoutputcannotvalidatepanelrun-uporrule158. No strategyselectedfromit. Data https://runway.adityaperswal.workers.dev/data/stock-bars ; engine /app/agent/dist/backtest/engine.js read; cli.jsonandexactengineassumptionsrecorded. Return/DDpercenton100%abstractcapital; notactualfundprofit. OutcomeREJECTforentryvalidation.

Required CLI executed: node /app/agent/dist/backtest.js --symbols BTC/USD,ETH/USD,SOL/USD --strategy rsi --days978 --stop-pct4 --target-pct8 (actualflagsseparated). Built-in CLI exposes price strategies, no fundinghistoryinput; this is a price-reversion control, NOT replication of registeredfundingrule. Engine samebarcloseentry,no explicitcosts, currentdayincluded,no independentchronologicalholdout; exactpercentagebarriers lackgapfill/15minmanagement. No parameterchosenfromit. Custom vectorbt tests127-129 implementregistered externalfundingwithdaily/hourlybars,nextopen,costs andfixedsplit; theirnegativeverdictcontrols. CLI measuredsummary {"symbol": "BTC/USD", "bars": 979, "returnPct": 2.2415344708743623, "buyAndHoldPct": 80.03656269903307, "maxDrawdownPct": 26.456840736476657, "winRate": 52.77777777777778, "avgTradePct": 0.14347217951893762, "sample": 36} {"symbol": "ETH/USD", "bars": 979, "returnPct": -23.895629029964528, "buyAndHoldPct": 4.268071789682248, "maxDrawdownPct": 31.919598465806025, "winRate": 42, "avgTradePct": -0.4697585757388448, "sample": 50} {"symbol": "SOL/USD", "bars": 741, "returnPct": 13.882398172950804, "buyAndHoldPct": -35.233704292527825, "maxDrawdownPct": 25.914827218441012, "winRate": 48.93617021276596, "avgTradePct": 0.3897149986778772, "sample": 47} Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; /app/agent/dist/backtest/engine.js . Do not interpret differences betweenRSIandfundingtestsas isolatingonebias.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Freeze signal ADR20:stop1ADR,target2ADR,conservative stopfirst/adversegap fills; includeentryday. Close>=+1R schedules1/3trimnextopen andbreakevenstop; maximum5sessions. 2023-Jan3 through2026-Sep4 chronological diagnostic holdout,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee25bp+slippage25bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-80.2992%,DD80.3135%,N549,p32.9690%,W1.3415%,L1.9662%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Freeze signal ADR20:stop1ADR,target2ADR,conservative stopfirst/adversegap fills; includeentryday. Close>=+1R schedules1/3trimnextopen andbreakevenstop; maximum5sessions. 2023-Jan3 through2026-Sep4 chronological diagnostic holdout,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee5bp+slippage5bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-14.8384%,DD16.9591%,N549,p45.1730%,W1.6719%,L1.5224%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Hold5sessions thenexitopen; no mandatory stops/targets so notvenueeligible. 2023-Jan3 through2026-Sep4 chronological diagnostic holdout,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee25bp+slippage25bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-72.2625%,DD73.0675%,N549,p38.7978%,W1.6277%,L2.1422%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Hold5sessions thenexitopen; no mandatory stops/targets so notvenueeligible. 2023-Jan3 through2026-Sep4 chronological diagnostic holdout,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee5bp+slippage5bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns19.8968%,DD18.4248%,N549,p54.4627%,W1.8538%,L1.9576%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Freeze signal ADR20:stop1ADR,target2ADR,conservative stopfirst/adversegap fills; includeentryday. Close>=+1R schedules1/3trimnextopen andbreakevenstop; maximum5sessions. 2017-2022 training,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee25bp+slippage25bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-92.3758%,DD92.5984%,N896,p32.4777%,W1.6771%,L2.0536%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Freeze signal ADR20:stop1ADR,target2ADR,conservative stopfirst/adversegap fills; includeentryday. Close>=+1R schedules1/3trimnextopen andbreakevenstop; maximum5sessions. 2017-2022 training,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee5bp+slippage5bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-16.8486%,DD31.2305%,N896,p44.7545%,W1.9125%,L1.6316%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Hold5sessions thenexitopen; no mandatory stops/targets so notvenueeligible. 2017-2022 training,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee25bp+slippage25bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-92.7967%,DD92.7967%,N896,p36.7188%,W2.0795%,L2.5193%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Frozen note161. Nine original sector ETFs, completed5session returns less equal-sector mean; every5sessions buy bottom3 negative residuals in equalcash sleeves at followingopen, unusedslotscash. Hold5sessions thenexitopen; no mandatory stops/targets so notvenueeligible. 2017-2022 training,2016warmup; datesoverlapotherresearch, notgloballyuntouchedsample. Fee5bp+slippage5bp eachside are assumed, notlivecostmeasurement. vectorbt explicitlongonlyorders,cashsharing; independentcash/quantitydailyledger agrees<1e-6. Synthetic4event/dayordering isNOTobserved intradaydata; reportedDDdailyclose. Finalincompleterotation excluded, zeroopenpositions/terminalmarks inpayoffs. Returns-21.4363%,DD52.7304%,N896,p50.7813%,W2.2027%,L2.3420%. Populationiscurrentninewrappersnotpoint-in-timeallETF universe; originalNagel48industrylongshortdailyweights differ. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.nber.org/system/files/working_papers/w17653/w17653.pdf . Conclusion specificimplementation, notallreversaltheory; noadoption/tuningfromholdout. Actual15minute/dividend/whole-shareexecution unvalidated.

Executed node /app/agent/dist/backtest.js --symbols TSN,PPC,JBS --strategy breakout --lookback 20 --days 730 --stop-pct 3 --target-pct 6. TSN502bars return-1.468%,DD8.739%,win36.364%,N22;PPC502 return-13.312%,DD22.030%,win24.138%,N29;JBS309 return5.029%,DD12.038%,win40%,N25. Sameclose/noexplicitcosts, differentrule andnoindependentholdout, so notexecutablevalidation. Registeredtests133–135 useexplicitfees/slippage andnextopen; no parameterchosenfromCLI. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; cli.json archived.

Registered note159 before pricefetch: prior20 completed-close breakout AND SMA50; nextopen entry, signalADR20 stop1/target2, 5sessions exitnextopen, no pyramiding; entryday barriers included, stopfirst both-touch, adversegapactualopen/favorabletargetgapcapped. Training2022–2024, chronological diagnostic holdout2025–Sep4 2026;2021warmup. Fees5bp andslippage10bp eachside separately in finalvectorbt replay; stress10/20bp. Currentselected names make this exploratory, not prospective or causal policy study. All-adjusted SIP bars paginated, datesunique; synthetic9h/16h rows encode ordering, not observed intradaybars. Independentcashledger reconciles below2e-12; terminalopenfalse; naturalexitpayoffs only. Trainingtarget1.5/2 stop1/1.25 sweep saved, fixed1/2 unchanged. 309bars startJune13 2025; no training, so laterperiod not independently trained OOS. Positive estimate does not justify adoption. No actual15minute/+1R/rawdividend/whole-share replay, so daily barrier model is ineligible. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; artifacts/test.py/results.json/trades.json and archive. QuarterKelly authorized0, no current signal.

Registered note159 before pricefetch: prior20 completed-close breakout AND SMA50; nextopen entry, signalADR20 stop1/target2, 5sessions exitnextopen, no pyramiding; entryday barriers included, stopfirst both-touch, adversegapactualopen/favorabletargetgapcapped. Training2022–2024, chronological diagnostic holdout2025–Sep4 2026;2021warmup. Fees5bp andslippage10bp eachside separately in finalvectorbt replay; stress10/20bp. Currentselected names make this exploratory, not prospective or causal policy study. All-adjusted SIP bars paginated, datesunique; synthetic9h/16h rows encode ordering, not observed intradaybars. Independentcashledger reconciles below2e-12; terminalopenfalse; naturalexitpayoffs only. Trainingtarget1.5/2 stop1/1.25 sweep saved, fixed1/2 unchanged. Training/holdout result does not establish positive stable expectancy; reject implementation, not all supply signals. No actual15minute/+1R/rawdividend/whole-share replay, so daily barrier model is ineligible. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; artifacts/test.py/results.json/trades.json and archive. QuarterKelly authorized0, no current signal.

Registered note159 before pricefetch: prior20 completed-close breakout AND SMA50; nextopen entry, signalADR20 stop1/target2, 5sessions exitnextopen, no pyramiding; entryday barriers included, stopfirst both-touch, adversegapactualopen/favorabletargetgapcapped. Training2022–2024, chronological diagnostic holdout2025–Sep4 2026;2021warmup. Fees5bp andslippage10bp eachside separately in finalvectorbt replay; stress10/20bp. Currentselected names make this exploratory, not prospective or causal policy study. All-adjusted SIP bars paginated, datesunique; synthetic9h/16h rows encode ordering, not observed intradaybars. Independentcashledger reconciles below2e-12; terminalopenfalse; naturalexitpayoffs only. Trainingtarget1.5/2 stop1/1.25 sweep saved, fixed1/2 unchanged. Training/holdout result does not establish positive stable expectancy; reject implementation, not all supply signals. No actual15minute/+1R/rawdividend/whole-share replay, so daily barrier model is ineligible. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; artifacts/test.py/results.json/trades.json and archive. QuarterKelly authorized0, no current signal.

Preregisterednote158beforeprice retrieval; generic pricefalsification NOTpanelbacktest. SIPalladjustedpaginated560GRALbarsJun12 2024-Sep4 2026,SPY672fromJan2024. Fifty-barwarmup excludesearlywhen-issuednoise. Signalcompletedclose>prior20closes ANDSMA50,next-openbuy; simplesignalATR14frozenstop1ATR/target2ATR,fivesessions timeoutatnextopen; nooverlap;entrydaybarriers,badgapopen/favorablegaptarget/both-touchstopfirst. Fee5bp+slippage5bp eachside; standalonecashledger reconcilesvectorbt explicitlongonlyorders to<4e-12. SyntheticroworderingNOTobservedminuteexecution. Training390bars/28trades+23.4898%,DD26.8884%,p46.43%,W11.3945%,L7.5400%. Holdout170bars/13trades-36.2152%,DD39.0181%,p30.7692%,W7.54887%,L7.74074%,mean-3.03624%,quarterKelly0;noopenmarks. Stress25bpfee+25bpslip eachsideholdout-42.5155%,DD44.821%,same13outcomes. Training-onlysweepstop1/1.25,target1.5/2produces23.49–112.94%; keptregistered1/2,noholdoutselection. ReturnandDDare100%abstractcapitalstrategy, not$49.14actualfund; dailycloseDDcouldmissintradayshocks. No+1Rpartial/actual15minuteexits,noeventconditionedselection; evenpositivecouldnotauthorizepanelentry. Thisnegativeholdoutrejectsthisimplementation, notallcatalystrunups. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; artifacts/test.py,tests.json; rawGRALSHA9459602a452562126bbf0d5db729213359f239a125fdfcc1dca061757f2fd4d8;SPYSHA7ecf3b2c2b253af24930d116e16f2c6635dc8fe8049167b3687025121a34b1d6. Quarterlydates andpanelprobabilitiesnotretrofitted topricehistory.