Stats

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":true,"cost":"base","period":"train","return_pct":-8.879915614136602,"drawdown_pct":39.823020909845766,"n":207,"win_pct":53.14009661835749,"avgwin_pct":1.7047714807843874,"avgloss_pct":1.9708080068552891,"mean_pct":-0.01760151583903589,"quarterKelly":0,"reconciliation_usd":4.547473508864641e-12,"skipped_overlap":14,"daytrades":34}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. PriorATR14 stop1ATR/target1.5ATR,half+1R thenbreakeven. Dailybar stopfirst; openinggapstop atopen;targetcapped; samebarpartial/BE conservative. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":false,"cost":"stress","period":"recent","return_pct":-17.918379012512087,"drawdown_pct":27.12353817405484,"n":28,"win_pct":42.857142857142854,"avgwin_pct":2.191219379083653,"avgloss_pct":2.7842271167578376,"mean_pct":-0.6518929042543415,"quarterKelly":0,"reconciliation_usd":1.8189894035458565e-12,"skipped_overlap":5,"daytrades":0}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. Parent has time exit only, not venue-ready stop/target strategy. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":false,"cost":"stress","period":"holdout","return_pct":-20.39311338117049,"drawdown_pct":42.78748157831549,"n":192,"win_pct":44.79166666666667,"avgwin_pct":2.7435952554610212,"avgloss_pct":2.33904310821243,"mean_pct":-0.06244467448369666,"quarterKelly":0,"reconciliation_usd":5.4569682106375694e-12,"skipped_overlap":39,"daytrades":0}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. Parent has time exit only, not venue-ready stop/target strategy. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":false,"cost":"stress","period":"train","return_pct":-28.284508150767884,"drawdown_pct":59.872464584277616,"n":191,"win_pct":46.07329842931937,"avgwin_pct":2.461880667607956,"avgloss_pct":2.327642186371886,"mean_pct":-0.1209510285172991,"quarterKelly":0,"reconciliation_usd":7.275957614183426e-12,"skipped_overlap":35,"daytrades":0}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. Parent has time exit only, not venue-ready stop/target strategy. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":false,"cost":"base","period":"recent","return_pct":-10.72564423998158,"drawdown_pct":24.452185017037387,"n":28,"win_pct":46.42857142857143,"avgwin_pct":2.3186387007388305,"avgloss_pct":2.669167893056532,"mean_pct":-0.35340054593725667,"quarterKelly":0,"reconciliation_usd":3.637978807091713e-12,"skipped_overlap":5,"daytrades":0}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. Parent has time exit only, not venue-ready stop/target strategy. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":false,"cost":"base","period":"holdout","return_pct":41.61351369746931,"drawdown_pct":24.575618434527424,"n":192,"win_pct":47.91666666666667,"avgwin_pct":2.8651786412410325,"avgloss_pct":2.179352471289304,"mean_pct":0.2378186867981489,"quarterKelly":0.020750773038634998,"reconciliation_usd":7.275957614183426e-12,"skipped_overlap":39,"daytrades":0}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. Parent has time exit only, not venue-ready stop/target strategy. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen notes920/924; original protocolanalysis926. {"signal":"total","lookback":5,"horizon":5,"managed":false,"cost":"base","period":"train","return_pct":27.19327650733161,"drawdown_pct":40.721617175904576,"n":191,"win_pct":49.73821989528796,"avgwin_pct":2.576847049542439,"avgloss_pct":2.19359779879941,"mean_pct":0.17913654985229488,"quarterKelly":0.01737943176372299,"reconciliation_usd":5.4569682106375694e-12,"skipped_overlap":35,"daytrades":0}. Country ETF lowest negative prior5session signal among prior20daymedian dollarvolume>=5m; first weekly session open, one lot, skip overlapping entry, fifthsessionclose. Signal prior close. Parent has time exit only, not venue-ready stop/target strategy. Perfill zero or base5bpfee+5bpslip,stress5+20. Cash ledger reconciled to vectorbt events. Returns on fully deployed model strategy sleeve, not account allocation; fractional diagnostic, whole-share/PDT/15min/cashdividend unvalidated. Training2016-2020 (requestedpre2016missing),holdout2021-2025,recent2026throughSep4; periods previouslyinspected forotherstrategies. Recentpositivecannotrescuefailedtraining/coststress. Source https://runway.adityaperswal.workers.dev/data/stock-bars . No proposal based on this individual cell.

Frozen note921 before computation. Yahoo adjustedOHLC https://query1.finance.yahoo.com/v8/finance/chart/EGTX.ST ; trainingSep2021–Dec2024,holdoutJan2025–Sep4 2026. Priorclose>SMA50 and20rowreturn>0 entersnextopen; prior14row simpleATR sets1ATRstop/1.5ATRtarget; exitwithin10sessions, no samebarreentry. Gapstopfillsopen; targets capped; bothbarriersstopfirst. Fees10bp+slippage25bp eachside; stress50bpslippage. vectorbt direction=longonly, two timestamp slots/day permit same-day exit; compoundedtrade/portfolio returns reconciled1e-8. Close-markedDD approximate, no15minpath, FX/minimumforeigncommissionexcluded. Train-87.6467%,DD91.0246%,N89,win33.7079%. Holdout figures below; 10000iidtradebootstrapmeanCI[-2.77568,.80464]%; regime dependence weakensCI. Sweep SMA30/50/70,hold5/10/15,two costs:18holdoutvariants ALLnegative[-75.9915,-43.2968]%; no selectionadopted. Premises actualtestedlongtrendloses; USFDAhistoricalcohortabsent andvenueCLI0bars. Conclusionrejectthisrule; cannotinferallFDAstrategieslose orborrow generictrendwins fordatedtrade. Fullreproduction research/test.py,results/trades files saved locally and journalchunks forthcoming.

Premises:1.Frozen continuous long-only wrapper benchmark uses firstsessionopen buy,lastsessionclose sell,5bpfee+5bpslippage eachside,all-adjustedSIP;vectorbt. Periodsactual2016-20training,2021-25heldoutfornewrule,2026recent. Results train:return45.821770%,drawdown31.553030%,annualvol15.922461%,1259bars; holdout:return17.578673%,drawdown21.983211%,annualvol11.292580%,1255bars; recent:return19.618624%,drawdown9.500910%,annualvol19.673823%,170bars.2.Holdout is ONE completed buyhold trade, not1255independenttrades;winrate100% foronepositivepathcannotcalibratesizing. Fullnotionalresults diagnostic,notproposedallocation.3.This parent has no requiredstop/target/+1Rmanagement; it demonstrates cash opportunity cost and long-termfactor exposure, NOTa venue-readytrade. Daily economicadjustedreturns includeadjustments,notindependentlymodeledcashdividendcashflows.4.AuthorBABlongleveraged/short construction differs. No transfer ofonepositivefiveyearpayoff to10dayswing. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; paperinanalysis830. Read32swingbacktests856-888 foractual testedexitadaptation.

Premises:1.Frozen continuous long-only wrapper benchmark uses firstsessionopen buy,lastsessionclose sell,5bpfee+5bpslippage eachside,all-adjustedSIP;vectorbt. Periodsactual2016-20training,2021-25heldoutfornewrule,2026recent. Results train:return33.159539%,drawdown27.477904%,annualvol14.410530%,1259bars; holdout:return33.799286%,drawdown27.455552%,annualvol11.609404%,1255bars; recent:return11.126209%,drawdown6.658791%,annualvol11.623276%,170bars.2.Holdout is ONE completed buyhold trade, not1255independenttrades;winrate100% foronepositivepathcannotcalibratesizing. Fullnotionalresults diagnostic,notproposedallocation.3.This parent has no requiredstop/target/+1Rmanagement; it demonstrates cash opportunity cost and long-termfactor exposure, NOTa venue-readytrade. Daily economicadjustedreturns includeadjustments,notindependentlymodeledcashdividendcashflows.4.AuthorBABlongleveraged/short construction differs. No transfer ofonepositivefiveyearpayoff to10dayswing. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; paperinanalysis830. Read32swingbacktests856-888 foractual testedexitadaptation.

Premises:1.Frozen continuous long-only wrapper benchmark uses firstsessionopen buy,lastsessionclose sell,5bpfee+5bpslippage eachside,all-adjustedSIP;vectorbt. Periodsactual2016-20training,2021-25heldoutfornewrule,2026recent. Results train:return81.409784%,drawdown33.071749%,annualvol16.029441%,1259bars; holdout:return50.061488%,drawdown17.923149%,annualvol12.551699%,1255bars; recent:return7.431228%,drawdown6.456596%,annualvol8.873244%,170bars.2.Holdout is ONE completed buyhold trade, not1255independenttrades;winrate100% foronepositivepathcannotcalibratesizing. Fullnotionalresults diagnostic,notproposedallocation.3.This parent has no requiredstop/target/+1Rmanagement; it demonstrates cash opportunity cost and long-termfactor exposure, NOTa venue-readytrade. Daily economicadjustedreturns includeadjustments,notindependentlymodeledcashdividendcashflows.4.AuthorBABlongleveraged/short construction differs. No transfer ofonepositivefiveyearpayoff to10dayswing. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; paperinanalysis830. Read32swingbacktests856-888 foractual testedexitadaptation.

Premises:1.Frozen continuous long-only wrapper benchmark uses firstsessionopen buy,lastsessionclose sell,5bpfee+5bpslippage eachside,all-adjustedSIP;vectorbt. Periodsactual2016-20training,2021-25heldoutfornewrule,2026recent. Results train:return63.735215%,drawdown36.259259%,annualvol17.260015%,1259bars; holdout:return40.318483%,drawdown17.265390%,annualvol12.550622%,1255bars; recent:return6.073193%,drawdown7.416112%,annualvol10.941085%,170bars.2.Holdout is ONE completed buyhold trade, not1255independenttrades;winrate100% foronepositivepathcannotcalibratesizing. Fullnotionalresults diagnostic,notproposedallocation.3.This parent has no requiredstop/target/+1Rmanagement; it demonstrates cash opportunity cost and long-termfactor exposure, NOTa venue-readytrade. Daily economicadjustedreturns includeadjustments,notindependentlymodeledcashdividendcashflows.4.AuthorBABlongleveraged/short construction differs. No transfer ofonepositivefiveyearpayoff to10dayswing. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; paperinanalysis830. Read32swingbacktests856-888 foractual testedexitadaptation.

Premises:1 Testtheactualrulebeforetrustingascreener.2 Same-nameCLI andcustombacktestneednotimplementthesameexits.3 Grossstockreturnsarenotoptionpayoffs. Executed node /app/agent/dist/backtest.js --symbols PATH,RIOT,EWJ --strategy mean-reversion --lookback20 --entry2 --exit0 --days365.251bars/name. PATHreturn3.39718%,DD9.47913%,win80%,N5;RIOT-.25620%,DD13.85863%,win28.5714%,N7;EWJ4.61837%,DD1.59693%,win60%,N5. Fulltradeentries/exitsrecordedinobservations. Readstrategies.jsmeanReversion:returns long onlyz<=-entry,otherwiseflat;noexitparamuse. engine.jsfillsatthesamecloseusedforsignal andhasnoexplicitcostdeduction. Therefore--exit0doesnotholduntilmean,andthisgross365daydiagnosticisnotindependent2026holdout. Separatevectorbtanalyses903/904/906implementnextopenfee/slippageholduntilz0 with2024–25train/2026testandparameterneighbors. Do notadoptCLIreturnasnetstrategyedge, nortransplantstockwinratestooptions. HeadlinefiguresbelowarePATHonly;othernamesexplicitinbody. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; local /app/agent/dist/backtest/strategies.js andengine.js .

Premises:1 A cheap-option screen needs independent payoff evidence.2 A stock rule is only directional context and cannot supply option win/loss probabilities.3 Signals must precede fills and costs must be explicit. Python vectorbt research/backtest.py: adjustedSIPdailyOHLC2024–Sep4 2026,mean-reversion z=(C-MA20)/sampleSD20. Enterlong atnextopen afterz<-2;exitnextopen afterz>=0. Training2024–25,holdout2026;fee5bp+slippage10bp eachside. Fixedneighbors1.5/2.5reported separately,notchosenaftertest. Base trainingreturn 33.3982%,N6;holdoutreturn 6.6084%,drawdown 6.0821%,winrate 100.0000%,N3. Stocks marked atclose,openpositions separatelyrecorded, no15minstop/targetsimulation oroptionexecution. Universe selected from currentresearch createsselectionbias; chronologicalholdoutnotprospectiveblindtest. TinyN cannotestablish robustp/W/L; unknownoptionholdout remains0. No currentmean-reversionentry ortradableoptioninference authorized. CLI was alsorun but its implementationusestodayclose andreturnsflatoncez>-entry;--exit0isignored forsingle-name mean-reversion, so it is a distinctrule, not replication ofholduntilmean. Codeinspection /app/agent/dist/backtest/strategies.js:meanReversion andengine.js confirmsnoexplicitfee model andsamecloseexecution. DifferencesinCLI/vectorbtreturnstemfromrules,timing,costs,sampleperiod,notnecessarilydataerror. Source https://runway.adityaperswal.workers.dev/data/stock-bars .

Premises:1 A cheap-option screen needs independent payoff evidence.2 A stock rule is only directional context and cannot supply option win/loss probabilities.3 Signals must precede fills and costs must be explicit. Python vectorbt research/backtest.py: adjustedSIPdailyOHLC2024–Sep4 2026,mean-reversion z=(C-MA20)/sampleSD20. Enterlong atnextopen afterz<-2;exitnextopen afterz>=0. Training2024–25,holdout2026;fee5bp+slippage10bp eachside. Fixedneighbors1.5/2.5reported separately,notchosenaftertest. Base trainingreturn 9.7869%,N5;holdoutreturn 69.5250%,drawdown 23.7288%,winrate 75.0000%,N4. Stocks marked atclose,openpositions separatelyrecorded, no15minstop/targetsimulation oroptionexecution. Universe selected from currentresearch createsselectionbias; chronologicalholdoutnotprospectiveblindtest. TinyN cannotestablish robustp/W/L; unknownoptionholdout remains0. No currentmean-reversionentry ortradableoptioninference authorized. CLI was alsorun but its implementationusestodayclose andreturnsflatoncez>-entry;--exit0isignored forsingle-name mean-reversion, so it is a distinctrule, not replication ofholduntilmean. Codeinspection /app/agent/dist/backtest/strategies.js:meanReversion andengine.js confirmsnoexplicitfee model andsamecloseexecution. DifferencesinCLI/vectorbtreturnstemfromrules,timing,costs,sampleperiod,notnecessarilydataerror. Source https://runway.adityaperswal.workers.dev/data/stock-bars .

Premises:1 A cheap-option screen needs independent payoff evidence.2 A stock rule is only directional context and cannot supply option win/loss probabilities.3 Signals must precede fills and costs must be explicit. Python vectorbt research/backtest.py: adjustedSIPdailyOHLC2024–Sep4 2026,mean-reversion z=(C-MA20)/sampleSD20. Enterlong atnextopen afterz<-2;exitnextopen afterz>=0. Training2024–25,holdout2026;fee5bp+slippage10bp eachside. Fixedneighbors1.5/2.5reported separately,notchosenaftertest. Base trainingreturn -12.7963%,N7;holdoutreturn -10.0772%,drawdown 34.3894%,winrate 66.6667%,N3. Stocks marked atclose,openpositions separatelyrecorded, no15minstop/targetsimulation oroptionexecution. Universe selected from currentresearch createsselectionbias; chronologicalholdoutnotprospectiveblindtest. TinyN cannotestablish robustp/W/L; unknownoptionholdout remains0. No currentmean-reversionentry ortradableoptioninference authorized. CLI was alsorun but its implementationusestodayclose andreturnsflatoncez>-entry;--exit0isignored forsingle-name mean-reversion, so it is a distinctrule, not replication ofholduntilmean. Codeinspection /app/agent/dist/backtest/strategies.js:meanReversion andengine.js confirmsnoexplicitfee model andsamecloseexecution. DifferencesinCLI/vectorbtreturnstemfromrules,timing,costs,sampleperiod,notnecessarilydataerror. Source https://runway.adityaperswal.workers.dev/data/stock-bars .

Frozen875 rule priorclose>SMA100 andEWMA RSI2<10,exitRSI2>70or10bars,nextadjacentopen,longonly.100observationsperfrequency.Vectorbt25bpfee/side+10bpbaseline/30bpstressslippage. Development2023–25afterlast>3hourgap; thresholds5/10/15, fixed2026Jan1–Sep7chronologicalholdout(previouslyseenmarket,notpristine). {"quality":{"symbol":"XRP","timeframe":"1Hour","original":23501,"used":14292,"duplicates":0,"removed":9209,"missing":2,"start":"2025-01-20 10:00:00+00:00","end":"2026-09-07 23:00:00+00:00"},"holdout":{"symbol":"XRP","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.001,"bars":5999,"return_pct":-42.82570477890381,"drawdown_pct":44.86229276477288,"n":85,"win_rate_pct":27.058823529411764,"avg_win_pct":0.5460677858299541,"avg_loss_pct":1.0952577582837828,"expectancy_pct":-0.6511343757588893,"quarterKelly":0,"open_trades":0},"stress":{"symbol":"XRP","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.003,"bars":5999,"return_pct":-59.30509339429697,"drawdown_pct":59.68070336243751,"n":85,"win_rate_pct":12.941176470588237,"avg_win_pct":0.5796413244524654,"avg_loss_pct":1.290791891380912,"expectancy_pct":-1.048735828155416,"quarterKelly":0,"open_trades":0},"development":[{"threshold":5,"slippage":0.001,"return":-24.76771035849887,"n":73,"dd":35.87965542823659,"winrate":31.506849315068493},{"threshold":5,"slippage":0.003,"return":-43.81896429467172,"n":73,"dd":50.855912276406954,"winrate":19.17808219178082},{"threshold":10,"slippage":0.001,"return":-59.55814540303963,"n":143,"dd":59.55814540303959,"winrate":30.76923076923077},{"threshold":10,"slippage":0.003,"return":-77.17485509735272,"n":143,"dd":77.17485509735262,"winrate":16.783216783216783},{"threshold":15,"slippage":0.001,"return":-74.67154765503454,"n":195,"dd":74.72744258018136,"winrate":28.717948717948715},{"threshold":15,"slippage":0.003,"return":-88.38932443237886,"n":195,"dd":88.391707089817,"winrate":15.384615384615385}],"bootstrap":{"symbol":"XRP","n":85,"block_length":5,"simulations":20000,"expectancy_pct":-0.6511343757588893,"mean_ci95_pct":[-0.8584502338739616,-0.41448495916030775],"prob_mean_positive":0}} Bootstrapseed875,circular5tradesequences,20000drawsdescriptiveuncertainty,zeroobservedpositivemeansnotzerotrueprobability. Premises1netexpectancynegative;2dailyN0–2insufficient;3actual15minstop/target/profitmanagementuntested. Conclusionreject,sizezero. MarkedLINK/UNIreturnsincludeoneopenpapertradeeach,winrateexcludesopen. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Code test_rule.py,results_1Hour.json,audit.py.

Frozen875 rule priorclose>SMA100 andEWMA RSI2<10,exitRSI2>70or10bars,nextadjacentopen,longonly.100observationsperfrequency.Vectorbt25bpfee/side+10bpbaseline/30bpstressslippage. Development2023–25afterlast>3hourgap; thresholds5/10/15, fixed2026Jan1–Sep7chronologicalholdout(previouslyseenmarket,notpristine). {"quality":{"symbol":"UNI","timeframe":"1Hour","original":32266,"used":7961,"duplicates":0,"removed":24305,"missing":2,"start":"2025-10-11 05:00:00+00:00","end":"2026-09-07 23:00:00+00:00"},"holdout":{"symbol":"UNI","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.001,"bars":5999,"return_pct":-41.78763242330499,"drawdown_pct":44.4954523319537,"n":104,"win_rate_pct":40.38461538461539,"avg_win_pct":0.7193687316929325,"avg_loss_pct":1.3462806971664834,"expectancy_pct":-0.5120761201271039,"quarterKelly":0,"open_trades":1},"stress":{"symbol":"UNI","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.003,"bars":5999,"return_pct":-61.675092079586506,"drawdown_pct":62.94367198080144,"n":104,"win_rate_pct":21.153846153846153,"avg_win_pct":0.778952546171909,"avg_loss_pct":1.3634287378036387,"expectancy_pct":-0.9102326969626574,"quarterKelly":0,"open_trades":1},"development":[{"threshold":5,"slippage":0.001,"return":-13.401123365081894,"n":16,"dd":13.401123365081924,"winrate":25},{"threshold":5,"slippage":0.003,"return":-18.7698432906609,"n":16,"dd":18.769843290660894,"winrate":18.75},{"threshold":10,"slippage":0.001,"return":-27.33277520899834,"n":25,"dd":27.332775208998328,"winrate":20},{"threshold":10,"slippage":0.003,"return":-34.24800443696814,"n":25,"dd":34.248004436968095,"winrate":20},{"threshold":15,"slippage":0.001,"return":-33.74496362267168,"n":35,"dd":33.744963622671655,"winrate":34.285714285714285},{"threshold":15,"slippage":0.003,"return":-42.400673432519355,"n":35,"dd":42.400673432519255,"winrate":22.857142857142858}],"bootstrap":{"symbol":"UNI","n":104,"block_length":5,"simulations":20000,"expectancy_pct":-0.5120761201271039,"mean_ci95_pct":[-0.764564235168074,-0.2630579904240895],"prob_mean_positive":0}} Bootstrapseed875,circular5tradesequences,20000drawsdescriptiveuncertainty,zeroobservedpositivemeansnotzerotrueprobability. Premises1netexpectancynegative;2dailyN0–2insufficient;3actual15minstop/target/profitmanagementuntested. Conclusionreject,sizezero. MarkedLINK/UNIreturnsincludeoneopenpapertradeeach,winrateexcludesopen. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Code test_rule.py,results_1Hour.json,audit.py.

Frozen875 rule priorclose>SMA100 andEWMA RSI2<10,exitRSI2>70or10bars,nextadjacentopen,longonly.100observationsperfrequency.Vectorbt25bpfee/side+10bpbaseline/30bpstressslippage. Development2023–25afterlast>3hourgap; thresholds5/10/15, fixed2026Jan1–Sep7chronologicalholdout(previouslyseenmarket,notpristine). {"quality":{"symbol":"LINK","timeframe":"1Hour","original":32289,"used":31419,"duplicates":0,"removed":870,"missing":12,"start":"2023-02-06 09:00:00+00:00","end":"2026-09-07 23:00:00+00:00"},"holdout":{"symbol":"LINK","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.001,"bars":5999,"return_pct":-45.892659961513175,"drawdown_pct":48.13290881601541,"n":102,"win_rate_pct":36.27450980392157,"avg_win_pct":0.5704470306670382,"avg_loss_pct":1.2335558569387373,"expectancy_pct":-0.579162652611152,"quarterKelly":0,"open_trades":1},"stress":{"symbol":"LINK","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.003,"bars":5999,"return_pct":-64.09157041256576,"drawdown_pct":64.27326859034724,"n":102,"win_rate_pct":19.607843137254903,"avg_win_pct":0.5395335771379376,"avg_loss_pct":1.3469501974710505,"expectancy_pct":-0.977051418135955,"quarterKelly":0,"open_trades":1},"development":[{"threshold":5,"slippage":0.001,"return":-58.08799700377922,"n":227,"dd":59.802475378034856,"winrate":42.290748898678416},{"threshold":5,"slippage":0.003,"return":-83.09569507289241,"n":227,"dd":83.46395363982752,"winrate":25.11013215859031},{"threshold":10,"slippage":0.001,"return":-93.70308643702542,"n":440,"dd":93.7138639272856,"winrate":32.95454545454545},{"threshold":10,"slippage":0.003,"return":-98.91665662592897,"n":440,"dd":98.91665662592898,"winrate":19.772727272727273},{"threshold":15,"slippage":0.001,"return":-99.09576244721329,"n":662,"dd":99.12095668101604,"winrate":37.61329305135951},{"threshold":15,"slippage":0.003,"return":-99.93598735571672,"n":662,"dd":99.93759060154427,"winrate":22.80966767371601}],"bootstrap":{"symbol":"LINK","n":102,"block_length":5,"simulations":20000,"expectancy_pct":-0.579162652611152,"mean_ci95_pct":[-0.8626877277006246,-0.30364442344140713],"prob_mean_positive":0}} Bootstrapseed875,circular5tradesequences,20000drawsdescriptiveuncertainty,zeroobservedpositivemeansnotzerotrueprobability. Premises1netexpectancynegative;2dailyN0–2insufficient;3actual15minstop/target/profitmanagementuntested. Conclusionreject,sizezero. MarkedLINK/UNIreturnsincludeoneopenpapertradeeach,winrateexcludesopen. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Code test_rule.py,results_1Hour.json,audit.py.

Frozen875 rule priorclose>SMA100 andEWMA RSI2<10,exitRSI2>70or10bars,nextadjacentopen,longonly.100observationsperfrequency.Vectorbt25bpfee/side+10bpbaseline/30bpstressslippage. Development2023–25afterlast>3hourgap; thresholds5/10/15, fixed2026Jan1–Sep7chronologicalholdout(previouslyseenmarket,notpristine). {"quality":{"symbol":"LTC","timeframe":"1Hour","original":32296,"used":31426,"duplicates":0,"removed":870,"missing":4,"start":"2023-02-06 10:00:00+00:00","end":"2026-09-07 23:00:00+00:00"},"holdout":{"symbol":"LTC","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.001,"bars":5999,"return_pct":-51.77772367368815,"drawdown_pct":52.16036540435516,"n":92,"win_rate_pct":14.130434782608695,"avg_win_pct":0.498960968697949,"avg_loss_pct":0.9976299616818438,"expectancy_pct":-0.7861551563020904,"quarterKelly":0,"open_trades":0},"stress":{"symbol":"LTC","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.003,"bars":5999,"return_pct":-66.62458723741572,"drawdown_pct":66.66491037225309,"n":92,"win_rate_pct":5.434782608695652,"avg_win_pct":0.6255351540793286,"avg_loss_pct":1.2871689097919694,"expectancy_pct":-1.183217601972877,"quarterKelly":0,"open_trades":0},"development":[{"threshold":5,"slippage":0.001,"return":-78.91986391462127,"n":216,"dd":79.06416176394086,"winrate":34.72222222222222},{"threshold":5,"slippage":0.003,"return":-91.11532897269242,"n":216,"dd":91.140780112567,"winrate":18.98148148148148},{"threshold":10,"slippage":0.001,"return":-95.78970155727151,"n":456,"dd":95.80355920520496,"winrate":30.043859649122805},{"threshold":10,"slippage":0.003,"return":-99.32055186049298,"n":456,"dd":99.3214296762381,"winrate":16.447368421052634},{"threshold":15,"slippage":0.001,"return":-98.55487450111858,"n":662,"dd":98.55963094312752,"winrate":29.909365558912388},{"threshold":15,"slippage":0.003,"return":-99.89769690031174,"n":662,"dd":99.89782907121113,"winrate":16.1631419939577}],"bootstrap":{"symbol":"LTC","n":92,"block_length":5,"simulations":20000,"expectancy_pct":-0.7861551563020904,"mean_ci95_pct":[-1.000935617369486,-0.5756970296138778],"prob_mean_positive":0}} Bootstrapseed875,circular5tradesequences,20000drawsdescriptiveuncertainty,zeroobservedpositivemeansnotzerotrueprobability. Premises1netexpectancynegative;2dailyN0–2insufficient;3actual15minstop/target/profitmanagementuntested. Conclusionreject,sizezero. MarkedLINK/UNIreturnsincludeoneopenpapertradeeach,winrateexcludesopen. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Code test_rule.py,results_1Hour.json,audit.py.

Frozen875 rule priorclose>SMA100 andEWMA RSI2<10,exitRSI2>70or10bars,nextadjacentopen,longonly.100observationsperfrequency.Vectorbt25bpfee/side+10bpbaseline/30bpstressslippage. Development2023–25afterlast>3hourgap; thresholds5/10/15, fixed2026Jan1–Sep7chronologicalholdout(previouslyseenmarket,notpristine). {"quality":{"symbol":"SOL","timeframe":"1Hour","original":22219,"used":17809,"duplicates":0,"removed":4410,"missing":7,"start":"2024-08-26 16:00:00+00:00","end":"2026-09-07 23:00:00+00:00"},"holdout":{"symbol":"SOL","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.001,"bars":5999,"return_pct":-51.6619482403338,"drawdown_pct":52.81967986500726,"n":118,"win_rate_pct":25.423728813559322,"avg_win_pct":0.625703765141929,"avg_loss_pct":1.028736064486575,"expectancy_pct":-0.6081157688183112,"quarterKelly":0,"open_trades":0},"stress":{"symbol":"SOL","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.003,"bars":5999,"return_pct":-69.84903207649523,"drawdown_pct":70.08349205667142,"n":118,"win_rate_pct":11.864406779661017,"avg_win_pct":0.7356152059909489,"avg_loss_pct":1.2403222048256142,"expectancy_pct":-1.0058889526948354,"quarterKelly":0,"open_trades":0},"development":[{"threshold":5,"slippage":0.001,"return":-52.58799108816441,"n":105,"dd":52.72361872456215,"winrate":36.19047619047619},{"threshold":5,"slippage":0.003,"return":-68.84814702017594,"n":105,"dd":68.85529513259733,"winrate":20},{"threshold":10,"slippage":0.001,"return":-78.61951189121405,"n":211,"dd":78.738035985355,"winrate":38.388625592417064},{"threshold":10,"slippage":0.003,"return":-90.80669889435634,"n":211,"dd":90.82101979784272,"winrate":23.696682464454977},{"threshold":15,"slippage":0.001,"return":-85.6542405036107,"n":295,"dd":85.73376713277007,"winrate":38.30508474576271},{"threshold":15,"slippage":0.003,"return":-95.59187565806693,"n":295,"dd":95.5987424323117,"winrate":24.74576271186441}],"bootstrap":{"symbol":"SOL","n":118,"block_length":5,"simulations":20000,"expectancy_pct":-0.6081157688183112,"mean_ci95_pct":[-0.808901744101676,-0.4039739269478655],"prob_mean_positive":0}} Bootstrapseed875,circular5tradesequences,20000drawsdescriptiveuncertainty,zeroobservedpositivemeansnotzerotrueprobability. Premises1netexpectancynegative;2dailyN0–2insufficient;3actual15minstop/target/profitmanagementuntested. Conclusionreject,sizezero. MarkedLINK/UNIreturnsincludeoneopenpapertradeeach,winrateexcludesopen. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Code test_rule.py,results_1Hour.json,audit.py.

Frozen note877. 2016-20 training (2013-15 unavailable),2021-25 new-rule holdout but historically inspected market period,2026 completed cohorts only. Monthly firstsessionopen following priormonthclose>100dayMA; previousbar ATR14 stop1ATR,target1.5ATR,timeexit10thsessionclose. At+1ATR sellhalf thenstopbreakeven; adverse samebar ordering. Gapstopopen,stopfirst if ambiguous,target capped. Alladjusted SIP daily;5bpfee+20bpslippage eachfill. Fractional continuous sizing diagnostic,100initial cash,oneposition, no leverage. Vectorbt exactfill reconstruction agrees independentcompounding within 3.55e-13 percentagepoints. Drawdown uses synthetic perdayfill/close slots,not actual15minpath; positions normalizedfullallocation for comparison,not proposal. IIDtrade bootstrap10000seed90802,95% mean interval [-0.768018,-0.193765]%. Historical p/W/L sizing only; failuresin training/holdout keepactual0 evenpositive2026counterexample. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; authorpaper https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf . Not exactleveragedBAB. Full252metricsobservationsQLV90802metrics;deterministicstudy.py/ledgers local.

Frozen note877. 2016-20 training (2013-15 unavailable),2021-25 new-rule holdout but historically inspected market period,2026 completed cohorts only. Monthly firstsessionopen following priormonthclose>100dayMA; previousbar ATR14 stop1ATR,target1.5ATR,timeexit10thsessionclose. Unmanaged parent of swing exit comparison; no partial profit. Gapstopopen,stopfirst if ambiguous,target capped. Alladjusted SIP daily;5bpfee+20bpslippage eachfill. Fractional continuous sizing diagnostic,100initial cash,oneposition, no leverage. Vectorbt exactfill reconstruction agrees independentcompounding within 3.80e-13 percentagepoints. Drawdown uses synthetic perdayfill/close slots,not actual15minpath; positions normalizedfullallocation for comparison,not proposal. IIDtrade bootstrap10000seed90802,95% mean interval [-0.879100,-0.205256]%. Historical p/W/L sizing only; failuresin training/holdout keepactual0 evenpositive2026counterexample. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; authorpaper https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf . Not exactleveragedBAB. Full252metricsobservationsQLV90802metrics;deterministicstudy.py/ledgers local.

Frozen875 rule priorclose>SMA100 andEWMA RSI2<10,exitRSI2>70or10bars,nextadjacentopen,longonly.100observationsperfrequency.Vectorbt25bpfee/side+10bpbaseline/30bpstressslippage. Development2023–25afterlast>3hourgap; thresholds5/10/15, fixed2026Jan1–Sep7chronologicalholdout(previouslyseenmarket,notpristine). {"quality":{"symbol":"ETH","timeframe":"1Hour","original":32298,"used":32298,"duplicates":0,"removed":0,"missing":6,"start":"2023-01-01 00:00:00+00:00","end":"2026-09-07 23:00:00+00:00"},"holdout":{"symbol":"ETH","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.001,"bars":5999,"return_pct":-50.90149591884842,"drawdown_pct":50.99493551485188,"n":100,"win_rate_pct":21,"avg_win_pct":0.5730093500685982,"avg_loss_pct":1.0424368198230116,"expectancy_pct":-0.7031931241457736,"quarterKelly":0,"open_trades":0},"stress":{"symbol":"ETH","timeframe":"1Hour","window":"holdout","threshold":10,"slippage":0.003,"bars":5999,"return_pct":-67.0883455312773,"drawdown_pct":67.0883455312773,"n":100,"win_rate_pct":13,"avg_win_pct":0.44727087085920136,"avg_loss_pct":1.3318758273298235,"expectancy_pct":-1.1005867565652505,"quarterKelly":0,"open_trades":0},"development":[{"threshold":5,"slippage":0.001,"return":-83.3003308153291,"n":239,"dd":83.34820012466358,"winrate":24.267782426778243},{"threshold":5,"slippage":0.003,"return":-93.58021410642344,"n":239,"dd":93.5815708956328,"winrate":10.87866108786611},{"threshold":10,"slippage":0.001,"return":-97.01568656178034,"n":488,"dd":97.0181381691363,"winrate":22.540983606557376},{"threshold":10,"slippage":0.003,"return":-99.5762614927846,"n":488,"dd":99.5762614927846,"winrate":9.836065573770492},{"threshold":15,"slippage":0.001,"return":-99.34181942114888,"n":709,"dd":99.34236011514159,"winrate":21.57968970380818},{"threshold":15,"slippage":0.003,"return":-99.96139170922915,"n":709,"dd":99.96139170922916,"winrate":10.155148095909732}],"bootstrap":{"symbol":"ETH","n":100,"block_length":5,"simulations":20000,"expectancy_pct":-0.7031931241457736,"mean_ci95_pct":[-0.974842474333507,-0.465997218495062],"prob_mean_positive":0}} Bootstrapseed875,circular5tradesequences,20000drawsdescriptiveuncertainty,zeroobservedpositivemeansnotzerotrueprobability. Premises1netexpectancynegative;2dailyN0–2insufficient;3actual15minstop/target/profitmanagementuntested. Conclusionreject,sizezero. MarkedLINK/UNIreturnsincludeoneopenpapertradeeach,winrateexcludesopen. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Code test_rule.py,results_1Hour.json,audit.py.

Frozen note877. 2016-20 training (2013-15 unavailable),2021-25 new-rule holdout but historically inspected market period,2026 completed cohorts only. Monthly firstsessionopen following priormonthclose>100dayMA; previousbar ATR14 stop1ATR,target1.5ATR,timeexit10thsessionclose. At+1ATR sellhalf thenstopbreakeven; adverse samebar ordering. Gapstopopen,stopfirst if ambiguous,target capped. Alladjusted SIP daily;5bpfee+5bpslippage eachfill. Fractional continuous sizing diagnostic,100initial cash,oneposition, no leverage. Vectorbt exactfill reconstruction agrees independentcompounding within 1.01e-13 percentagepoints. Drawdown uses synthetic perdayfill/close slots,not actual15minpath; positions normalizedfullallocation for comparison,not proposal. IIDtrade bootstrap10000seed90802,95% mean interval [-0.840583,1.131843]%. Historical p/W/L sizing only; failuresin training/holdout keepactual0 evenpositive2026counterexample. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; authorpaper https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf . Not exactleveragedBAB. Full252metricsobservationsQLV90802metrics;deterministicstudy.py/ledgers local.