Stats

Diagnostic note2092 added after initialdailytraining inspected; not independentpreregistration/no winnerselection. Period2016-12-29–2022-12-30. Priorcalendar-month finalclose above10monthlycloseSMA holds followingmonth; only firsttradingdayopen signals, no catchup atsplitinitialsession; terminalclose liquidation. Fivefixedindependent equalcapital sleeves,1000each,no monthlycapitalrebalance,cash0%;SIP/alladjustment not independentlybuilt totalreturn. vectorbt from_orders longonly, fee1bp+slip15bp eachside; excludes taxes, cashyield, operatorcosts, no stops/targets. Authorpaperuses same-monthsignalclose andTbillcash; thislaggedproxy differs. Bothperiods profitable; WATCHmonthlyresearch, notlivepermission. Missing current>=3R stop/target and15min/whole-share implementation; modifying rule requires independenttest, existingobserved2023–2026 no longer untouched. Fiveassetmatch from200dailyreturns baseline train-3.20348/holdout12.65007%,stress-7.77259/+9.01443%; differentrule/exposuretiming, cannot attribute difference solelytostops orcausality. Sources https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars . Code research/monthly_control.py andmonthly_control.json retained.

Edge test: MLKN Oct16 P17.5 reference .28 Sep9, S20.80 Sep10: requires -17.211538%; eight issuer-dated earnings close moves, median absolute10.745876%, coverage1/8=12.5% (<40%), intraday lows also1/8. HITI frozen N9 medianabs5.494505%, C.31/P.16 atS2.56 require+9.765625/-8.593750%; closecoverage1/9each, putdaily-low4/9 is hindsight path only. Neither documents expected optionpayoff>=3x premium. Premises: testwhetherbearishpre-eventstockpathsupportsputrun-up, independentofearningsgap. Vectorbt Portfolio.from_signals short_entries/short_exits, 1000initialcash, fullstocknotional,10bpfees+10bpslippageeachside(assumptionsnotvenuetariff), entryT-2/T-3/T-5open beforeAMCday, exitT-1close. Six2024–25events train, two2026 holdout; noholdoutselection. Fullresults [{"lag":2,"phase":"train","return":-3.997024599531062,"drawdown":6.0269945892936665,"win_rate":33.33333333333333,"sample":6},{"lag":2,"phase":"holdout","return":-1.973521635520217,"drawdown":4.115226417384676,"win_rate":50,"sample":2},{"lag":3,"phase":"train","return":-8.454414289599754,"drawdown":9.786059101244893,"win_rate":33.33333333333333,"sample":6},{"lag":3,"phase":"holdout","return":-7.388058106205415,"drawdown":10.096569869161353,"win_rate":0,"sample":2},{"lag":5,"phase":"train","return":-8.350924632949239,"drawdown":9.444998722783293,"win_rate":50,"sample":6},{"lag":5,"phase":"holdout","return":-10.13648642032798,"drawdown":12.642233751916987,"win_rate":0,"sample":2}]. Training-bestT2 loses3.9970%; holdout-1.973522%,DD4.115226%,win50%,N2. Allvariantslosebothwindows. A stockshortproxy is researchonly: venueandmandatepermitlongoptions, notactualshortposition. It doesnotmodel optionIV/theta/spreads, borrowcost, or15minstops; noindependentpositiveoptionedge. CurrentBMOtimingdiffersfromhistoricAMC, so no claimofidenticalexecution. Counterexample: optionvolatilityexpansioncouldoutperformstock; historicaloptionpricesneeded. SerializationinitiallyfailedonNumPyint64; fixedexplicitint and reranNUMBA_DISABLE_JIT=1, successfulJSON. Noeconomicschanged. Source upcoming https://news.millerknoll.com/2026-08-25-MillerKnoll-Schedules-First-Quarter-Fiscal-Year-2027-Conference-Call-and-Webcast ; eightprimaryschedules23066–73; stock-bars SIPadjustmentall675bars2024–Sep10. No backtestCLItechnicalruletrusted, vectorbtcustomdatedtestused.

Diagnostic note2092 added after initialdailytraining inspected; not independentpreregistration/no winnerselection. Period2016-12-29–2022-12-30. Priorcalendar-month finalclose above10monthlycloseSMA holds followingmonth; only firsttradingdayopen signals, no catchup atsplitinitialsession; terminalclose liquidation. Fivefixedindependent equalcapital sleeves,1000each,no monthlycapitalrebalance,cash0%;SIP/alladjustment not independentlybuilt totalreturn. vectorbt from_orders longonly, fee1bp+slip5bp eachside; excludes taxes, cashyield, operatorcosts, no stops/targets. Authorpaperuses same-monthsignalclose andTbillcash; thislaggedproxy differs. Bothperiods profitable; WATCHmonthlyresearch, notlivepermission. Missing current>=3R stop/target and15min/whole-share implementation; modifying rule requires independenttest, existingobserved2023–2026 no longer untouched. Fiveassetmatch from200dailyreturns baseline train-3.20348/holdout12.65007%,stress-7.77259/+9.01443%; differentrule/exposuretiming, cannot attribute difference solelytostops orcausality. Sources https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars . Code research/monthly_control.py andmonthly_control.json retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2023-01-03 to 2026-09-10; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(0); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage15bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":26.38377201205344,"IWM":5.354821332148041,"EFA":28.51371131393328,"EEM":17.639756103504478,"TLT":-27.905545909065754,"IEF":-13.115800110816622,"GLD":7.4620880456991445,"DBC":2.371501098360529,"VNQ":0.28147353536602804}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2023-01-03 to 2026-09-10; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(0); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage5bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":29.45367737624258,"IWM":7.054057957035864,"EFA":30.84790536983071,"EEM":21.222423989776452,"TLT":-22.986772474113316,"IEF":-8.295059481656253,"GLD":8.759402828612984,"DBC":6.336479925047089,"VNQ":6.695718881876278}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2016-12-29 to 2022-12-30; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(2); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage15bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":3.4078171497209597,"IWM":-5.579949231820303,"EFA":-22.07376939450205,"EEM":-17.321886586017243,"TLT":-22.131031210091756,"IEF":-6.905174309487084,"GLD":-9.300987020688344,"DBC":1.167379643909805,"VNQ":3.167766729063204}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2016-12-29 to 2022-12-30; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(2); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage5bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":6.557057491275531,"IWM":-2.1188985278281947,"EFA":-17.089419400321816,"EEM":-13.602826228651354,"TLT":-18.95313479617693,"IEF":-3.4927009060862213,"GLD":-5.031379677702052,"DBC":4.875759672186408,"VNQ":9.109997748285537}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2023-01-03 to 2026-09-10; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(2); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage15bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":38.497527925049326,"IWM":35.866113507717756,"EFA":47.203643961592626,"EEM":51.00243420798159,"TLT":-20.113842950253492,"IEF":-5.57555833209975,"GLD":-4.154784612903684,"DBC":7.723165262803899,"VNQ":4.05070421375362}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2023-01-03 to 2026-09-10; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(2); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage5bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":41.01305800455537,"IWM":39.44495973061672,"EFA":51.68666726452622,"EEM":58.11079273058517,"TLT":-14.663466606171571,"IEF":-0.5355780434952995,"GLD":-2.608925598131484,"DBC":11.895420869374473,"VNQ":12.042515396131659}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2016-12-29 to 2022-12-30; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(3); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage15bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":-14.591613736966558,"IWM":-18.039455740205433,"EFA":0.9680625445164681,"EEM":2.144943042648273,"TLT":-5.227152560452353,"IEF":-5.568513117853302,"GLD":-5.939073491687397,"DBC":-19.153644482547826,"VNQ":-19.11324054522835}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2016-12-29 to 2022-12-30; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(3); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage5bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Sensitivity only; positive heldout cell cannot replace preregistered200. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":-10.749739141364945,"IWM":-11.3905909962984,"EFA":4.878684405730804,"EEM":5.8891565243266,"TLT":-0.5671043743658788,"IEF":-0.9252498726996072,"GLD":1.4882277489038698,"DBC":-14.66791161903952,"VNQ":-13.767234606003974}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2023-01-03 to 2026-09-10; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(0); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage15bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Primary trainingfails and holdoutCIcross0, noentry. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":31.272070484275673,"IWM":31.205680328045627,"EFA":26.51437548689872,"EEM":14.862385099747444,"TLT":-38.72252456967193,"IEF":-11.934389193601714,"GLD":1.6505574515651489,"DBC":-6.386220804456178,"VNQ":5.606317983344633}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2023-01-03 to 2026-09-10; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(0); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage5bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Primary trainingfails and holdoutCIcross0, noentry. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":33.389318723808124,"IWM":33.05548024265854,"EFA":27.785867483440185,"EEM":18.83485989336602,"TLT":-33.08571830303656,"IEF":-7.04809837677055,"GLD":2.877713681930416,"DBC":-2.565765778254603,"VNQ":11.689004062946529}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2016-12-29 to 2022-12-30; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(2); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage15bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Primary trainingfails and holdoutCIcross0, noentry. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":0.0505793718993397,"IWM":-13.34127185943893,"EFA":-14.271283766176168,"EEM":-6.144195636142136,"TLT":-4.900973439429867,"IEF":-6.50069987376588,"GLD":0.705828948464915,"DBC":-14.739594243152975,"VNQ":-3.401966080917651}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregistered note2085 primary200day,150/250 diagnostic sensitivities never selected. Period 2016-12-29 to 2022-12-30; SIP adjustmentall daily histories, first250sessionswarmup. Fixed9 independent equalcapital sleeves initially1000each, allcash at split, no crosssleevecapitaltransfer; cash0%, fractional proxy. Buy nextopen only on priorclose upwardSMAcross; stop2*prior20TRsimplemeanATR; target6ATR=3R; sell nextopen after belowSMA or126sessions. Openinggaps fillopen; intraday touches filllevel at dailycloseeventslot, stopfirst ifboth, terminalsampleclose forced(2); no15minactualfill validation. vectorbt from_orders directionlongonly, fee1bp andslippage5bp/side. Independent customtradecount and capitalreturn vs vectorbtfee-denominator reconciled exactly. Portfolio return/dailycloseDD, trade winrate. Primary trainingfails and holdoutCIcross0, noentry. Entire matrix6trainingnegative/6holdoutpositive, regime dependence. Gapfills, taxes, dividends reconstruction, whole-share187.22allocation and venue15minexits remain limitations. This adapter is not Faber's monthly model. Perassetreturns {"SPY":3.510793125049716,"IWM":-7.428824705779736,"EFA":-8.787867938913008,"EEM":-2.118295376597046,"TLT":-1.6120073816427882,"IEF":-3.073400101186974,"GLD":8.007833637821825,"DBC":-9.828786886720199,"VNQ":2.1618631241606234}. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf . Local research/test_trend.py SHA dff619d846bd555021e1f319301ecac71170ed7ef8700e3e61c036e18dff9b5d; rawJSON/results/tradeCSVs retained.

Preregister2088 beforebars. SIPadjustedall2021-Sep10 2026,1428bars https://runway.adityaperswal.workers.dev/data/stock-bars . Close>prior20closehigh AND SMA50; nextopen;stop2prior20ADR,target3R,20sessions exit. Daily stop-first, openinggaps atopen; training2021-24 and flatstart2025-Sep10holdout; entrydate highlow cantriggerexit. vectorbt longonly portfolios,fees5bp/slip20bp eachside;50bpsstress; all10/20/40cells in observation. Assert roundtripcounts and compoundedmanual fee-inclusivereturn equalsportfolio. Unpublished default-short diagnostic discarded; this replaces it. Frozen20 train+48.9822%,N28;holdout+47.4497%,DD23.3388%,N15;stress+34.7587%. MeanCI crosseszero; chronological holdout is untouched by parameterselection but symbolselection is current-survivor biased. No15min/+2Rtrailing policy or cashdividend/rawstop replication; daily price-adjustment proxy is not liveexecution. Event-marked DD omits intraday equity lows. Current54.41<55.20priorhigh; entryfalse. Conclusion retain as watch, not reject positive economics merely because CI includeszero; no presentorder. Alternative10/40holdouts positive; none selected to replace20. Finalstatistics in figures; oldCLI2342differentIEXsignalexit not confirmation.

Required CLI run: node /app/agent/dist/backtest.js --symbols LPG --strategy breakout --lookback20 --days1500 --stop-pct10 --target-pct30.82trades1030bars;return94.52997%,DD11.30342%,win39.02439%,buyhold235.20345%. Source https://runway.adityaperswal.workers.dev/data/stock-bars . Inspected /app/agent/dist/backtest/bars.js: CLI explicitlyIEX and noadjustment; not comparable to SIPadjusted bespoke3R study, not independently heldout, no assurance cost model matchesvenue. Treat diagnosticonly, never sizing from this number.

Executed node /app/agent/dist/backtest.js --symbols BTC/USD,ETH/USD,SOL/USD --strategy breakout --lookback20 --days730 --stop-pct8 --target-pct24. 731 supplied daily bars. Premises:1 gross returns BTC25.66649/ETH26.22267/SOL55.50067%;2 engine uses same-close entry, no fees/slippage, no holdout and close-series breakout, which differs from frozen prior-high ATR rule;3 therefore positive benchmark cannot authorize trade. Valid inference is only a harness diagnostic; soundness limited by timestamp aggregation and last partial bar. Counterexample: after-cost next-open version can lose despite positive gross return. Full trade outputs local cli.json. Source https://runway.adityaperswal.workers.dev/data/crypto-bars .

HITI Oct16 K2.5 call/put close .31/.18 (Sep9); underlying2.555 throughSep10 19:20UTC. N9 issuer-dated AMC earnings: directional break-even +9.98043%/-9.19765%, coverage1/9=11.111% each (<40%); medianabs close5.494505%, medianabs gap2.34375%. Thresholds exceed close median by4.485925/3.703146pp, so no hold-through exception. Recent put .16 still1/9 at8.414873%. Intrinsic-only mean payoff .166696/.471908x premium, NOT option expected value; no documented >=3x mispricing. Premises: test a distinct pre-event run-up rather than infer option profit from earnings gaps. Vectorbt from_signals, adjusted daily SIP HITI Jan2022–Sep9 2026; nine dates from observations. Entry open T-2,T-3,T-5 sessions relative to AMC release; exit close T-1. $1000 initial,100% stock exposure when active,10bp fee+10bp slippage EACH side (assumptions, not venue tariff). Six pre2026 train,three2026 holdout. T2 train-0.9083%,DD7.8305%,win33.33%; holdout+2.1345%,DD1.3973%,win66.67%. T3 train+.8830%,DD7.0722%,win50%; holdout-4.7624%,DD5.3978%,win33.33%. Training-best T5 +4.2836%,DD10.8128%,win50%; holdout-2.6317%,DD5.3640%,win66.67%. No holdout-based selection; all cells retained. Conclusion: no stable run-up alternative, and this stock-price proxy cannot establish actual option spread/theta/IV P&L. Counterexample: option volatility expansion could outperform stock; requires historical option prices, not inferred here. Daily signal marks do not replicate venue15min option exits. All historical primary URLs stored with event observations. Upcoming date https://hightideinc.com/high-tide-to-announce-third-fiscal-quarter-2026-financial-results/. Scripts analyze.py/backtest_hiti.py and raw JSON saved local; no claimed trading backtest of options.

Frozen note1880/1882: buy observed09:30gap<=-1.0prior20ADR at09:45;1ADRstop/1.5target;half+1RthenBEatnext15mincheck;15:45exit. Fullsessionsonly; costsfee1bp+slippage15bp eachside, assumednotmeasuredspreads. Exactordercashflowsverifiedagainstvectorbt; portfolioeightdailyrebalanced1/8cashsleeves, markevery15minutes, inactivecash. Windowrecent:train2017–19/holdout2020–25/recent2026throughSep9;recentnotvirginprospective. Allneighborsretained,noselection. Return-2.188312%,DD2.579277%,N90,win35.555556%,avgwin0.858982%,avgloss0.776409%. Researchportfolioisnotwhole-share/PDTfeasibleand09:45entrynot09:50managerfill. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; https://personal.lse.ac.uk/polk/research/TugOfWar.pdf . Adaptation,notoriginalstockmonthlyportfolio.

Frozen note1880/1882: buy observed09:30gap<=-1.0prior20ADR at09:45;1ADRstop/1.5target;half+1RthenBEatnext15mincheck;15:45exit. Fullsessionsonly; costsfee1bp+slippage5bp eachside, assumednotmeasuredspreads. Exactordercashflowsverifiedagainstvectorbt; portfolioeightdailyrebalanced1/8cashsleeves, markevery15minutes, inactivecash. Windowrecent:train2017–19/holdout2020–25/recent2026throughSep9;recentnotvirginprospective. Allneighborsretained,noselection. Return0.034781%,DD2.190892%,N90,win45.555556%,avgwin0.848952%,avgloss0.701396%. Researchportfolioisnotwhole-share/PDTfeasibleand09:45entrynot09:50managerfill. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; https://personal.lse.ac.uk/polk/research/TugOfWar.pdf . Adaptation,notoriginalstockmonthlyportfolio.

Frozen note1880/1882: buy observed09:30gap<=-0.5prior20ADR at09:45;1ADRstop/1.5target;half+1RthenBEatnext15mincheck;15:45exit. Fullsessionsonly; costsfee1bp+slippage15bp eachside, assumednotmeasuredspreads. Exactordercashflowsverifiedagainstvectorbt; portfolioeightdailyrebalanced1/8cashsleeves, markevery15minutes, inactivecash. Windowrecent:train2017–19/holdout2020–25/recent2026throughSep9;recentnotvirginprospective. Allneighborsretained,noselection. Return-7.922128%,DD7.942104%,N259,win35.135135%,avgwin0.703699%,avgloss0.770646%. Researchportfolioisnotwhole-share/PDTfeasibleand09:45entrynot09:50managerfill. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; https://personal.lse.ac.uk/polk/research/TugOfWar.pdf . Adaptation,notoriginalstockmonthlyportfolio.

Frozen note1880/1882: buy observed09:30gap<=-0.5prior20ADR at09:45;1ADRstop/1.5target;half+1RthenBEatnext15mincheck;15:45exit. Fullsessionsonly; costsfee1bp+slippage5bp eachside, assumednotmeasuredspreads. Exactordercashflowsverifiedagainstvectorbt; portfolioeightdailyrebalanced1/8cashsleeves, markevery15minutes, inactivecash. Windowrecent:train2017–19/holdout2020–25/recent2026throughSep9;recentnotvirginprospective. Allneighborsretained,noselection. Return-1.770402%,DD4.245520%,N259,win43.629344%,avgwin0.749110%,avgloss0.673702%. Researchportfolioisnotwhole-share/PDTfeasibleand09:45entrynot09:50managerfill. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; https://personal.lse.ac.uk/polk/research/TugOfWar.pdf . Adaptation,notoriginalstockmonthlyportfolio.

Frozen note1880/1882: buy observed09:30gap<=-0.25prior20ADR at09:45;1ADRstop/1.5target;half+1RthenBEatnext15mincheck;15:45exit. Fullsessionsonly; costsfee1bp+slippage15bp eachside, assumednotmeasuredspreads. Exactordercashflowsverifiedagainstvectorbt; portfolioeightdailyrebalanced1/8cashsleeves, markevery15minutes, inactivecash. Windowrecent:train2017–19/holdout2020–25/recent2026throughSep9;recentnotvirginprospective. Allneighborsretained,noselection. Return-15.255118%,DD15.255706%,N424,win30.896226%,avgwin0.626137%,avgloss0.728750%. Researchportfolioisnotwhole-share/PDTfeasibleand09:45entrynot09:50managerfill. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; https://personal.lse.ac.uk/polk/research/TugOfWar.pdf . Adaptation,notoriginalstockmonthlyportfolio.

Frozen note1880/1882: buy observed09:30gap<=-0.25prior20ADR at09:45;1ADRstop/1.5target;half+1RthenBEatnext15mincheck;15:45exit. Fullsessionsonly; costsfee1bp+slippage5bp eachside, assumednotmeasuredspreads. Exactordercashflowsverifiedagainstvectorbt; portfolioeightdailyrebalanced1/8cashsleeves, markevery15minutes, inactivecash. Windowrecent:train2017–19/holdout2020–25/recent2026throughSep9;recentnotvirginprospective. Allneighborsretained,noselection. Return-5.789755%,DD6.552816%,N424,win41.981132%,avgwin0.635228%,avgloss0.650198%. Researchportfolioisnotwhole-share/PDTfeasibleand09:45entrynot09:50managerfill. Sources https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; https://personal.lse.ac.uk/polk/research/TugOfWar.pdf . Adaptation,notoriginalstockmonthlyportfolio.