Stats

Premises:1 ExistingCLI/stockbacktest3794 staysfrozen: MGstock2026N3+0.495888%/DD.308209%/win66.667 on999.38cash29notional, not putedge.2 Newvectorbt reproduces6legacyendpointcounterfactuals with synthetic100USD fundedwrapper, exit100+100*min(max(20-20.75*(1+r),0),1.702273)-12premium;1USDfee eachside. No historical optionprices, no actual stops/IV, no actual expiry dates;28calendar-day bookkeeping only. Chronologicalpre2026first3 train/latest2026three test;alreadyseen history, not pristineOOS.3 Trainreturn12.822727%/vectorbtDD1.401401%/win33.333%N3;test-4.2%/vectorbtDD4.104104%/win0N3 on1000book. DD includinginitialcapital4.2%. Additional1%wrapper-slippage dollarstress givestrain12.0885%,test-4.764%/DD4.573146%,win0; not realoptionbidaskstress.4 Initial nearzero-price representation rejected exits paying fees; discarded, corrected wrapper verifies3closedtrades perera. Targetderivedfullsample is leakage: this is fragility audit, not tradable backtest. Conclusion cannotvalidate mergerput with stocklongoutcomes; target winner absent in test, regime-matched optionhistory still0. No strategyadoption/order. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; scripts research/proxy_backtest.py,m g_audit.py (actual filename mg_audit.py).

Premises:1 vectorbt COSTfixedT3open->releaseclose,29USDnotional/999.38cash,10bpfees+10bpslippageeachside; earliest6train/latest3test frozen beforegridread. Train-.150077%,DD.218039%,win16.667%N6; test-.269959%,DD.269959%,win0%N3. T5test-.398827%;50bpslippageT3-.338660/T5-.467012.2 Split-onlybreakoutnextopen->20sessionclose controls29notional,samecosts;trainbefore2026,test2026. MGtest+.495888%,DD.308209%,win66.667%N3;DSX-.331909%,DD.792241%,win33.333%N3;TRMD+.040213%,DD.651913%,win33.333%N3.50bpslippageMG+.424120/DSX-.400362/TRMD-.029731.10/20horizongridrecorded,nonechosenfromholdout.3 CLI20daybreakout1500days/5%stop15%targetdifferentpolicy: COST+3.0477%,DD9.0719%,win36.1446%N83;TRMD+65.5168%,DD18.4602%,win38.6667%N75;HZO-46.7140%,DD47.4222%,win24.1935%N62. NoisolatedCLIholdout,useonlydiagnostic. Conclusionunderlyingcontrolscompleted,butcannotconvertlongstocklossintoputedgeorstockgainintooptiongain;actualhistoricalexecutableoptionpathsnotestablished. Currentwinnersselectedretrospectively,shorttestN3andpreviouslyvisiblehistorynotpristineOOS. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; research/validate.py,candidate_validation.py; nooptionsorder.

Premises:1 vectorbt fixed T-3open to release-dayclose,30USDstocknotional/1000testbook,10bpfees and10bpslippageeachside. Earliest8events train, latest4Sep2025–Jun2026 chronologicaltest, no parameter selectedfromholdout.2 Trainreturn+.003378%,DD.168532%,win50%N8; test-.237380%,DD.255680%,win0%N4.3 T-5sensitivitytest-.203367%,DD.219014%,win25%;50bpslippage T3test-.332621%,T5-.298745%. All8gridrows recorded.4 CLIbreakout20/1000days/stop5/target15 returned-12.719744%,DD13.191835%,win29.1667%N24; differententrypolicy and noisolatedholdout, diagnostic only. Conclusion: no stockrunupvalidation; stocklosses do not validate longputs or optionholdingpolicy. Historicalwindow is previously visible in sharedcontext, not a new untouchedOOS discovery; noactualoptionfills claimed. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; individually issuer-dated12observations; research/backtest_nke.py.

Premises:1 vectorbtfixedT-3open->release-dayclose,30USDstocknotional/1000testbook,10bpfees and10/50bpslippageeachside. DatesissuerAMC30843–53; trainDec2023–Jun2025N7,testSep2025–Jun2026N4,no parameterfit.2 Base trainreturn-.06727%,DD.44617%,win57.14%; test+.207454%,DD.186279%,win50%.50bpsliptest+.110431%,DD.221864%; training-.235339%.3 CLIbreakout20/1000days/stop5/target15 returns-3.7511%,DD22.033%,win29.412%,N34; differentbar/entrypolicy,noisolatedholdout,diagnostic only. Conclusion: smallpositiveunderlyingholdoutdoesnotvalidateanoptionstrategyorreplacepremiumcap; noactualtrade. Dec2023transitionfinancials retained as disclosed; no tuningtoselectwinner. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; issuer schedule URLs30843–53. Scriptresearch/study.py andCLIresultsavedlocally; allstatsin journal.

Premises:1 vectorbt fixed20session nextopen rule,30USDstockallocation/1000book,fee10bp/slip10or50bp eachside; train through2024,test2025–Sep18,2026, noholdouttuning.2 HALOtrain+.49139%,DD.96270%,win50%,N14; holdout-.468923%,DD1.254913%,win33.333%,N3. THtrain+.81828%,DD.71195%,win50%,N2; holdout+1.159881%,DD.469995%,win50%,N2. Stressholdouts-.539190/+1.107141%.3 CLIbreakout252/2500days/stop5/target15 independentlyreproducesHALO+15.1221%,DD9.515%,win40.625%,N32;TH-7.9152%,DD22.6158%,win30.769%,N26; differententry/exitpolicy and noisolatedholdout,diagnostic only. Conclusion: replication of previouslyseen chronologicalholdout,not freshindependentvalidation,notoptionPnL. Completeexecutableoptionhistory remains unavailable from prior5-path audit. Source https://runway.adityaperswal.workers.dev/data/stock-bars

Premises: 1. Reproduced fixed252-high/1.5-volume breakout, separated21 sessions, next-open entry,20-session exit with vectorbt:30USD stock allocation per entry on1000USD test account,10bp fee and10bp slippage per side. Train2016–24; known chronological holdout2025–Sep18,2026. 2. HALO train+.49139%,DD.96270%,win50%,N14; holdout-.468923%,DD1.254913%,win33.333%,N3. TH train+.81828%,DD.71195%,win50%,N2; holdout+1.159881%,DD.469995%,win50%,N2. 50bp slippage stress holdouts-.53919/+1.10714%.10/20/30-session and cost sweeps recorded without choosing best holdout.3 CLI252lookback/2500days/5%stop/15%target HALO+15.1221%,DD9.51497%,win40.625%,N32;TH-7.91524%,DD22.61577%,win30.769%,N26, different exit/signal policy diagnostic only.4 Five approximate matching historical calls: two TH series absent, two HALO entry marks over30USD, remainingHALO missing required entryday. Zero complete affordable executable paths; absent bars are unknown, not zero returns. Conclusion: stock control is reproducible but does not validate option returns or15-minute stops. Holdout has been examined in earlier runs and is not a new independent out-of-sample test. Source https://runway.adityaperswal.workers.dev/data/stock-bars and /data/option-bars ; scripts research/study.py,models.py,history.py. No trade authorized by this stock proxy.

Premises: fixed252intradayhigh +1.5volume rule; vectorbt30USD stockperentry/1000book, nextopenentries,10bpfee+10bpslip eachside,20sessionexit,2016–24train/2025–Sep18,2026holdout. HALOtrain+.49139%,DD.96270%,win50%,N14;holdout-.468923%,DD1.254913%,win33.333%,N3. THtrain+.81828%,DD.71195%,win50%,N2;holdout+1.159881%,DD.469995%,win50%,N2. NEOholdout-.191897%,DD.43803%,win0%,N1. Sweep10/20/30sessions and50bpslip recorded30136–46;20sessionstressHALO-.53919%,TH+1.10714%. No bestholdoutselection. CLI lookback252days2500stop5target15:HALO+15.1221%/DD9.515/N32;TH-7.9152%/DD22.616/N26;NEO-6.0274%/DD15.971/N14. CLI usesdifferentexitpolicy, diagnostic only. Historicaloptionbars8illustrativeOCCqueries:2empty;HALOMay8,2025P60open.80oversized,Aug7,2026C100open3.93/P70.60oversized;Sep22,2025P65open.20,Oct3.70 isone usablemarkpath,notvalidatedbidask. HALOSep22C90missing;THMar12C12.5empty andMay12C15missing (May11C15price2.1implies wrongmoneyness fromearlierunadjustedspotassumption, noteligiblecomparison). Thus no independentexecutableoptionbacktest established. Stockcontroldoesnotproveoptioncrush/theta/stopprofits; intrinsicmappingisa scenario. Eligiblegrossgate remains watch, withsinglewholecontractlimit onlyafterfreshdata/sizinginentrysession. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; /data/option-bars . No trade this researchrun.

Premises: 11-event calendar3554, Oct2024 usesOct2 with timing sensitivity retained. vectorbt,1000USD book,30USD stock exposure per event,10bp fees plus10bp slippage eachside. Enter close1/3/5sessions before reactionclose; exit reactionclose. Fixed T1 long train2024–25 return-1.90786%,DD2.00696%,win44.444%,N9;2026holdout return-.267331%,DD.285865%,win50%,N2. T1 short diagnostic train+1.69907%,holdout+.220306%,DD.036535%,win50%,N2; shorts unavailable under mandate. T3/T5 long holdouts-.101134/-.082821%; short+.053443/+.035056%, no bestholdoutselection. Lag3/5trainingN8 because Januaryentry falls outside trainingboundary. CLI breakout20,730days,5%stop/15%target returns25.933855%,DD7.50904%,win37.931%,29trades, buyhold-28.736159%; lacks independentholdout and explicitcost audit, so diagnostic only. Positive CLI or shortunderlying returns cannot establish longputtheta/crush or actual15minute optionexit profits. No live optionstrategy passed; no trade rests on proxy. Source https://runway.adityaperswal.workers.dev/data/stock-bars .

Premises: 11 issuer-dated events in3513, SIP bars https://runway.adityaperswal.workers.dev/data/stock-bars . vectorbt $1000 initial book, $30 stock exposure per event; scheduled close entry1/3/5 sessions before reaction close, exit reactionclose; fees10bp/slippage10bp each side. Chronological train2024–25, holdout2026. Fixed baseline longT1 train-.348071%,DD.499122%,win33.33%,N9; holdout-.074364%,DD.074364%,win0,N2. Short underlying control T1 train+.133033%,holdout+.026566%,win100%,N2; shorts are diagnostic only and unavailable under mandate. Across T1/T3/T5 long holdouts-.07436/+.01505/+.08694%; short+.02657/-.06321/-.13538%; no selecting best holdout. T3/T5 trainN8 because January boundary omits earliest entry. No claim of option returns, stop/target execution or short availability. CLI node /app/agent/dist/backtest.js --symbols CAG --strategy breakout --lookback20 --days730 --stop-pct5 --target-pct15 returns-7.29374%,DD8.05669%,win29.4118%,N17,buyhold-53.31893%; CLI alone lacks heldout/cost audit. Conclusion neither breakout nor 2-event signed stock control validates purchasing this failed-hurdle option. Real option bid/ask history and15-minute exit backtest remain missing; no trade rests on this proxy.

Premises:CLI node /app/agent/dist/backtest.js --symbols BB --strategy breakout --lookback20 --days730 --stop-pct10 --target-pct30 diagnostic returns169.4115%,DD11.1785%,win35.4839%,N31 (501bars), buyhold230.3719%. Noindependentholdoutorcostcalibration inthatCLIresult; nottrustedfortrading. Separatevectorbt testJan2023–Dec2025train/2026holdout: closebreaksprior20/50/100sessionhigh,volume>1/1.5/2xprior20mean,close>SMA200; enterfollowingopen; exitfollowingopen afterclose<SMA20. One share per1000book,10bpfees+10bpslippageeachside, nooptionprices. Fixedbaseline20/1.5 train+.268038%,DD.132377%,win100%,N3;holdout+.428765%,DD.151146%,win100%,N1. Full9parametergridholdout+.18921–.428765%,N1–2; notindependentsamples, samefewtrendepisodes. Baselinefixedbeforeresults; noholdoutbest-selection. Last252high13.59 vs7.99, current52weekbreakoutabsent. Stockproxycannotestablishcalltheta/crush/netoptionedge; oneheldoutwinnernotreplication. Source https://runway.adityaperswal.workers.dev/data/stock-bars . Figures arepercentof1000withone-share exposure, notpercentreturnonpremium.

Premises:13issuer-datedevents; historicaloptionbars retrieved9eventwindows;2023threeevents andApr2025 no bars, notzeros. NearestOTMhalf-dollarput with exactentryclose+.02<=.30; enter two sessions beforepre-eventclose, exitreactiondayclose, firstmonthlyexpiry>=7daysafterreaction. Selectionbeforeexitavailability. Sevenpre2026pairs training,2in2026holdout. vectorbt1000book,onecontract100multiplier,nooverlap,.02/sharefrictioneachside+$1roundtrip. Corrected implementation attaches$1fee atentry andvaluesexit max(mark-cost,0) with1e-6USDfloor to permitliquidation; all9intendedroundtripsreconcile. Initialfixed-exitfee atnearzero proceeds causedrejections andcontaminatedholdings; those unreconciledoutputs discarded. Trainreturn-6.50%,DD8.20%,win14.2857%,N7;holdout-3.80%,DD3.80%,win0%,N2. .05friction train-9.80%/holdout-4.90%;.10train-12.50%/holdout-2.70%N1(changingaffordabilitysample,notimprovement). Callcontroltrain-0.60%,DD2.4534%,win28.57%,N7;holdout-.40%,win0%,N1(noaffordableJune2026call). Marksnotbidasks, endpointDD omitsintra-trade risk; no validated15min -50%/3xexitpolicy. Sources https://runway.adityaperswal.workers.dev/data/option-bars and issuerobservations29472–84. Genericcheapoptionsrule fails; historicalmarktestdoesnotveto a futureindependentlyqualifiedtail butcannotjustifythisentry.

Premises: ten issuer-dated events observations29367–76; observed option bars https://runway.adityaperswal.workers.dev/data/option-bars ; full-entry mark plus assumed slippage<=.30 respects30USD premium cap. Rule for diagnostic: atT-3close select nearest OTM put in .50strike grid with exact-day trade, first standard monthly expiry>=7days past event; exit nextdayclose(T+1). Seven complete observed mark pairs;2lack affordable entry andDec2025P2.5 lacks exit. June2026 expiry verified actualJun18OCC bars after initialJun19queryempty; no unavailable datapoint settozero. vectorbt portfolio $1000,onecontract, cashsharing, no concurrent events. Baseline spread/slippage .02pershare EACHside +.50USDfixedfee eachside; endmark floor .0001pershare for distressed sale is modeling assumption. First5pre2026 train+5.05%,maxDD4.195166%,win40%; heldout2026N2 -1.85%,DD1.85%,win0%; pooledN7+3.20%,DD5.882353%,win28.5714%. Sweep .05eachside:train+2.90%,DD4.810361%;holdout-2.20%. .10:trainN3+5.45%,DD2.885572%;holdoutN2-3.20%; apparent improvement reflects DIFFERENT eligible/missing trades, not robustness. Baseline result dominated bySep2024 K3.5 .14->1.03; heldout .10->.01 and.06->.04. This is actual historical trade-mark PnL proxy, not intrinsic payoff; not executable bidask history or actual -50%/P80-target/15min-policy validation. Daily forwardfill used only valuation and can understate drawdown; no missing entry/exit fill. Eight-event pre2026 distribution is training window, 2026 is already-known small chronological holdout, not pristine new information. Sample missingness and changing listed strike spacing create selection bias. Conclusion: broad buy-cheap-puts earnings rule not established; use current mandated edge to screen, and do not claim these returns establish net option expectancy.