Stats

Numberedpremises1 Read ownlessons83/82/81 thennotes/observations/analyses andtradehistory; zeroopen/pendingholdings meansnoexits toreset. Trade10singleoperatorliquidation isnotnaturalexit calibration. 2 Expanded36paircompletedcensusfindsnewUNI/LTC/LINKbreakouts; majorBTC/ETH/SOLfail20closelevel. PriorUNI81/LINK80rejectnotreversedbynewsignal; XRP6tradesandSOLwatchstillinsufficient. 3 Preregister214 fixedLTCdaily20+SMA50/rangeexit survives neitherdaily norhourlycostholdout: daily-7.7225%N8, hourly-19.4063%N11,p27.2727%,W3.4177%,L3.8692%,E-1.88186%; doublecostworse. Training15/20/30allnegative. AllKellysizes0. 4 ExistingfundingBTC/ETH/SOL negativepayoffs unchanged; currentOKXfields notdailysettledBinancesignal. IBITshareincreaseSep3/4 isdatedissuer evidence, notweekendcausalflow. 5 Dailyhourlyopens/ranges/volume discrepanciesconfirmed; sameLTCsignalbutnoexactexecutiontruth. Spot59bpLTCspread and15minstop/partialprofitimplementation remain gaps; noapproveddelayedentryafter00UTC. Conclusion zeroPROPOSE/EXIT/ADJUST/FUND,hold235.86cash. Validunderpositiveexecutableedgerequirement; soundnessboundedbydataquality,currentuniverseselection,smallN andsimulatedcosts,notuniversalclaimcryptocannotprofit. Cashbearcase missedrotation consideredin202; opportunitycostcannotfixnegativeexpectancy. Startcash235.86,goalsoundrepeatableedge,bindingconstraintvalidation. Hardcaps212.7475position,235.86fund,8.509plannedloss applyonlyafterpositiveedge; no inventedstop/target. Ratevsperiod: hourlyhistoricalmean-1.88186%pertrade, total-19.4063%over11Jan-Sep2026trades; notannualized orforecast20tradesbymonthend. Who benefits/threeorderchains202; simulations217illustrateerosionnotcalibratedfutureodds. Lessons98-100recorded. Researchscopecompleteearly: audit8mincapcutcause/majorfullhistoryexpansion; LTCtestfinishedwithin17minallowance,final5minutephasefrom~00:14. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://docs.alpaca.markets/us/docs/crypto-fees ; https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf ; https://www.okx.com/api/v5/public/time .

Premises1 Calendarcensus36publishedUSDpairs,35have21uniquedates;USDGfourrecentdays remainsincomplete. Observedv*vwAug30-Sep5 rankingrawtop10 USDC,USDG,BTC,ETH,SOL,XRP,UNI,LTC,PAXG,LINK;USDG noteligibleandvolumehistoricalaggregationdefects meannotcertifiedcurrentAssets/globalranking. RemovingincompleteUSDG bringsUSDTinto10, stillsame3nonstablebreakouts. EightwholecensusrawbreakoutsUNI,LTC,LINK,ARB,GRT,CRV,SUSHI,BAT; outsideobservedtop10notpromoted. 2 BTCclose79841.951<81269.4915,ETH2479.918<2515.933,SOL103.165<109.15. UNI7.04908>6.32576(+13.9735%day),LTC54.72185>53.14(+7.8051%),LINK12.0527>11.9919(+3.5909%). 3 LTChourlyaggregatedprior14meanrange2.30976697USD,current50daymean47.245769,dayhigh55.33815/low50.6473;20breakoutandSMAfiltertrue. Thesearelevelsandrangeestimates,notorders. At00:08:17UTC quote54.4977/54.82 spread58.96575bp; quoteisnotdailyclose, and delayedentryafter00UTCnotregistered. UNI/LINKspread19.40368/18.07014bp. 4 Strongestbullcase rotation cancontinue;counterexamplefailedcostadjustedpayoffs: LTC212hourly-19.41%/daily-7.72%,UNI81natural-exitexpectancynegative,LINK80fragile+.295%returnturns-7.786%doublecost. Noqualifiedentry,stops/targetsnotinventedbecausecash. Validity undernetedgegate sound; no inference allfuturebreakoutsfail. Quotesnotguaranteedfills; historicalquoteOHLC cannot certify actualexitprices. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://docs.alpaca.markets/us/docs/historical-crypto-data-1 .

Premises1 Hourlyholdout212 has11naturalnettrade returns, mean-1.88186%,p27.27%,W3.41768%,L3.86918%. QuarterKelly max(0,(p-(1-p)/(W/L))/4)=0. 2 Numpyseed20260906,20000scenario paths of20trades, circularblocks2 sampledfrom11returns preserve neighboringpair order. Hypothetical10%/25%equityallocations compound eachsamplednetreturn from850.99equity. No annualization or20tradesbydeadlineclaim. At10% endingp5/median/p95=$798.22/$819.35/$841.94;98.845% paths lose. At25% $724.59/$773.64/$828.14;98.905% lose,48.58%maxdrawdown>10%. These conditionalempiricalscenario frequencies are NOT calibratedfuture probabilities: smallselectedsample,no unseenfat-tails, changingregimes, gaps, dynamicspreads orcashfundcapmodel.25%scenario maybreakplanned1%lossconstraint and isnotproposal. Mean20trade-scenario2.5/97.5percentiles[-3.43274,-.26580]% are a distribution ofscenario means, NOT a confidenceinterval ofthe11tradepopulationmean. Separateactual11tradebootstrap in212 spanszero. 3 Negativeplug-inKelly=>authorizednotional0 beforeposition212.75,fund235.86,plannedloss8.509dollar caps. Conclusion holdcash; reducingstake canreduce modeleddrawdown butdoesnotcreatepositiveedge. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; recordedtest212; local ltc_trades.json/simulation.json.

Preregisternote214 beforeLTCresults. Fixed completedclose>prior20closemax AND>SMA50; enter nextUTCopenwhenflat; stopprior14daymean(high-low),target1.5range,timeout7days; after observed+1R moveprice stoptoentry. Check at each observedbaropen, fill actualopen withslippage, noidealbarrierfill. Oneentryperday, no same-dayreentryafterexit. vectorbt explicitorders/cash10k,25bpfee+5bpslip eachside. Chronologicaltraining2024-2025;holdoutJan1-Sep5 2026 startsflat,warmuptrainingprices, no carriedposition. Currentuniverse/knownhistory selectionnotindependentprospective. Canonicalhourly uses completedhourlyaggregateddaily signals; vendor-daily-signalhourly identicalholdoutpayoffs despitehistoricalfeeddefects. Hourlyholdoutreturn-19.40627%,DD23.92950%,N11,p27.2727%,W3.41768%,L3.86918%,E-1.88186%,tradebootstrap95mean[-3.85956,+.22622]%,quarterKelly0. Doublecostreturn-24.55378%,DD27.49468%,N11,mean-2.46881%,CI[-4.43469,-.37334]%. Dailyvariantreturn-7.72250%,DD20.19482%,N8,p25%,W9.35989%,L4.19682%,E-.80765%,Kelly0; doubled-12.04725%. Differentexitfrequency changestrades/payoffs, no independentreplications. Trainingonlylookback15/20/30 returns-34.3387/-19.2603/-26.5399%; fixed20retained,allnegative,notoptimized. No terminalpositions or heldtradescrossing4knownmissinghours; independentledgermaxerror1.82e-12. Missingbars canstillaffectrange/history. Quote spread58.9657bp exceeds base10bp roundtripslip; doubledtotalcost120bp roughly exceeds currentspread+50bpfees109bp, butnotvalidatedhistoricalspreadmodel. Pricebreakeven doesnotcoverfees; no mandatorypartialprofit or15minuteimplementation. Negativeinallregisteredvariants sufficesREJECT,size0; don'tretuneinspectedholdout. Serializationnp.int failurefixedwithoutchangingrules; partialartifactdiscarded. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://docs.alpaca.markets/us/docs/crypto-fees .

Requiredcontrol executed node /app/agent/dist/backtest.js --symbols LTC/USD --strategy breakout --lookback20 --days979 --stop-pct5 --target-pct7.5 (actual flags spaced). Vendor980dailybars,currentpartialSep6included; no chronologicalholdout, fees orslippage; samebarcloseentry and fixedbarrierexits, signalexit unlike registered7dayrange rule. Thus a diagnostic control, NOT registeredrule replication or parameterselection. Return-24.538177%,DD31.073446%,win26.415094%,N53,avgtrade-.434447%,BH-26.736173%. Source engine/code /app/agent/dist/backtest/engine.js ; https://runway.adityaperswal.workers.dev/data/crypto-bars . vectorbt prereg214 separately usescausalnextopen,rangeexits,fees,holdout and hourlycomparison. ThisgrosscontrolcannotjustifyLTCpurchase, and disagreement withcustomtestcannotisolateonebias.

No holdings or pendingorders exist; no live stop/target to reset. Cash is the thesis being attacked. Premise1 NewSep5 completed bars show UNI+13.97%,LTC+7.81%,LINK+3.59% and prior20close breakouts; BTC/ETH/SOL below thresholds. Premise2 UNI/LINK prior costadjustedtests81/80 rejected; LTCnewtest214pending; funding127-129negative. Premise3 issuerIBITsharecounts support netissuance across Sep3/4, not timestampedSaturdayflow. Chains (conditional mechanisms, not forecasts): A Cash bear case: missed breakout -> compounding upside opportunity cost -> lower chance meeting monthly runway; strongest rebuttal no demonstrated netedge means chasing can lose principal and shortenrunway further. B Breakout bull case: price cross attracts followers -> dealer/liquidityresponse can amplify thinbook move -> continuation if sustained demand; bear chain marginal buyer exhausted -> reversal through wide spread -> delayed15minuteexit exceeds planned loss. C Funding squeeze: negative carry attracts arbitrage/shortcover -> basis normalizes -> spot need not keep rising; failedlongpayoffs alreadyshow this isnotreliable direction. D ETFcreation: newshares reflect investor/APdemand -> cashpurchase orinkindinventorytransfers absorb/reallocateBTC -> secondarymarketprice depends on offsettingsupply and hedgeunwind, not sharecountalone. Who benefits: venues/liquidityproviders earn fees/spread whether strategywins; issuer gainsassets/fees; users of underlyingnetworks can benefit withouttokenprice gains. Soundness: barproxy isprimaryvenue butnotallvenues; issueroneorigin despite multipleURLs; no liquidations/participants or globalnetflow observations establish causality. Resolution completedUTCsignals and net executablepayoff; unverifiedmedia rotation narratives discarded asentryevidence. Start235.86cash;goalsoundrepeatableedge;bindingconstraint validation, not349.01shortfall. Evidence movingprior: independentcostrobustforwardmeanwithuncertaintyabove0 plus reconciledexecutioninputs, not largerheadline. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf ; https://www.okx.com/api/v5/public/funding-rate .

Premises1 Refetch directproxy UTCdaily/hourly Dec1-3 forBTC/ETH/SOL; all72hour rows each complete. DailyDec2opens exceed hourlyaggregate by6.2064/7.9068/10.2992%; dailyvolume underreports sums82.7396/89.8276/80.2971%. Dailyclose matches hourlylast across3days; this does not validate open/range/volume. 2 CoinbaseBTCDec2open86282.35 agrees approximately Alpaca hourly86270.82, not daily91625.132. Crossvenue comparison is corroboration of morninglevel, not samevenue executiontruth. 3 LTC979calendar days/23492hour rows: four missing hours on3days;976complete. Maxcompleteopen discrepancy6.55235%,low6.20774%,volume98.4400%; dailyvsaggregatedclose tiny differences558days,max.240743%. Fixed20breakout+SMA50signals identical77days; stopranges can stilldiffer. 4 Alpacaofficialdocs include quote midprices; no-trade prices possible. This alone cannot explain omission of morning range and most volume. No causal vendor explanation obtained within8minuteaudit allowance; fullmajorhistoricalexpansioncut, namedgap. Initial30pagecap failed because responsepages~200rows despite requested10000, not repeatedtokens; rejectedtruncatedinputs. LTCfetch repartitioned90daywindows, paginationplusendcoverage, final979days completeendSep5 23UTC. Valid conclusion: rawdaily figures cannot support exact stopsize/volume claims; use prereg same-source hourlyaggregation only as sensitivity, not certified repair. Require 15minute/executablequotes and prospective evidence before adoption. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://api.exchange.coinbase.com/products/BTC-USD/candles ; https://docs.alpaca.markets/us/docs/historical-crypto-data-1 .

Premises:1 Issuer product page asofSep4 shares1,384,960,000; issuer CSV asofSep3 shares1,375,120,000, quantity784301.22680BTC. 2 Delta9,840,000 shares (+0.71557%) supports net issuance across snapshots, subject to asynchronous publication; using Sep4NAV45.15 gives $444.276m illustrative share-delta valuation, NOT verified cashflow or timestamped spot purchases. Basket22.65BTC, value$1,805,812.52 reflect Sep4 date. 3 Same page NAV fell2.14% Sep4; creations can coincide with falling price. Do not turn a dated Friday level into fresh Saturday demand. Strongest bullish argument institutional accumulation can absorb float and seed a squeeze; strongest bear counterexample arbitrage creations follow price/hedging and existing supply absorbs flow. Single issuer is not allETF totals; cashversusinkind, dealerinventory and intraday purchases unknown. Goal retain only pricepredictive demand; bindingconstraint historical timestamped flows plus validated conditional payoff. Conclusion WATCH structural demand, no flowtrade. Independent sources count1issuer despite2URLs. Primary issuer has marketing incentive; register facts closer to event than media. Sources https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf ; https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf/latest-holdings.csv .

Numbered premises:1 Read lessons72/71/70, notes then observations/analyses and fund trades. Onlylive BTC trade10 ended by operator liquidation -.081%; cannot calibrate natural stops/targets or execution slippage. Noopen/pending holdings, no due crypto monitors; ETH10/PAXG16armed. 2 Recorded fixed daily/hourly cost-adjusted funding tests127-129 reject longs: BTC p47.368% W3.2757% L3.3857%,N19;ETH p42.105% W3.6153% L4.0385%,N19;SOL p30% W4.0207% L5.3599%,N30. Recomputed pW-(1-p)L negative for all, quarterKelly clipped0. Dailyholdouts also negative; prior input gaps/cross-timeframe differences and15minfills unresolved. Not new test. 3 XRP79positive49.30% on6exits doesnot override meanCI[-1.19,+18.77]% and untested venue exits; SOL priorWATCH similarlyinsufficient. New completedcalendar refresh has no major or XRP entry. OnlyUSDT rawbreakout has no validated pricepayoff. 4 Primaryfee schedule reconfirmed15bpmaker/25bptaker; marketablelimit stilltaker. CurrentOKXnegative forecasts aren't settledBinancedaily signal and cannot reverse rejectedfunding verdict. 5 Starts235.86cash;goalrepeatableprofit;bindingconstraintvalidatedexecutableedge. Equity850.99 implies212.75positioncap and8.509plannedlosscap; fundcap235.86. Zeroqualifiededge=>zerosize before caps, no target/stop invented. Conclusion noPROPOSE/EXIT/ADJUST/FUND; cash. Valid given entry/edge mandate; soundness limited by historical selection, incomplete true universe, bars and execution modeling, not evidence every future rule fails. Strongestcounterargument profitableXRP/SOL exploratorytrend can justify future independent forwardstudy, not today's entry. Beneficiariesofturnover includevenue/liquidityproviders; first order unvalidated entry paysfees, second repeatedturnover reducescapital, third lesscapitalshortensrunway; consequencechain not priceforecast. Monthly349.01shortfall addsnoedge. Repairjob20 isolates6OKXrequests, timestampsfundingandOI independently, exitsnonzeroonmissing/stale; job19cancelled after savedreplacement. Tested locally6/6fresh and injectedfailure; firstsandboxpending. Jobs11/14retained. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://fapi.binance.com/fapi/v1/fundingRate ; https://www.okx.com/api/v5/public/funding-rate ; https://www.okx.com/api/v5/public/open-interest ; https://docs.alpaca.markets/us/docs/crypto-fees . No headline acted on, no orders placed.

1. Direct venue-proxy daily bars fetched16:35UTC, partialSep5 discarded,35/36symbols have21uniqueUTCdates Aug15-Sep4; USDG incomplete4/7recentdays. Observed-volume ranking includes stablecoins; onlyUSDT meets close>prior20closes. This is a price diagnostic, not a tested stablecoin strategy. 2. BTC/ETH/SOL/XRP completed closes79676.7475/2456.376/101.95/1.398615 versus thresholds81269.4915/2515.933/109.15/1.51966; allfalse. Quotes16:34:42–16:35:45 mid79951.753/2467.853/103.18/1.413905 are1.6214/1.9110/5.4695/6.9591% below thresholds; quotes are not completed signals. 3. 36-symbol observations2838-2873 and quote observations2874-2877 retain measured rows/source dates; calendar volume computed v*vw, no global-volume or live-Assets completeness claim. 4. Strongest counterexample: a new intraday surge could cross a level; that would not satisfy existing completed-bar entry timing or repair unvalidated exits. Watch status retained, no new rule/backtest or proposal. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots . Saved cross-timeframe inconsistencies167 remain unresolved; this refresh does not certify historical feed integrity.

Numbered premises:1 Crypto starts/ends235.86cash,zeroopen/queuedpositions; trade10operatorliquidation-.081%doesnotcalibratenaturalstops/targets orslippage. Applicablelessonscoverage53/smallN54implemented.2 Questiontree154/160 investigated settledfunding versus forecast,3venues,OIquantityprice,arbitrageincentives,conditionalstatistics andvenuecosts. DirectfundingJan2024-Sep2026 Binance/Bybit;OKX93completedcomparabledays.3 Fixeddaily/hourlynettest127-129: hourlyBTC-5.514%,N19,p47.368%,W3.276%,L3.386%;ETH-16.220%,N19,p42.105%,W3.615%,L4.038%;SOL-55.718%,N30,p30%,W4.021%,L5.360%. Dailyallnegative,doubledcostworse.4 Statistical130 SOLextremegrossforwardmean-3.543%,overlap/dependenceexplicit; preciseexecutionlimitedbybarinconsistencyandmissinghouraudit.5 Allcurrentregisteredentrysignalsfalse; singlevenueSOLnegativefundingdoesnotrepresentallvenues. Allplug-inquarterKellynegative =>size0. Conclusion noPROPOSE/EXIT/ADJUST/FUND; holdcash. Deductionvalid underpositiveexecutableedgerequirement;soundnessboundedbyhistoricalselection,feedinconsistency,feesimulation,missing15minfills andnocausalparticipantdata. Rejecttestedrule,notuniversalclaimnegativefundingmustcausefalls. Strongestcounterargument BTCtrainingpositiveanduptrendstrata+1.5%grosscanmotivatefutureindependentstudy;cannotoverridefailedholdout. SourceURLs https://fapi.binance.com/fapi/v1/fundingRate ; https://api.bybit.com/v5/market/funding/history ; https://www.okx.com/api/v5/public/funding-rate-history ; https://www.okx.com/api/v5/public/open-interest ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Fundamental122,OI123,simulation140,technical141,consequences150,CLI153recorded. PriorSOL/XRPbreakoutWATCHunchanged;nonewdatedcatalystneedsmonitor. Jobs11/14verified;newlocaltestedjob19every120min22runsreplacescancelled13,firstsandboxpending04:14. ETH10/PAXG16monitorsretained. Archive229observationsreadbackexactmatch,code/results/dailybars/settledratespersist;hourlyrawlocal/refetchrevisiongapnamed. Researchsaturatedforthisdecision, no threadoverrun; final5minreserveenteredabout04:12,finishabout04:15within43minuteplan/45ceiling. Monthlyfundingpressureaddsnoedge;nootherfundactions.

1. RepairedhourlyseriesendSep4 23UTC; BTC/ETH24428rows eachfromNov22 2023,SOL17737fromAug26 2024 16UTC. Missing4/4/7hours; holdout7823/7822/7823vs7824expected. No missingholdoutmidnight; SOLmissingOct11midnightisinpurge. 2. Comparingcompleteholdoutdays325/324/325: largestabs(dailyopen/hourlyopen-1) BTC6.2064%,ETH7.9068%,SOL10.2992%,allDec2 2025. Daily/hourlyopens91625.132/86270.82,3020.1375/2798.84,139.716/126.67 respectively. Causes unresolved; do notassume identicalinputprices or attribute everydaily/hourlyreturn differencetomanagementfrequency. 3. Actualholdoutentry comparatorsBTC19,ETH19,SOL30: maximumopendifference.467143/0/0%; oneBTCentrydiffers>10bp. 4. BTCMar10-Mar13 2026timeoutwin+2.335254%crossesmissingMar12 17UTC; ETH/SOLnaturalholdouttradescrossnogap. ThatBTCoutcomehasunobservedbarrierpath; cannotclaimexactstoptest. 5. Bothdaily/hourlyimplementationslossesandno currententry preserve rejection, but reportedreturnmagnitudesarediagnosticswithdatauncertainty, not executable validated performance. Independentcashledgerreconciliationprovesaccountinggiveninputs,notinputtruth. Primarydata https://runway.adityaperswal.workers.dev/data/crypto-bars ; fields/ranges,missinglistsandcomparisonsarchivedaudit.py/audit2.py. Need vendoraggregation reconciliation or independentquote/tradehistory before futurepositive adoption. RemainingmissingglobalOI/borrow/liquidationparticipantdata limits causalclaims. No optimistic repair/holdoutretuning performed.

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.

Start235.86cash,goalrepeatable netprofit,bindingconstraint independently validated executable edge. Premise1 negativefunding transfers short-to-long perpetualpayments,notspotyield. First order: a perpdiscount can reward longperp/shortspot arbitrage and penalize persistentperpshorts; liquidityproviders benefitfromspread/turnover. Second: shortspot hedges can offset outrightspotbuying while margin/collateralconstraints limit arbitrage, so rateconvergence neednot liftspot. Third: in sustainedselloffs negativefunding may persist with forcedlongliquidation or hedgedeleveraging rather than short squeeze. These chains are plausible hypotheses, not measuredparticipantflows; no borrow/collateral or liquidations panel establishes causality. Alternativechain: unhedgedshortscover -> perp/spotbasisnarrows -> spotarbitrage purchases amplifyrally, but test127-129 givesnegative netreturns and rejects treating that alternativeasbasecase. Premise2 singlevenue sign is incomplete: lastSOLdaily Binance+.2422bp vsBybit-.9447/OKX-.2478; sameperiod/differentvenue makesallpartlyright. Premise3 quantityOI increasesETH.487%/SOL.307% overabout114minutes do notrevealshortdirection. Conclusion retaincash,do notinvertfailedlongtest intoashorttrade/newfundwithoutseparateevidence. Strongestcounterargument BTCtraining+13.84% anduptrendconditionalpositivegrosswindows meritfuturestudy; theydo notrepairnegativeholdoutorjustifyretuning. Fundingdeadline cannotchange payoffgeometry. Sources https://www.okx.com/en-sg/help/how-to-calculate-future-funding-fee ; https://www.okx.com/api/v5/public/open-interest ; https://fapi.binance.com/fapi/v1/fundingRate ; https://api.bybit.com/v5/market/funding/history ; https://runway.adityaperswal.workers.dev/data/crypto-bars . Facts endatsettlements,OIandtests; causalchains remain assumptions.

CompletedSep4 spot closes BTC79676.7475,ETH2456.376,SOL101.95 exceed50daymeans68776.3257,2077.1341,83.2150;20day returns+26.425/+30.567/+35.419%. These correlated price indicators are not independent confirmations. Registered previousday Binance fundingdaily sums+1.9511,+1.6131,+.2422bp,so none qualifies negative-and-below-prior30day10thpercentile; thresholds+1.24586,+.69702,-1.07644bp. All currententriesfalse. Illustrative1ADRstop/2ADRtarget prices BTC76754.90/85520.44,ETH2337.63/2693.88,SOL95.42/115.01 derive from ranges but are not live orders, currentquotes or evidence of expectedprofit. FundingforecastSOL-.4808bp at04:04 is a different field/period/venue from completed Binance daily+.2422bp; no contradiction after definitions align. Binance/Bybit/OKX lastSOLdaily sums+.2422/-.9447/-.2478bp;93commoncomplete days have Binance/Bybit signagreement79.57%,Binance/OKX83.87%. Negativeonevenue is not a global consensus. Premises absententry and failednetbacktests independently prohibit purchase. No openholdings or queuedorders,so no R-progress or stopadjustment needed. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://fapi.binance.com/fapi/v1/fundingRate ; https://api.bybit.com/v5/market/funding/history ; https://www.okx.com/api/v5/public/funding-rate-history . Currentdaily snapshot not executablebidask.

NumPyseed90504,10000paths perasset: draw3 complete24h logreturnblocks from diagnostic holdout, concatenate72h; BTC/SOL325 blocks,ETH324. ReferenceSep4close,1ADRstop/2ADRtarget widths BTC3.667/7.334%,ETH4.834/9.669%,SOL6.405/12.810%. Hourlyclose target-before-stop frequencies5.57/7.02/3.73%,stopfirst30.27/31.76/24.10%; these are unconditional historical resampling frequencies, not forecast probabilities or intrabar/15minute fill odds. 72h terminal5/95percentiles BTC-6.423/+6.226%,ETH-9.029/+9.244%,SOL-9.596/+9.651%. Separate10-trade scenario draws observed net naturalpayoffs,invests10% of changing fundcapital: medianterminal-.241/-.881/-2.581%,lossfrequency57.70/71.76/93.33%. This is10trades,not monthlydeadline simulation; IIDpaths miss clustering/unknown tails. QuarterKelly is negative for all3 using observed win/payoff distributions, so authorizedsize0 beforecaps. Stop-only ceilings min(235.86,212.77,8.5108/stopwidth) BTC212.77,ETH176.05,SOL132.87; actual fills/fees/gaps can exceed nominal1%loss, these are illustrations not approvals. Premises negative netpayoff and modest72h targetfrequency jointly reject deployment; no target extension to meetfundinggap. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://fapi.binance.com/fapi/v1/fundingRate ; observations1762-1764 and code risk.py.

1. Oct14 2025-Sep1 2026 next3day open/open overlapping observations N323 perasset. Extreme negative previousday funding per fixed rule yields grossmean BTC-.6221%(28windows),ETH-.0827%(29),SOL-3.5426%(44), versus otherdays-.2193,-.3447,+.0184%. 2. Seven-calendar-day circular blockbootstrap10k draws gives extreme-minus-other95%CI BTC[-2.863,2.351]pp,ETH[-3.003,3.021],SOL[-6.204,-.904]. Returns overlap; event counts are NOT independent trades. 3. OLS forward3 return on previousday funding inbp and prior3day return, intercept; HAC7 SE funding coefficients BTC-.003560(SE.002443),ETH-.004857(.003435),SOL+.003588(.001685) fractionalreturn perbp. SOL positive coefficient means more negative funding associates with weaker subsequentreturns; association, not causal identification or validatedshort. 4. ExploratorySMA50 strata show BTC/ETH uptrend extreme means+1.52/+1.51% on6/9windows; SOL-2.12% on10. These inspectedholdout strata cannot justify adding a trendfilter and calling it OOS. 5. Net naturaltrade circular3trade intervals BTC[-2.133,1.622]%,ETH[-3.370,1.723],SOL[-4.589,-.656] reinforce zero sizing. Currentuniverse, multiple comparisons, selectedregimes and datarevisions limit inference. Conclusion reject fixedrebound rule; Solana evidence favors continuedweakness in this period, not universal prediction. Prior unknown edge revised against this rule; independent preregistered prospective positive net mean with positive uncertainty bound and exact exits would change sizing, a freshnegative quote would not. Sources https://fapi.binance.com/fapi/v1/fundingRate ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; rawmodels androws saved.

Frozen notes154/160: prior completedUTCday Binance settledsum<0 and below previous30daily10th percentile; next00spotopen, signaldayADR20 stop1,target2,72h timeout,+1R completedbar trims1/3 nextopen and stoptoentry. Daily/hourly use same dailyentry signal but management cadence differs; neither15minute venue replication. Jan1 2024-Oct10 2025training,Oct11-13purge,Oct14 2025-Sep4 2026holdout; SOL spot startsAug26 2024,20daywarmup required. Historical price windows previously inspected for unrelatedrules, so chronological diagnostic holdout, not untouched prospective sample. Per-side25bpfee+10bpslippage, twiceeach stress. Explicit vectorbt longonly orders on synthetic withinbar event timestamps; independent cashledger error<1e-6. Gapstops open, both-touchstopfirst, targetgap cappedtarget, entrybar barriers included, no terminalopen in base results. Daily holdoutreturn -55.1896368834248%,DD60.06658062115767%,N30; hourlyreturn -55.71775975879778%,DD61.324549216845185%,N30,win30%,W4.0207271602503845%,L5.359893722342115%. Hourly training -14.717679400439575%; doubledcostholdout -64.10559722985161%. Training-only5/10/20quantile sweep recorded,10 retained without holdout tuning. Conclusion REJECT this contrarian implementation, not every funding strategy. Negative plug-in quarterKelly clipped0; no shorting inference authorized. Missinghours, publicationlatency, cash-fee approximation vs creditedassetfee and actual15minute fills remain limitations. Source https://fapi.binance.com/fapi/v1/fundingRate ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Complete rows/trades/code archived separately.

Frozen notes154/160: prior completedUTCday Binance settledsum<0 and below previous30daily10th percentile; next00spotopen, signaldayADR20 stop1,target2,72h timeout,+1R completedbar trims1/3 nextopen and stoptoentry. Daily/hourly use same dailyentry signal but management cadence differs; neither15minute venue replication. Jan1 2024-Oct10 2025training,Oct11-13purge,Oct14 2025-Sep4 2026holdout; SOL spot startsAug26 2024,20daywarmup required. Historical price windows previously inspected for unrelatedrules, so chronological diagnostic holdout, not untouched prospective sample. Per-side25bpfee+10bpslippage, twiceeach stress. Explicit vectorbt longonly orders on synthetic withinbar event timestamps; independent cashledger error<1e-6. Gapstops open, both-touchstopfirst, targetgap cappedtarget, entrybar barriers included, no terminalopen in base results. Daily holdoutreturn -15.623693834952867%,DD29.962922885468313%,N19; hourlyreturn -16.220236291290966%,DD31.642970314482778%,N19,win42.10526315789473%,W3.615324451398019%,L4.038497907270158%. Hourly training -21.556906551456233%; doubledcostholdout -26.653849568026633%. Training-only5/10/20quantile sweep recorded,10 retained without holdout tuning. Conclusion REJECT this contrarian implementation, not every funding strategy. Negative plug-in quarterKelly clipped0; no shorting inference authorized. Missinghours, publicationlatency, cash-fee approximation vs creditedassetfee and actual15minute fills remain limitations. Source https://fapi.binance.com/fapi/v1/fundingRate ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Complete rows/trades/code archived separately.

Frozen notes154/160: prior completedUTCday Binance settledsum<0 and below previous30daily10th percentile; next00spotopen, signaldayADR20 stop1,target2,72h timeout,+1R completedbar trims1/3 nextopen and stoptoentry. Daily/hourly use same dailyentry signal but management cadence differs; neither15minute venue replication. Jan1 2024-Oct10 2025training,Oct11-13purge,Oct14 2025-Sep4 2026holdout; SOL spot startsAug26 2024,20daywarmup required. Historical price windows previously inspected for unrelatedrules, so chronological diagnostic holdout, not untouched prospective sample. Per-side25bpfee+10bpslippage, twiceeach stress. Explicit vectorbt longonly orders on synthetic withinbar event timestamps; independent cashledger error<1e-6. Gapstops open, both-touchstopfirst, targetgap cappedtarget, entrybar barriers included, no terminalopen in base results. Daily holdoutreturn -6.9180930956961095%,DD22.964373866351174%,N19; hourlyreturn -5.514402823344822%,DD22.96892955424241%,N19,win47.368421052631575%,W3.2757027686202886%,L3.385744168911633%. Hourly training 13.83697777317403%; doubledcostholdout -17.281279901086677%. Training-only5/10/20quantile sweep recorded,10 retained without holdout tuning. Conclusion REJECT this contrarian implementation, not every funding strategy. Negative plug-in quarterKelly clipped0; no shorting inference authorized. Missinghours, publicationlatency, cash-fee approximation vs creditedassetfee and actual15minute fills remain limitations. Source https://fapi.binance.com/fapi/v1/fundingRate ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Complete rows/trades/code archived separately.

Premises:1 Direct OKX quantities saved02:11 permit matched04:05 comparison.2 USD OI equals coin OI times an implied mark, so logUSD change decomposes exactly into logquantity plus logprice.3 Contracts contain paired long and short exposure and can be hedges; no liquidation-side or participant data was collected. BTC contracts-.070625%, implied price+.012197%, USD-.058436%; ETH contracts+.487341%,price-.000408%,USD+.486931%; SOL contracts+.306840%,price-.039289%,USD+.267431%. Conclusion: ETH/SOL OI growth in this short interval mostly reflects quantity, not price, but neither magnitude nor direction proves an extreme or squeezable short book. Funding forecasts BTC-.173037bp,ETH+.884251bp,SOL-.480786bp; lastsettled+.007019,+.575933,-.852469bp respectively. Valid descriptive decomposition, not causal direction. Strong bull counterargument negative SOL funding with increasing quantity is consistent with short buildup; equally compatible with hedged long spot or persistently weak demand. Need historical conditional price evidence and broader venue baselines. Sources https://www.okx.com/api/v5/public/open-interest ; https://www.okx.com/api/v5/public/funding-rate ; recorded1187-1189/1681-1683. Existingjob13 saves onlyUSD; add separate compact quantity collector if needed, not trading authorization.

Premise1: OKX's direct mechanism description says negative rates transfer funding from short to long perpetual positions; its contract notional determines payment. Premise2: this fund can hold cash spot, not short perpetuals or borrow spot; no funding transfer enters its cash returns. Premise3: negative funding can accompany a perp discount and arbitrage demand but does not identify whether an account is directionally short or hedging spot. Conclusion: only a subsequently measured spot return, net of both-sided costs, can qualify the proposed contrarian signal. Valid deduction; soundness depends on instrument definitions, not a claim funding cannot predict price. Illustrative -1bp every8h totals -3bp/day or -9bp/72h to shorts, not guaranteed constant and not income to our spot. Baseline modeled round-trip fee/slippage about70bp, far above that hypothetical9bp; do not confuse transfer and price profit. Strongest bull case forced short covering can produce gains much larger than funding; bear case negative rates persist with spot selling. No intrinsic token valuation derived from this metric. Primary sources readSep5: https://www.okx.com/en-sg/help/how-to-calculate-future-funding-fee ; https://developers.binance.com/docs/derivatives/usds-margined-futures/market-data/rest-api/Get-Funding-Rate-History ; https://bybit-exchange.github.io/docs/v5/market/history-fund-rate . Exchange methodology sources close to mechanism but commercial, three venues are independent records rather than copied commentary.

1. Start state: crypto ledger has one previously closed operator liquidation, zero open or pending positions, $235.86 deployable. No holding has R-progress, time horizon or exits to adjust; no order authority exercised. Prior BTC forced loss does not calibrate natural payoffs. 2. Expanded all36 published USD pairs;35 have complete7day observations. Calendar correction78 supersedes initial67 USDG volume4.103M: observed2.103M across4days, missing3 unknown. Published inventory is not live account Assets census, and USD-only ranking is not cross-quote/global volume. Only USDT has apparent completed breakout; prior-price anomaly and stablecoin economics unvalidated. All shortlisted nonstable entries false. 3. Registered tests79-82: XRP daily20 WATCH (net+49.30%,6 exits,p66.67%,W13.299%,L3.691%,meanCI[-1.194,18.765]); LINK daily20 REJECT (train-39.71%,holdout+.295%,doublecost-7.786%); UNI daily20 REJECT for adoption (train-35.39%,markedholdout+13.07% includesopenposition,13closedmean-.346%,quarterKellynegative); PAXG daily20 REJECT (net-4.10%,8exits,no pre2026training). All separate20observed-hour rules REJECT after71.6-83.4%losses. Identicaldaily signals onhourly matchreturns, higherDD; independentledger verifies. 4. Primary USDC netissuance+.61B/7d, XRPlogo sponsorship, Circle/Chainlinkreserve integration and UNIburnmechanism are real source claims, but do not quantify current marginaltoken demand or a short-horizon edge. PAXGfee change is issuerconversion notAlpacaspot fee. OIUSDchanges confound price/quantity. Statistics90 andsimulation91 leaveactualsize0. Conclusion: no PROPOSE/EXIT/ADJUST/FUND. Valid deduction under required independent executable edge; soundness bounded by selected universe, smallsamples, quote-based bars, missinghours, IIDbootstrap and untested15min/stop-target/+1Rexecution. Strong counterargument XRPpositive training/holdout and realinfrastructure meritresearch; they do not remove absenceofentry or independentconfidence. Prior66 core verdict unchanged (BTC/ETHdaily rejected,SOLwatch). XRP joins watch only. Goal repeatable net gains; bindingconstraint validatededge, not fundingdeadline. No risk added to recover$348.92. Continuity: existingjobs11/13 retained; newtestedlocal job14 daily30runs explicitcalendarcoverage/eligible0; firstsandboxpending atfinalcheck. PAXGmonitor16 Sep30 noon chosenreviewtime beforewaiverexpiry; ETHmonitor10 unchanged. XRP frozen prospective assessment Sep6 onward, earliestOct6 AND>=30closedtrades; extend ifinsufficient, no retuning inspectedhistory. Gaps: liveAssets/fullcrossquotes, missingUSDGdays, matchedpriorIBITshares, exchangeflows, currentUNIburn/float and completeunlockmap, actualvenueexits. FullUSstock/nonUSstockscreen outside crypto mandate; no newfund thesis justified. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://www.circle.com/transparency ; https://www.circle.com/blog/cirbtc-a-clearer-reserve-standard-for-wrapped-btc ; https://developers.uniswap.org/docs/ecosystem/governance/uni ; https://support.paxos.com/articles/2899561282-PAX-Gold-Fees ; sourceURLs inobservations318/319. Codearchive131-134, recipe135. Research completed~14minutes within43minuteplan; finalrecordingreserve preserved, no branchoverrun.

Executed required CLI: node /app/agent/dist/backtest.js --symbols XRP/USD,LINK/USD,UNI/USD,PAXG/USD --strategy breakout --lookback 20 --days 730. Individual return/drawdown/win/sample observations saved. Source inspection confirms same-signal-close entry, no fee/slippage and current-day inclusion; full-window control not chronological validation. Custom registered vectorbt tests79-82 use nextopen and net costs, daily and hourly, training-only sweep, separately excluded terminal marks, independent cash ledger. No parameter chosen from CLI. Positive control cannot override failed costs, small N or absent live entry. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; /app/agent/dist/backtest/engine.js.

Direct OKX USDT swaps, Sep5 02:11 versus 00:15 baseline: USD OI BTC -.831%, ETH +.241%, SOL +.122%. Funding forecasts -.115/.801/-.599bp versus last settled .007/.576/-.852bp. Premise1: single venue USD changes confound price and contract quantities. Premise2: prior collector retained only USD OI, so net contract growth unknown. Premise3: daily breakouts false and core verdict66 unchanged. Conclusion no leverage-extreme or squeeze entry. Current raw contract/coin quantities now observed for future comparison. Strong bull case negative SOL forecast may reflect shorts that cover; bear case persistent adverse expectations; neither gives conditional payoffs. Sources https://www.okx.com/api/v5/public/funding-rate and https://www.okx.com/api/v5/public/open-interest . Existing job13 remains research only.