1 Querytrades shows no holdings/pending/closing; all3closedlosses, MIRM-.4393%,IONS-2.8424%,RARE-45.8601%. 2 Twooperatorliquidations censor intendedhorizons; do not fit strategywinrate/slippage distribution toN3. 3 RAREactual6.44R shows unboundedstop risk; contemporaneousquotes/sessiondata missing, so do not attribute all shortfall tobroker slippage. Conclusion no liveexits toreset and no empiricallypositiveKelly here. Counterexample: twoforcedclosures could laterhavewon, which prevents universalnegativeedge inference, not entryvalidation. Full49.14fund at historicalRAREloss wouldlose22.5356=2.6482%account850.99; this is stressscenario,notworstcase. Wholefundtotal loss5.774%account exceedsfewpercent. Source runway:query_trades?fund=catalyst .
sample3
uncensored_sample1
realized-31.94
win_rate0
RARE_loss_pct45.86
RARE_loss_R6.44
fund_gap_scenario_loss22.54
account_gap_scenario_pct2.65
QTR9110000 three-step adverse chains and beneficiaries
Premises are the primary cash/procedure evidence in this run's fundamental analysis, not probabilities. AVAV: first manufacturing expansion and inventory/unbilled growth tie up cash; then poor conversion and SCDE losses can require liquidity or slower investment; then financing costs or lower expected returns can compress equity value despite funded backlog. Suppliers paid for expansion benefit before shareholders; counterchain timely acceptance yields cash and operating leverage. FTK: first infrastructure delays postpone generation while chemistry forecast must become delivered invoices; then slower receipts tie up working capital and weaken customer funding; then financing or idle capacity reduces equity cash flow despite attractive H2 revenue narrative. Infrastructure providers and customers benefit if suppliers fund ramp; counterchain documented fuel/restart/paid invoices reverses it. TXNM: first intervenor objections may force protections or more process; then concessions and elapsed time alter financing/transaction economics; then merger-spread expected return falls if close probability or date worsens. Ratepayers can benefit from protections even if shareholder transaction economics weaken; counterchain accepted conditions accelerate closing. These are conditional causal scenarios, not claims that any new adverse event occurred tonight. An issuer publication and the same filing mirror are one origin, not independent corroboration. Empty ledger means no thesis-breaking live holding to sell. Sources https://investor.avinc.com/node/21396/html ; https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf ; https://www.txnmenergy.com/investors/rates-and-filings/pnm-nmprc-filings.aspx .
theses attacked3
consequence steps per thesis3
live holdings0
new verified adverse events0
QTR9110000 attack the empty book's watch theses: cash and regulatory risks
No live holdings exist, so no actual stop/target resets. Numbered premises: 1 AVAV Sep9 Q1 endedAug1 primary original unchanged; CFO13.496m less property44.033m and capitalizedsoftware5.417m gives -35.954m broader FCF, versus -30.537m property-only definition. Inventory +31.2978%, unbilled +11.8203%; AxS EBITDA margin18.0031%, SCDE-6.6131%. Strongest bear: demand/backlog requires cash investment while SCDE losses and delayed acceptance erode cash return; positive EAR could be beta-relative relief, not durable price drift. Counterexample: profitable funded backlog converting promptly can reverse working-capital use, so neither insolvency nor price target inferred. 2 FTK unchanged full investorPDF forecasts H2 international >40m versus H1 12.6m; forecasts are not paid receipts. Strongest bear: infrastructure/fuel/logistics dependencies delay revenue and consume working capital; counterexample actual profitable receipts/restart can support value. 3 TXNM mirror latest acquisition entryAug18; original order allows responsesSep9, not approval; actual docket responses unreadable. Strongest bear: ratepayer protections or financing changes impair timing/economics, while delay annualizes downside carry; counterexample accepted protections remove uncertainty. Conclusion no upgraded deal odds or live valuation target; claims remain bounded to originals read. Valid inference that missing cash/odds cannot substantiate payoff; soundness constrained by issuer bias and missing regulator text. Prior remains watch; independently verified cash conversion, accepted protections and validated price-rule odds would move it. Sources observations23003/23004; https://investor.avinc.com/node/21391 ; https://investor.avinc.com/node/21396/html ; https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf ; https://www.txnmenergy.com/investors/rates-and-filings/pnm-nmprc-filings.aspx ; https://e360.prc.nm.gov/portal/public/ .
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 .
BTC_return25.67
BTC_DD13.54
BTC_win32.65
BTC_N49
ETH_return26.22
ETH_DD24.94
ETH_win21.05
ETH_N38
SOL_return55.5
SOL_DD18.61
SOL_win52.94
SOL_N34
POST910 final decision: no orders; correct Kelly units and preserve prospective tests
Start state: Catalyst49.14cash, no holdings/queued orders. Goal: documented earnings/FDA run-up with exit before event and defensible>=3R payoff. Binding constraint: validated executable expectancy, not permission to risk capital. 1 Original EPM1928 training−6.551%,N6 holdout CI[-1.779,3.065]% fails; finished day +.989R is not a completed independent event outcome. Dividend.12 afterclose adds news, not payout coverage. 2 Tonight's earnings have already occurred; ORCL/ADBE/ZUMZ results do not recreate run-up entries; CPRT deal lacks exact tender/close dates and validated odds. 3 TLX/SRRK early-decision/full-pipeline risk unresolved. Conclusion no orders, retain49.14cash. Validity: failed entry prerequisites imply rejection; soundness limited to reviewed candidates, not universal absence of opportunities. Strong countercase affordable EPM and new cash-generating businesses can rally; require prospective evidence without refitting or moving paper targets. CORRECTION supersedes Kelly labels2322/2324 and priorjournal: p4/6,W.02643209564,L.03262590862 yields loss-budget halfKelly.1276117927 but invested-capital halfKelly .5*(p/L−q/W)=3.9113636407. Simulation actually used fixed12.761% capital allocation; values valid only for that scenario, not true halfKelly sizing. Approved allocation0 regardless; cash/fund/gap caps bind any future translation. Accountnormalrisk25.5297/convex42.5495,positioncap212.7475,fund49.14; monthly349.01shortfall excluded. Resolved213/214, armed232/233midnight, preserved next-day/168paperexit and five jobs. Sources: runway:query_research?fund=catalyst ; runway:query_trades?fund=catalyst ; https://ir.evolutionpetroleum.com/news-releases/news-release-details/evolution-petroleum-schedules-fiscal-q4-and-full-year-2026 ; https://telixpharma.com/feed/ ; https://investors.scholarrock.com/rss/news-releases.xml ; https://investor.oracle.com/files/content_files/1q27-pressrelease-September_FINAL.pdf ; https://www.adobe.com/cc-shared/assets/investor-relations/pdfs/01906202/au56y4ter.pdf .
fund_cash49.14
orders0
approved_allocation0
EPM_paper_R0.99
train_pct-6.55
halfKelly_capital3.91
halfKelly_loss_budget0.13
simulation_fixed_allocation0.13
POST910 cash recipients and obligations after announcements
Edge reference: frozen HITI N9 issuer-dated earnings daily responses, median absolute5.494505%; at S2.56 Oct16K2.5 C.31 requires+9.765625%,P.16 requires-8.593750%,each1/9=11.111%<40%; no >=3x expected-payoff mispricing documented. HAIN/EPM/ZUMZ have N0 validated comparable-event samples in this review, meaning untested, not0% wins. Premises: (1) HAIN .33 Sep9close was a13:59 trade with stock.681–.6839, intrinsic.3161–.319; newer.37 Sep10 13:55 traded with stock.6484–.6583,intrinsic.3417–.3516. Thus both marks exceed entire matched-minute intrinsic ranges. (2) HITI empirical responses22929–37 unchanged; NumPy recomputation uses signed returns. Fourofnine needed: price ceiling.0015525call/.08065934put atfixedS2.56, beforecosts. Validity: matching time removes asynchronous false comparison; soundness: minute data do not establish simultaneous executable quotes, adjusted deliverables or fairvalue. Counterexample: an unobserved executable cheaper ask could change decision, but none documented. HAINstandardK1 at.37 maximum payout2.702703xpremium even atS0, less than3xexpectedpayout and less than4xprice neededfor3Rnettarget onfullpremiumrisk. Size0. HAIN>=8eventcollection deferred because present candidate alreadyfails payoffhardbound; not presented as failedbacktest. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; /data/option-bars; option_contracts. Issuer datesSep14HAINBMO https://ir.hain.com/news-releases/news-release-details/hain-celestial-announces-fiscal-2026-fourth-quarter-and-full andHITIAMC https://hightideinc.com/high-tide-to-announce-third-fiscal-quarter-2026-financial-results/ .
Premises: 1. Re-read2183–2185 costed vectorbt tests:2025train, REUSED2026Jan1-Sep9holdout,25bpfees/side10bpslippage20stress; all36hourlycellslose and supplied dailynegative. 2. Python recomputation22952 pW-(1-p)L and .5*(p-(1-p)/(W/L)) produces negative expectancy and allocation for all3. 3. Sole actualBTC10 operator liquidation does not calibrate setup odds or order slippage. Conclusion zero allocation to recorded compression rule. Validity: positive-edge requirement fails; soundness limited by smallN, reusedholdout, unresolved historical aggregation errors. Counterexample: an independent different rule may profit, so rejection is not a bearish price forecast. Current>=3Rtarget/+2Ratmostquarter exits differ from frozen1.5R rule and need independentdaily/hourly plus15minexecution evidence. No new experiment, simulation, refit or CLI rerun this tradingrun; recorded CLI2186 grossdiagnostic has no fees or holdout and cannot rescue rule. Source https://runway.adityaperswal.workers.dev/data/crypto-bars .
BTC_E_pct-1.27
ETH_E_pct-1.77
SOL_E_pct-0.74
BTC_halfKelly-0.23
ETH_halfKelly-0.35
SOL_halfKelly-0.1
BTC_N12
ETH_N9
SOL_N8
approved_notional0
QTR9102100 existing holdout and capital audit retain zero live sizing
Premises:1. Fresh fundledger3closed operatorliquidations,0active/pending/closing; censoring precludes stop/target odds and execution-slippage inference.2. Saved2284/2208/2209 costedgap holdoutN2151 return-26.212876%,DD27.351241%,win43.979544398%,mean monthlyclusterCI[-.165928,-.052452]%,stressreturn-56.878654%; not rerun.3.Python p=.43979544398,W=.617581664%,L=.682768517% yields mean-.1108804318176%,halfKelly=max(0,.5*(p-(1-p)/(W/L)))=0. Conclusion cash187.22; positioncap212.7475 and normalaccount risk25.5297 do not force use. Validity arithmetic correct conditional on empiricalpayoffs; soundness constrained by saved vendorhistory/costmodel/selection. Regimechange is counterexample but requires independentcosted evidence. No newMonteCarlo or simulatedexitodds claimed: negativeexpectancyalreadyfailsgate. Sources22967 and saved2284; https://runway.adityaperswal.workers.dev/data/stock-bars .
saved return pct-26.21
saved drawdown pct27.35
win rate43.98
sample2151
mean net pct-0.11
half Kelly0
fund cash USD187.22
QTR9102100 original evidence has not established incremental cash or merger odds
Premises: (1) Frozen note1683 sets t=Sep9 issuer release16:10EDT, t-2=Sep4 (Sep7 holiday), t+1=Sep10, t+2=Sep11. (2) SIP adjustmentall 28bars each, no pagination, end20:30Z obtained afterclose; Sep4/10 AVAV144.65/147.065 and SPY770.19/757.83 imply +1.669547% minus -1.604799%=+3.274346pp EAR. (3) Prior20session ADR Aug13–Sep10=8.759085. Conclusion: freeze prospective Sep11actualopen paper entry, stop entry-8.759085, target entry+13.1386275; half at+8.759085 then BE following session for dailyproxy, five-session horizon exitSep17close if not exited earlier. No entry price yet exists. Keep historical44cells and cost assumptions1bpfee+5bpslippage eachside/stress1+15 unchanged. These are paper rules, not current live exit instructions: original1.5R target and+1Rhalf differ from current>=3R and+2Rquarter. Saved N9dailyCI[-.8066,4.6100]%/15min[-.7874,5.5167]% cross0; no independently validated live sizing. Validity positiveEAR->paper follows frozen rule; soundness conditional on vendor daily corrections, no inferred execution. A positive event could drift but cannot calibrate success probability from one observation. Initial future-ended SIP request403 fixed by delayed cutoff with samefeed. Sources obs22964/22965; https://investor.avinc.com/node/21391 ; https://runway.adityaperswal.workers.dev/data/stock-bars .
EAR pp3.27
AVAV return pct1.67
SPY return pct-1.6
ADR USD8.76
paper target distance USD13.14
sample9
live allocation USD0
POST910 trading decision: no qualifying after-hours limit order
Premises observations22954/57/59/60 from https://investor.avinc.com/news-events/events-and-presentations ; https://www.sec.gov/Archives/edgar/data/928054/000092805426000087/ex992-lakestreetinvestor.htm ; https://ir.eia.gov/wpsr/table7.csv ; https://www.census.gov/wholesale/pdf/mwts/currentwhl.pdf . Firstorder demand/flowmix requires equipment,inventory,shipping. Secondorder scarce-capacity vendors may capture rent, but fuel,delays and receivable financing absorb cash. Thirdorder profitable accepted deliveries fund capacity and erode scarcity; weak collections transfer returns to lenders/suppliers. These are conditional mechanisms, not measured stock returns. Start278.67cash; goal positive asymmetric net payoff; bindingconstraint executable validated edge plus incremental cash evidence. Strong counterexample: firm profitable orders may price before receipts; current failed/small selected tests still do not identify entry odds.
fund_cash278.67
approved_usd0
POST910 frozen FTK expectancy and current flow arithmetic
Python recalculation of obs22959-61, not a new backtest or independent sample. https://ir.eia.gov/wpsr/table7.csv Sep4 total11.868m b/d times7=83.076m barrels; weeklyratechange-.028 means-.196m barrels overweek. Components3.417crude/8.451products do not establish tanker tonne-miles. Census July rounded +2.3%inventory/+.1%sales implies1.023/1.001-1=2.197802% intensitygrowth; nominal survey excludingmanufacturers branches https://www.census.gov/wholesale/pdf/mwts/currentwhl.pdf . FrozenFTK1895 historical https://runway.adityaperswal.workers.dev/data/stock-bars N10p.30,W19.6007341/L19.2995217% gives expectancy-7.62944496% andhalfKelly-.1946214086 clipped0. Small selected sample cannot identify future distribution, but cannot support positive sizing from this rule. No simulation needed to overturn negative arithmetic.
FTK_expected_pct-7.63
raw_half_Kelly-0.19
approved_half_Kelly0
FTK_n10
total_week_m_barrels83.08
total_week_change_m-0.2
inventory_intensity_pct2.2
POST910 new GD filing is tax withholding; cash-capture gaps persist
Premises1 observations22954-60 and GD raw XML: https://www.sec.gov/Archives/edgar/data/40533/000119312526387811/ownership.xml newSep10Form4 coversSep8taxwithholding, not aircraft sales. 2 https://data.sec.gov/submissions/CIK0001120370.json latestAug21amendment leaves Abilenecompletion unknown. 3 https://www.sec.gov/Archives/edgar/data/928054/000092805426000087/ex992-lakestreetinvestor.htm retainsH2>40m versusH112.6mforecast; no incremental receipts verified. 4 https://investor.avinc.com/news-events/events-and-presentations archive same404replay leaves commentary unknown; retained priorQ1FCF-35.954m is not refreshed cash. 5 https://www.census.gov/wholesale/pdf/mwts/currentwhl.pdf July inventory outgrows sales; https://investor.tsmc.com/english/monthly-revenue/2026 revenue unchanged. Conclusion no new fundamental trigger. Validity: these sources do not establish incremental capture; soundness limited to bounded pages, not proof no receipts exist. Counterargument cash can lag profitable investment, so no short thesis inferred. Signed aircraft instrument remains unverified at https://www.whitehouse.gov/presidential-actions/ .