Premises:1.Census July nominal inventories grew faster than sales; ratio1.20 remains below1.28yearago. Both partly support stockbuild concerns and longer-run inventory normalization; different horizons are not contradiction. Use revised June794907/946687 million, not old rounded baseline. Survey excludes manufacturers branches and agents/brokers; nonresponse and price changes limit inference. 2.AugustPPI+.4%MoM/+5.4%YoY persists; cannot deflate July wholesale with August totalPPI. 3.TXNM actual response texts unavailable; originalAug18order permittedSep9responses,notdecision. 4.FTKfullPDFidentical, forecastcashconversion unverified. Conclusion: mixedmacrocontext, no issuerdealprobability/stockedge. Counterexamples: inventory builds can anticipate demand; nominal sales gains can be inflation; filings may exist offmirror; FTKmayreceivecash elsewhere. No claim of recession, approval or default. Observations22842–46. Sources https://www.census.gov/wholesale/current/index.html ; https://www.census.gov/wholesale/pdf/mwts/currentwhl.pdf ; https://www.bls.gov/news.release/ppi.nr0.htm ; https://www.txnmenergy.com/investors/rates-and-filings/pnm-nmprc-filings.aspx ; https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf .
July sales bn801.3
July inventory bn958.9
sales MoM pct0.8
inventory MoM pct1.3
ratio1.2
June revised sales million794907
June revised inventory million946687
PPI MoM pct0.4
PPI YoY pct5.4
allocation0
QTR9101350 decision: retain $187.22 cash and no proposals
Premises: 1. Agency August PPI is current through web retrieval, despite direct403. Energy/transport account for concentrated pressure while underlying prices still rise; July upward revisions require same-vintage comparisons. 2. Census July is pending at13:52 before scheduled14UTC; June nominal values cannot establish current real activity. 3. TXNM latest mirrored acquisition filing is Aug18; original p3A permits responses bySep9, not approval. Actual responses remain unknown despite regulator summary and public JS/config inspection. 4. FTK full PDF hash matches12:33, with international forecast and paused Montana unchanged. Conclusion: new macro context, no validated stock edge. Counterexamples: energy pressure may reverse; wholesaler nominal growth may be price inflation; responses may exist offmirror; FTK forecasts may convert to cash. These limit directional claims. Who benefits: energy suppliers may gain pricing but margins require costs; customers may absorb higher freight before passing it on. Primary-source dates/periods and gaps recorded22809/10/12/14 plus fullPDFobservation. Sources https://www.bls.gov/news.release/ppi.nr0.htm ; https://www.bls.gov/news.release/archives/ppi_08132026.htm ; https://www.census.gov/wholesale/current/index.html ; https://www.txnmenergy.com/investors/rates-and-filings/pnm-nmprc-filings.aspx ; https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf .
PPI monthly pct0.4
PPI annual pct5.4
ex food energy trade annual pct4.7
July revision pp0.1
FTK PDF pages28
allocation0
QPRE91012 trading decision: retain $187.22 cash; no eligible recorded rule
1. TXNM Aug18 original order paragraphA authorizes responses bySep9; current issuer list has no newer acquisition entry. Regulator summary and e360JSshell do not establish response contents; no acceptance/nonfiling claim is sound. Prior fee175m/replacement396m are carried from304,not reverified today. 2. FTK28page deck asofSep9 forecasts360–370m sales/50–54mEBITDA and>40mH2international; initialMontana powerpaused andphase2underdiscussion. Forecast is not receipt, unchangedlinkhash is not unchangedPDFproof. H1FCF-7.819m carried2178,image table not newly extracted. 3. AVAV original Q1FY27 release cash13.496-44.033-5.417=-35.954m, not positive free cash. 4. LOVE issuer Q2FY27 sales161.2m,+.4%,grossmargin68.4% includes20m tariff recoveries in costofsales;21mrefund/interest cash received,adjustedEBITDA-1.3m. Strong cash-receipt evidence and weak adjusted earnings can coexist; do not label all margin growth organic. Counterexamples: TXNM may have timely filings not mirrored; FTK forecast may convert; AVAV investment may earn future returns; LOVE refund cash is real. None establishes an independently tested current price rule. Watch,actualsize0. Sources: https://www.txnmenergy.com/investors/rates-and-filings/pnm-nmprc-filings.aspx ; https://www.prc.nm.gov/pnm_acquisition_case_information/ ; https://ir.flotekind.com/image/Lake-Street-Investor-Presentation.pdf ; https://investor.avinc.com/node/21391 ; https://investor.lovesac.com/news-events/press-releases/detail/167/the-lovesac-company-reports-second-quarter-fiscal-2027 . Detailsobservations22779–22782.
FTK revenue low m360
FTK revenue high m370
FTK EBITDA low m50
FTK EBITDA high m54
AVAV cash after investment m-35.95
LOVE sales growth pct0.4
LOVE adjusted EBITDA m-1.3
LOVE refund receipt m21
QGAP91008 final: reject opening-gap adaptation and retain pre-open cash
Start: Quant187.22 cash/account850.99, no holdings or pending orders. Goal: repeatable net returns; binding constraint: validated expectancy with feasible execution.
1. The author-hosted Lou/Polk/Skouras2019 paper motivates a question; its monthly stock portfolios do not validate our ETF day rule.
2. Frozen gap≤−0.5ADR rule loses8.8855% in2017–19 training. In2020–25 it loses26.212876%, drawdown27.351241%, over2151 trades; stressed return−56.878654%, drawdown56.928996%. Net win43.9795%, average win/loss0.617582%/0.682769%; quarter Kelly0.
3. Monthly-cluster mean CI is entirely negative; even gross CI spans0. QQQ's profitable neighboring cell fails stress. All six portfolio holdout variants lose.
4. Fractional eight-sleeve/PDT/09:45 execution assumptions are not a deployable187.22 book. A2.455R sampled loss also prevents treating the nominal stop as a hard loss bound.
5. FTK's recovered deck retains negative H1 cash after capex and a Montana pause; AVAV keeps capital-intensive guidance. Neither validates this price rule. Preserve frozen AVAV EAR and issuer/macro monitors.
Conclusion: no PROPOSE, EXIT, ADJUST or FUND; actual size0. Validity: failed predeclared edge gates imply cash. Soundness remains conditional on vendor history, survivor universe, assumed costs and sampled prints; this does not refute all published reversal effects. Change requires independent cost-stressed evidence and feasible execution, not funding pressure or a new fit to this holdout.
Sources: https://personal.lse.ac.uk/polk/research/TugOfWar.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min ; issuer analyses2161/2178 and calendar note1889.
return-26.21
drawdown27.35
sample2151
month clusters72
win rate43.98
avg win pct0.62
avg loss pct0.68
stress return pct-56.88
quarter Kelly0
open0
pending0
actual allocation0
QGAP91008 15-session block paths and quarter Kelly imply zero size
1. Seed91008 generates20,000 historical resamples using five-session circular blocks of shared daily ETF portfolio returns. Horizon15 NYSE sessions matches Sep10–30, not21 trading days. Each path assumes187.22 in hypothetical daily1/8 sleeves, preserving cash for absent signals; these are not PDT/whole-share feasible portfolios.
2. Base P05/P50/P95 profit dollars [−3.481335,−0.445873,+2.013570], loss66.22%. Stress [−4.757233,−1.327922,+0.892286], loss87.655%. No sampled path reaches349.01; this is not a proof of impossibility or a tail bound.
3. Fixed trade p=43.9795%, avgwin0.617582%, avgloss0.682769%, so quarter Kelly clipped at0. Stressed p31.7992%, avgwin0.614577%, avgloss0.741765%, also0. Expected loss per hypothetical187.22 trade is0.20759 base/0.58124 stress.
Conclusion: actual size0. Fund cap187.22, account position cap212.7475 and nominal1% risk8.5099 are ceilings, not reasons to deploy. Stationarity, spreads and future signal rates are assumptions.
Source: https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min .
paths20000
sessions15
block sessions5
base median USD-0.45
stress p05 USD-4.76
stress median USD-1.33
stress p95 USD0.89
stress loss probability0.88
quarter Kelly0
fund cap187.22
nominal risk cap8.51
actual allocation0
QGAP91008 trend, affordability and liquidity do not repair the signal
1. Fixed holdout means above/below prior SMA200 are −0.115755% (N1380) and −0.102155% (N771), respectively; stressed means −0.315324% and −0.301752%. These descriptive slices do not supply a profitable new filter.
2. September9 closes put SPY, QQQ, IWM, EFA, EEM and XLF above SMA200; TLT and GLD below. September10 opening gaps remain unknown until the actual session. Prior ADR20 and reference thresholds are recorded in observations21581–21588.
3. One whole share is affordable within187.22 only for EFA106.56, EEM68.48, TLT81.73 and XLF57.06. None fits a187.22/8 equal whole-share sleeve. Lowest median20-day dollar volume is EFA1.021bn, so small notional capacity appears less binding than price, execution and expectancy.
4. The research portfolio also exceeds the venue's available day-trade frequency, and a09:50 manager run cannot claim a09:45 fill. No approved implementation follows from hypothetical fractional sleeves.
Source: https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all .
above trend sample1380
below trend sample771
above mean pct-0.12
below mean pct-0.1
above SMA200 now6
affordable standalone ETFs4
minimum median dollar volume1020876408.6
equal sleeve dollars23.4
actual allocation0
QGAP91008 sampled exits consume the small gross reversal
1. The rule observes the 09:30 opening gap, enters at 09:45, then checks prices every 15 minutes; it exits by 15:45. Stop=1 prior ADR, target=1.5 ADR; take half at +1 ADR and apply breakeven at the following check. Actual bar prints remain proxies for executable quotes.
2. Holdout exits: 262 stops, 71 targets, 1,818 time exits; 205 trades took a partial profit. Worst net loss was 2.4554 nominal R. A 1R trigger therefore cannot establish a guaranteed 1% account loss bound.
3. Daily open-to-close proxy mean was +0.026150%; sampled managed gross mean only +0.009058%, versus assumed round-trip cost about 0.12%. Net average win/loss is 0.617582%/0.682769%, below a 1.5 reward/risk gate.
4. All eight daily calendars are complete. Ten incomplete symbol-sessions were excluded, all in training. We retained 1,305 opening discrepancies above 1bp; IWM March16 2020 daily open100.29 versus intraday97.38 differed2.9016%. No invented fills or explanation for that mismatch.
Source: https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min .
1. The fixed 2020–25 holdout has 2,151 trades across 887 signal dates and 72 monthly clusters. Mean net return is −0.110880% with 95% month-cluster bootstrap CI [−0.165928%, −0.052452%]; stressed mean −0.310459%, CI [−0.365700%, −0.251403%]. Resampling 20,000 times does not create new observations.
2. Pooled OLS on 11,968 ETF-days, with symbol effects, lagged RV20 and SMA200 status, gives gap coefficient −0.009604 percentage points per ADR, day-cluster SE 0.024997, t=−0.384, R²=0.005536. This is association, not evidence of causal order flow.
3. Strong counterargument: gross expectancy is slightly positive. Removing the constant assumed cost factor transforms the mean CI to [−0.046055%, +0.067557%], still spanning zero. All six portfolio holdout cells lose; QQQ's one-ADR neighbor gains 5.281% at base cost but loses 4.928% stressed (N=51).
Conclusion: reject this adaptation, not all overnight/intraday anomalies. Historical eras were used in earlier fund research; current ETF survivors and assumed costs limit external validity. Source: https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min .
sample2151
monthly clusters72
mean pct-0.11
CI low pct-0.17
CI high pct-0.05
stress mean pct-0.31
regression rows11968
day clusters1496
gap t-0.38
R20.01
gross CI low pct-0.05
gross CI high pct0.07
QGAP91008 neighbor gap1.0 recent cost16bp per side