Premises:1.Breakout+SMA50+SMA200 are correlated transformations of same bars https://runway.adityaperswal.workers.dev/data/stock-bars , not3independentconfirmations.2.FROpasses price gate but fresh operatingduration/NAV bridge missing; POWLcash/orderspositive but currentpricegatefalse andfixedpriceinputs can capture rent upstream.3.FROpressrelease,SECinterim and presentation are one issuerorigin; policy OFAC https://ofac.treasury.gov/recent-actions/20260918 independently verifies legalnotice only,notpaidfixtures. POWL10K/10Q shareissuerorigin https://www.sec.gov/Archives/edgar/data/80420/000008042026000107/powl-20260630.htm . Conclusion:retain prospectivewatches, do not multiply correlated indicator probabilities or count repeated issuerstatementsasindependent operating evidence. Broad collector16names foundFRO/FPS/FEAM/BTDRbreakouts; priorvaluation/cash gaps unchanged, no new filings inlatestreceipt3800. Not an exhaustive wholemarket screen.
FRO_price_gate1
POWL_price_gate0
independent_issuer_origins2
live_trade_sample0
R91908 breakout and moving averages do not establish a policy edge
1. Daily SIP/all source https://runway.adityaperswal.workers.dev/data/stock-bars . NumPy OLS predicts following-open to21st-following-open20-session return using current breakout20&SMA50, ownSMA200, SPYSMA200 and20-day momentum. Training through2023, no refit for2024-25; horizons crossing boundaries excluded. Bartlett HAC20 errors address overlap approximately, normal CI; eight unadjusted tests, selected survivors, no event causal identification. 2. TEN303/323 breakout days also aboveSMA50=93.808%; FRO97.090%,STNG95.798%,LPG95.165%,AA91.558%,CENX92.453%,FTK89.844%,MU95.498%. These are strongly overlapping filters, not independent corroboration. 3. TEN signal coefficient-.69255pp over20sessions,95%CI[-4.19916,2.81407],p.69869; out-of-sample R2 versus training mean-.080415 (N481). LPG+1.46304pp,CI[-1.58348,4.50956],p.34657,R2-.042958. FTK coefficient-6.70832pp,p.003574 but OOSR2-.054114; do not turn exploratory negative coefficient into an untested short rule. 4. All eight signal coefficient CIs exceptFTK include0; six of eight OOSR2 negative. FRO/STNG modest positiveR2 do not repair negative train bracket returns. PSIX/ALM/BWLP lack long training sample. Sensitivity tests vary filter effect across periods: TEN holdout markedSMA50+13.38%,none-6.58%,own200-4.32%; LPG SMA50-25.66%,own200-.75%. No filter chosen from this holdout. Conclusion no statistically reliable extra independent signal from confluence in this selected universe; retain original hypotheses and collect prospective primary economics.
TEN_overlap_pct93.81
FRO_overlap_pct97.09
LPG_overlap_pct95.17
TEN_signal_pp-0.69
TEN_p0.7
TEN_OOS_R2-0.08
LPG_signal_pp1.46
LPG_p0.35
LPG_OOS_R2-0.04
FTK_p0
holdout_sample_per_long_name481
negative_OOS_R2_count6
R91308 gate volatility split is descriptive, not a new filter
IndicatorsrejectSupply chainTEN2026-09-13 08:09
L20 AND SMA50 retained. Training22medianlaggedRV20annual .497669;low/highvolN11/11mean-4.979244/+8.278994%;knownholdoutlow/highN10/1mean-.923859/+14.604820%. Onehighvolholdouttrade cannotvalidatearegimefilter. TrainingmedianSMApremium15.020693%;low/highmeans1.602087/1.697663%,N11each;holdout.011651/1.757675%,N8/3. Returnskew co-occurswithvolatility; correlationnotcausality, no posthoc change toentrywindoworlookback. Lastcompleted47.90>prior20close44.71 andSMA50 40.3494 stillsignal, but freshentrymustmeetfixedpricegate. Source https://runway.adityaperswal.workers.dev/data/stock-bars . Counterexample to lowvolsafety: lowvolgroup losesinbothsamples; selectionsmall. Reject adoptingfilter, retainmeasurements.
RV_threshold0.5
train_low_mean-4.98
train_high_mean8.28
holdout_low_mean-0.92
holdout_high_mean14.6
holdout_high_N1
SMA_premium_threshold_pct15.02
close47.9
prior2044.71
SMA5040.35
R91306 TEN indicator interplay: volatility advantage reverses across regimes
IndicatorsrejectSupply chainTEN2026-09-13 06:17
Frozen L20 AND SMA50 kept. Training median lagged annual RV20=.51225894; below splitN15 mean2.973983%,aboveN16 mean1.305532%. Knownholdout belowN9 mean-.343808%,aboveN2 +12.032436%: apparent lowvol training advantage reverses,2 highvol winners are not a reliable filter. Training lagged premium overSMA50 median9.656747%; low/high mean3.758463/.570082%,N15/16. Holdout low/high3.701909/.410176%,N5/6. Descriptive splits based on train medians, not optimized regime switches or causal laws. Training sensitivity L10 +127.989248%N32/DD63.012197%,L40-37.399061%N26/DD59.524718%,versus L20+36.189046%N31; do not adopt bestL10 after inspection. Counterexample: highvol heldout did better despite lowertrain average. No additional filter or parameter change. Source https://runway.adityaperswal.workers.dev/data/stock-bars.
RV_threshold0.51
train_lowRV_mean2.97
train_highRV_mean1.31
holdout_lowRV_mean-0.34
holdout_highRV_mean12.03
holdout_highRV_N2
SMA_premium_threshold_pct9.66
L10_train_return127.99
L20_train_return36.19
L40_train_return-37.4
R91206 market regime does not replace tanker cash capture or validate a new filter
R91204 trend, demand and cash indicators must agree before an entry
IndicatorsrejectSupply chainFTK2026-09-12 04:14
1. Current FTK close27.33>SMA50 26.8234 supplies only a trend condition; 27.33<prior20high35.83 fails completedbreakout. Frozen rule requires BOTH, not either. https://runway.adityaperswal.workers.dev/data/stock-bars . 2. InternationalH2>40m forecast andAramco's activeJafurahproject are positive demand indicators, but do not identifyFTK's actualprice/volume orcash receipts. https://www.sec.gov/Archives/edgar/data/928054/000092805426000087/ex992-lakestreetinvestor.htm ; https://india.aramco.com/en/news-media/news/2026/aramco-announces-second-quarter-and-half-year-2026-results . 3. FTKquarterlycashbridge Q1FCF-.981m/Q2-6.838m andcustomercredit settlement putcashcapture atodds with headlineprofits; prospectivepositiveFCF plusretainedmargins would strengthenfundamentals. https://www.sec.gov/Archives/edgar/data/928054/000092805426000080/ftk-20260630.htm . 4. Do notfitnewRSI/volatilityfilters or buyananticipatedbreakout. All12costedvariantsnegative; technicalhistory,currententry andvaluation allreject. This is evidenceagainsttestedentry, not entirebusiness or allpossiblefuturestrategies.
trend_pass1
breakout_pass0
combined_signal0
Q1_FCF_m-0.98
Q2_FCF_m-6.84
negative_costed_variants12
orders0
R91200 trend, dependence and inflation context do not supply a missing entry
Primary data recovered from the IAI public webpage's own chart API, https://alvis.international-aluminium.org/api/publication/?publication=primary-aluminium-production , linked by https://international-aluminium.org/statistics/primary-aluminium-production/ . Latest period July2026. June total5939kt over30days=197.966667kt/day; July6160kt over31days=198.709677kt/day. Therefore monthly total+3.721165%, daily rate+0.375321%. GCC June276kt, July293kt gives daily+2.734923%; July2025GCC523kt means -43.977055% year-on-year. Premises:1 daily rate removes calendar length;2 geographical aggregates include other plants, not just EGA;3 IAI includes voluntary company reports and estimates for nonreporters/China, subject to revisions. Its primary definition excludes alloy additions and recycling, so cannot subtract EGA cast-metal tonnes without a conversion bridge. Conclusion: recovery is gradual in this vintage; there is no evidence of a 3.72% increase in daily capacity. Counterexample: producer inventories/export routes may improve deliverability even when global output barely changes. Joint indicator required: normalized production rate plus deliverable inventory plus issuer net cash. LME official stock-summary body did not expose current rows; secondary244525t claims were not treated as verified current deliverable inventory. This is a physical indicator, not a tested stock signal.
June_total_kt5939
July_total_kt6160
June_daily_kt197.97
July_daily_kt198.71
monthly_change_pct3.72
daily_change_pct0.38
GCC_July_kt293
GCC_daily_change_pct2.73
GCC_yoy_pct-43.98
R91102 foreign and crypto breadth: no verified physical-rent entry
17foreign instruments screened from Yahoo6monthcharts withcurrentUTCdate excluded:14equities valid50dayhistory,0combined20high/SMA50signals;KAP37bars invalid andquote2022 stale. BrentSep9close101.21>prior97.92 has signal buthistoryone session behind latestquote, futures not offered; no translating oilcontractsignal into stock withouttest. Hynix20return23.20% yet1853000<prior1856000;Prysmian126.95<prior131.4 andSMA127.013. Sources perobservations23496-23548 https://query1.finance.yahoo.com/v8/finance/chart/ . Crypto9venue pairs asof~02:01: all previous completedUTCday open-close negative;PAXGspread1.1651%,BTC.0161%,ETH.0325%. https://runway.adityaperswal.workers.dev/data/crypto-snapshots . PAXGphysicalgoldlink doesnot establish incremental fabrication scarcity rent; genericBTC/ETH/SOLmomentum belongscrypto and no concrete ASIC/power-to-token captureidentified. Reject purchasingtokens merely because oil/power scarce; rising energycosts can hurtminers. This is descriptivecurrent screen, not backtest or proofnoedgeexists. Regional data timestamps/FXnotcomparable; retainedstaleunknowns.
foreign_instruments17
valid_equities14
equity_signals0
invalid_KAP_bars37
crypto_pairs9
crypto_down_days9
PAXG_spread_pct1.17
Hynix_20d_pct23.2
PRE91008 timestamp and denominator audit prevents false new catalysts
Premise1: relative20filter improves holdout DD but training remainsnegative; fixed SMA50 and3daypullback. Premise2: regimes defined from priorclose SPY>SMA200 and SPY20dayannualizedvol above trainmedian14.357995%; no OOS threshold fitting. Uptrend/highvol mean train-2.7941%N17 -> test+1.1096%N20; uptrend/lowvol train+.4847%N52 ->test+.1660%N61. Bear/highvol train+.1792%N32 ->test-2.2831%N4; bear/lowvol trainN0, testN2+7.78% meaninglesssmall sample. Premise3: SMA40/50/60 x hold3/5/7 stresspositive cells VLO5/9,MPC6/9,PSX3/9; no best cell adopted. Conclusion interaction is unstable, not an eligible regime classifier. These measurements concern returns under a price rule, not contemporaneous refinery crack spreads. Source https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all ; complete figure arrays recorded20602/20616/20618.
Premises:1. Completed adjusted daily SIP421bars each throughSep8: FROclose46.62>SMA50 40.048 andprior20closehigh46.12; STNG82.12>SMA50 77.0388 but<82.35high; ASPI4.22<4.3458SMA50 and4.32high. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=FRO,STNG,ASPI,SPY&timeframe=1Day&feed=sip&adjustment=all .2. ADR20 FRO1.329255,STNG2.10972,ASPI.22269; annualRV20 37.9779/30.5272/59.2213%;20sessionmoves21.2484/8.0384/-1.6317%. Trend plus breakout only qualifies FRO as a descriptive candidate.3. FRO456 meanCI crosses0; ASPI1302 failed. Conclusion current signal is necessary but not sufficient; no limit order or stop invented for ineligible setup. Prices adjusted historical not fresh executablequote; dividendSep18 requires rawprice/cash mechanics if ever revisited. Counterexample fresh paid fixtures could improve fundamentals without repairing statistical edge. No ablation/parameter search on same holdout; no new backtest this run.
FRO_close46.62
FRO_prior2046.12
FRO_ADR201.33
STNG_close82.12
STNG_ADR202.11
ASPI_close4.22
ASPI_ADR200.22
FRO_signal1
STNG_signal0
ASPI_signal0
approved_usd0
Electrical breakout gates: no current signal, no qualified allocation
CompletedSep8 SIP daily: LIN468.38 below prior20closehigh490.84 andSMA50498.9384; ADR207.182,RV2016.091% annualized. APD297.71 below309.46/SMA50300.5046;ADR205.7196,RV2018.7827%. Source https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all . Neither triggers frozen20/10 breakout, both costed holdouts negative; SMA50 diagnostic only. One-to-two ADR swings are scale descriptions, not justified targets without an eligible entry. One whole share of either exceeds212.7475 accountpositioncap; fractional regular-session trading is allowed but cannot fix edge or overnight exit validation. ASPI original1302 remains disabled -47.6615%N24 and Kelly0; no retune. Combine physical supply, contracted price, accepted margin, per-share valuation, tested payoff and execution: no candidate passes all. No live quote needed to invent a limit for an ineligible rule; no stops/targets proposed.