Stats

A news scan found the report from EGA on August 12, which stated that as of Monday, August 10, there were 227 cells. Another report from EGA on August 26 stated that there were 315 cells out of 1262, which is roughly a one week stabilization. The implied interval speed is 88 cells divided by 16 days, which equals 5.5 cells per day. With 947 remaining cells, it would take approximately 172.18 days to reach the target, plus 7 days for stabilization, which would be February 21, 2027. This rough estimate is consistent with, but not independent confirmation of, the issuer's Q1 target. The assumptions made include that all 315 existing cells are stable, have identical capacity, and that the cast output is a proxy of 1.6 million tonnes per year. A simulation using NumPy with 50,000 paths and a fixed speed per path of 3 to 5.5 to 8 cells per day was run. The completion days after August 26 were estimated to be 138.35 days at the 5th percentile, 178.92 days at the 50th percentile, and 256.53 days at the 95th percentile. The model estimated that there is an 81.662% chance of completion before March 31, but this is not a real-world forecast odds. The integrated missing cast output proxy through March 31 was estimated to be 237,423 tonnes at the 5th percentile, 304,145 tonnes at the 50th percentile, and 421,743 tonnes at the 95th percentile. However, the precision of delivered tonnes is lower due to factors such as heterogeneous cells, hot-metal reuse, alloy content, existing stocks, and exports. In conclusion, scenarios should be used to bound questions for the next operational update, but should not be used to derive a stock target or trade probability. New measured output and net shipped volumes are more important than this simulation.

A news scan found the text about automated trading fund manager notes. The manager uses training payoffs, not favorable holdout, from fixed tests 2521–2522, with data from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. For CSTM, the training payoff is 40%, with a mean win of 8.4956% and a mean loss of 7.6217%, resulting in a clamped half Kelly of 0. For KALU, the training payoff is 41.9355%, with a mean win of 10.7182% and a mean loss of 7.6758%, giving a rough half Kelly of 0.176252% of equity. This is a binary-payoff approximation and a sizing ceiling, not proof of optimal allocation. The manager notes that stress costs make KALU training mean negative and half Kelly 0. Both CSTM and KALU currently fail entry, so the actual allocation is 0. The manager also notes that whole-share KALU 158.42 and stock minimum 1 exceed the hypothetical probe of $0.375, and therefore should not be rounded up. The manager ran simulations using NumPy with 50,000 paths of 20 independent and identically distributed training roundtrips, with a seed of 91104. The results show that allocating 25% per trade yields a median account return of -6.56% for CSTM, with a loss probability of 73.34% and a 95th percentile drawdown of 22.50%. For KALU, the median account return is -0.98%, with a loss probability of 53.40% and a 95th percentile drawdown of 19.89%. The manager notes that these are conditional empirical simulations, not calibrated future loss bounds, and omit unseen gap and persistent regime risk. The historical training target-hit frequencies are 2/30 for CSTM and 2/31 for KALU, versus 15/30 and 18/31 for stops. The conclusion is that the venue maximum is not a position recommendation, and no funded probe is justified by these distributions. A funding shortfall does not change the odds, target, or size.

The documented reversal mechanism compensates liquidity suppliers, not all buyers of falling prices. Sector aggregation mixes stock-specific shocks with shared cash flow news. ETF share/NAV arbitrage requires capacity and creation units, and daily fundamental prices can continue to move. There are three consequence chains to consider. Chain A assumes a liquidity shock: first, a forced seller depresses the price; second, funded intermediaries absorb inventory/arbitrage ETF basket; third, a temporary discount may close before the next open, leaving a smaller edge for the cash-only buyer. Chain B assumes a cash flow shock: first, sector revenue/margin outlook falls; second, earnings revisions and fund outflows reinforce price change; third, fading losers exposes funds to persistent losses and gaps through stop. Chain C considers crowding: first, high volatility raises prospective liquidity compensation; second, other liquidity suppliers withdraw or face funding limits; third, surviving providers may earn more, but poorly funded retail positions can exit before the payoff. Who benefits depends on the cause: informed sellers avoid permanent loss; liquid market makers capture temporary inventory premium; issuer fee revenue does not prove holder alpha. These cause sequences are hypotheses, not observed September 11 flows. A tested weekly selection did not outperform a matched sector basket and stress failed, so no order was made. Sources include research papers from the National Bureau of Economic Research and the New York Fed, as well as information from State Street Global Advisors.

A news scan found the analysis of the XLI ETF. The analysis is based on the following premises: the XLI issuer's characteristics on September 9, which include a price-to-book ratio of 6.63, a forward one-year price-to-earnings ratio of 24.59, and an index price-to-cash flow ratio of 19.05, with 83 holdings. These aggregate numbers do not measure an individual forced seller or incremental cash flow revision. The net asset value of 171.75 versus the closing price of 171.79 shows a premium of 0.03%, with a median 30-day spread of 0.01%, and an annual expense of 0.08% embedded in the ETF's returns. Secondary shares are not individually redeemable, but a creation unit mechanism is available to specialist participants, with discounts possible in stress. The conclusion is that cheap execution and affordability do not establish asset undervaluation, a net asset value arbitrage, or reversal probability, and no fundamental price target can be inferred from the price-to-earnings ratio. The issuer collects the expense base, intermediaries may capture creation and redemption pricing gaps, and the retail buyer receives sector cash flow exposure and risk. The strongest counterargument is that true liquidity shocks can be independent of fundamentals, and dated flow or inventory evidence or a validated residual signal is needed to identify them. No issuer 10-K cash forecast was invented for an ETF. The source of this information is https://www.ssga.com/us/en/intermediary/etfs/state-street-industrial-select-sector-spdr-etf-xli, with observation 24106.

A news scan found the text: Premises: 1. Sample primary 5 stress holdout daily cash portfolio, not best neighbor, as recorded. 2. Seed 9110400 draws 10,000 paths from overlapping 21-session blocks, 12 blocks = 252 sessions. This method preserves dependence within each block but misses new regimes, long memory, and unseen fat-tail gaps; it does not forecast. 3. Applying the full $187.22 USD fund gives 21 session P5/median/P95 PnL - $13.13308 / - $1.058883 / + $14.38771 USD, with a 55.84% loss probability; zero empirical paths gain $349.01 USD, which is not proof of impossibility. The 252 session return P5/median/P95 is -25.72362 / -3.98388 / +24.46864%, with a drawdown median/P95 of 17.16606 / 30.95414%. 4. The half-Kelly formula, which is a betting strategy, is 0.5 * (p - (1-p) / (avg win / avg loss)), clipped at 0. The primary stress p is 44.3137%, and the win/loss ratio is 1.22364, resulting in a negative unclipped Kelly, which is set to 0. The training stress is also 0. The baseline plugin of 3.146% does not override the cost failure. The account normal 3% risk is $25.5297 USD, with a position ceiling of $212.7475 and a fund of $187.22; all additional caps are in place, and there is no entitlement to maximum size. A full fund gap loss would be 22% of the account, so the fund cap alone is not a gap budget. The conclusion is to size at zero; funding need does not determine barriers. A counterargument that resampling misses future bonanza is accepted, but there is no independent positive edge that justifies paying for it with this rule. The source is https://runway.adityaperswal.workers.dev/data/stock-bars and research/inference.py/json.

The automated trading fund manager analyzed the REV9110400 payoff shape, partial exits, and current weekly eligibility. The premises of the analysis are based on 255 trades with a win rate of 44.3137%, a mean win of 2.30310%, and a mean loss of 1.88218%, resulting in a realized payoff of 1.22364, despite a nominal 3R barrier. The analysis found that only 9 trades exited due to target or gap target, while 123 trades exited due to time, 106 trades stopped, and 17 trades had gap stops. Additionally, 35 trades had quarter partial exits. The mean holding period was 3.6784 sessions, with 30 same-day final exits and 31 first-day partial or final exits. The median trade was -0.91748%, with the worst trade being -5.56926% and the best trade being +8.43914%. The maximum loss streak was 7. The signals used prior complete bars. For the current week, the model candidate is XLY, with a stress entry at 114.641705 and a prior ATR of 1.486295. The stop is at 113.155410 and the target is at 119.100590. However, these are hypothetical historical levels and not new orders or prospective success. The current incomplete week is excluded from scored backtests. On September 10, the ranks were XLV -4.21509%, XLB -4.13598%, and XLY -2.52481%, with closes at 165.66, 50.76, and 111.96, respectively. The SPY 20-day volatility is 8.53648% below the 20% gate. There will be no midweek catchup, and the next week's rank requires a September 11 close. A $187.22 buy can purchase at most 1 XLV or 1 XLY or 3 XLB before fees. The quarterly fractional partial cannot preserve whole share after hour lot. The ADV 20 USD for all nine is greater than 571 million, and small notional impact is not binding. However, spreads, 15-minute polls, quarter exits, and account PDT shared with other funds are. The conclusion is that there will be no live entry, as the frequency-limited weekly still has entry day exit risk, and the exact 15-minute exit policy and integer sizing are untested. The strongest counterargument is that 9 tail wins may pay, but recorded costs and mean expectancy reject that claim for this rule. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the research and inference can be found in research/inference.json.

A news scan found the body of a research paper: Frozen2177 applies five separately normalized lag1..5 negative cross-sectional-return weights to nine ETFs and averages them. Individual overlays are long1/short1; offsets make aggregate gross less than 2, correcting shorthand in2177, with no renormalization. Cash-only control clips aggregate negatives and normalizes positives to 1. Next-open returns use weights known previous close; same-close reference is explicitly non-executable. Findings: long-short next open holdout gross +3.38%, mean 0.381bp/day; one-way turnover break-even 0.6897bp, versus 6bp baseline yielding -35.60% and 16bp -70.75%. Training gross +22.55%, 6bp -18.91%, break-even 3.164bp. Long-only baseline holdout +20.48% but stress -51.74%; this positive baseline retains market risk and does not validate short-term selection. High vol condition does not rescue holdout: long-short next open 6bp mean -4.61bp/day, long only -5.29. These are identification calculations with linear turnover costs, no borrow/financing/terminal-liquidation charge and no actual short order; do not rank as executable backtests. They are not the weekly ATR/3R portfolio, which separately reconciles with vectorbt. Daily industry results need not replicate monthly within-stock results. All 18 descriptive cells recorded; no parameter promoted. Source research/diagnostic.py/json and https://runway.adityaperswal.workers.dev/data/stock-bars; methodological primary https://www.nber.org/papers/w17653.

A news scan found the text of a report from an automated trading fund manager. The report discusses three premises. Premise 1 uses a 20-day high and 10-day low, with a 14-day average true range, or ATR, which is a measure of volatility, for risk management. It also has a 3R target, which refers to a three times risk reward ratio, and credits a 25 basis point, or 0.25%, asset fee, with a 10 basis point, or 0.10%, slip, and a 20-day timeout. The signal is known after the bar is completed, and the next open is exact. There is a 2R cost basis, or break-even point, and a peak minus 2 entry ATR trail, with a retained target. A 20 basis point slip stress leaves the target fixed. The report uses a native calendar-hourly timeframe, which is based on the actual calendar hours, and a rolling-hourly timeframe, which looks back 480 hours. Premise 2 uses a chronological train from 2024 to 2025, and a reused test from January 1, 2026, to September 11, 2026, with a reset to flat, or zero, at the start of the test. The report also discusses hourly and daily open sampled exits, with no guarantee of stop price, and raw wealth reconciliation using vectorbt, a backtesting library, with a precision of 1e-10. The full native calendar-hourly train has a return of 21.988671%, with 12 trades, and the holdout, or test, period has a return of -10.135134%, with a maximum drawdown of 35.033237%, and a win rate of 28.571429%, with 7 trades. The same-target stress test has a return of -11.384479%. Premise 3 discusses attribution, which refers to the process of assigning returns to specific factors, with a ladder of full native daily and daily, daily and hourly, and calendar and hourly timeframes. The report also discusses matched-full 24-hour daily and daily, and rolling 480-hourly timeframes. The matching process drops the incomplete hour day on March 12, and suppresses later entries, which is a selection effect, not a price alpha, or return, effect. The native versus aggregate matched 2026 returns coincide, but the source construction is not the explanation for the ladder. The report concludes that positive matched variants cannot replace previously frozen rejected policies, and there is no switching based on this reused sample. The current 20-day entry is absent. The validity of the report is based on controlled changes that attribute differences within implementation, and the soundness is limited by tiny reused samples, historical bar anomalies, and sampled quotes. The report uses 80 cells per asset, with 168 matched and 72 full native, for a total of 240, but these are not 80 independent trials. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the method notes are 2175/2180.

A news scan found the text of a report from an automated trading fund manager. The report discusses the performance of a trading strategy, called Premise1, which uses a 20-day high and 10-day low, with a 14-day average true range, or ATR, for risk management, and a net target of 3R, with a 25 basis point credited asset fee and a 10 basis point slip. The strategy has a 20-day timeout and a signal is known after a completed bar, with the next exact open. The report also mentions a cost basis, or BE, of +2R and a peak entry ATR trail, with a retained target. A 20 basis point slip stress leaves the target fixed. The report uses a native calendar-hourly timeframe, with 20 completed UTC days, and a rolling-hourly prior 480 hours. The report is based on two premises, or assumptions, with the first being a chronological train from 2024 to 2025, and a reused holdout from 2026 January 1 to September 11, at 00 UTC, with a reset to flat. The report also discusses the performance of the strategy, with a 41.228668% return on the full native calendar-hourly train, and a 7.324178% return on the holdout, with a 42.857143% win rate and a 19.830197% drawdown. The report also mentions a same-target stress test, with a 5.832105% return. The report breaks down the attribution of the strategy, with a full native calendar-hourly train, and a daily and hourly open sampled exit, with no guarantee of stop price. The report also mentions a raw wealth reconciliation, using a vectorbt 1e-10. The report concludes that the validity of the results is due to controlled changes, and that the soundness of the results is limited by tiny reused samples, historical bar anomalies, and sampled quotes. The report is based on 80 cells per asset, with 168 matched and 72 full native, for a total of 240, but notes that these are not 80 independent trials. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the method notes are 2175/2180.

A news scan found the text describing an automated trading fund manager's notes. The notes discuss a trading strategy for Bitcoin (BTC) with specific rules. The strategy has a 20-day high and 10-day low premise, with a 14-day average true range (ATR) risk, which is a measure of price volatility. The target return is 3 times the risk, with a 25 basis point (0.25%) asset fee and a 10 basis point (0.10%) slip, which is the difference between the expected and actual price. The strategy also has a 20-day timeout. The signal to buy or sell is known after the bar is completed, and the next exact open price is used. The strategy also includes a 2 times risk cost and a peak minus 2 times ATR trail, which is a way to adjust the stop-loss price. The notes also discuss the training and testing of the strategy using historical data from 2024-25 and 2026. The results show a return of 30.44% for the training period and 2.08% for the testing period, with a win rate of 42.86%. The notes also discuss the attribution of the returns to different factors, including the native calendar-hourly and daily returns. The conclusion is to watch the strategy, but not to switch to it based on the results. The validity of the results is limited by the small sample size and historical bar anomalies. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the method notes are numbered 2175/2180.

A preregistered note, number 2178, outlines a trading strategy for KALU, which involves closing above the prior 20-day close high and SMA50, with a next session open entry, stop at 2 times the prior 20-day average daily range, and a target of 6 times the average daily range, with a 3R ratio, and exiting on the 20th session close. The strategy was trained on data from 2020 to 2024 and tested on a holdout period from 2025 to September 10, 2026. The data source is from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. The results show a return of -14.05% during the training period, with a drawdown of 55.15%, and a return of 47.76% during the holdout period, with a drawdown of 28.95%. The stress test return was 42.82%. A four-trade circular block bootstrap test with 10,000 samples showed that the mean confidence interval crosses zero. The conclusion is that the positive holdout results alone cannot fix the negative compounded training results or support retrospective sizing, and therefore, it is a research watch only, with no current signal and no defensible new allocation.

A preregistered note, number 2178, outlined a trading strategy for the CSTM stock, which is a type of aluminum-chain benchmark. The strategy was to buy the stock if its price closed above the highest price of the previous 20 days and the 50-day moving average, and then sell it when it reached a target price or after 20 days. The strategy was tested using data from 2020 to 2024, and then its performance was evaluated using data from 2025 to September 10, 2026. The results showed that the strategy had a return of -38.61% during the training period and 31.78% during the holdout period. The holdout period is a time when the strategy is tested on new data to see how well it performs. The stress test return was 27.89%. The results also showed that the strategy had a high drawdown, which is the largest decline in value from a peak to a trough, of 47.12% during the training period and 19.56% during the holdout period. A drawdown is a measure of risk. The strategy was also tested using a bootstrap method, which is a statistical technique, and the results showed that the mean confidence interval crossed zero. The conclusion is that the strategy is only for research purposes and does not currently have a signal to buy or sell the stock, and the positive results during the holdout period are not enough to make up for the negative results during the training period.

The consequences of BB's earnings depend on the mix of licenses and production conversion. A test was done on September 11, with the stock price at $7.54 for BB. By October 2, the close price for a call option was $0.40, which requires an increase of 11.405836% to be profitable, and this covers 3 out of 12 months, or 25%. On the other hand, the close price for a put option was $0.32, which requires a decrease of 11.405836% to be profitable, and this covers 2 out of 12 months, or 16.67%. Over 12 quarters from September 2023 to June 2026, the median absolute change in stock price from before to after news was 9.987505%. The first 8 quarters had 2 types of options, and the last 4 quarters had 1 type of option. Neither the 40% gate nor the expected option payout passed. The option expires on October 2, 8 days after an event on September 24. The default exit is on September 23, but the hurdle exceeds the median by 1.418331 percentage points, so there is no exception to hold through. The sources of this information include event observations from the issuer and websites with stock and option data. The starting point is a profitable software supplier with uneven cash generation and a stock price below the 20 and 50-day averages. The goal is to find an affordable option with a high payoff. The main constraint is the current directional coverage, not the conviction in the turnaround. There are three orders of factors that can improve reported revenue: stronger development licenses and deliveries, automaker production and renewals, and sustained receipts. However, there is also an adverse chain of events that can weaken receipts and compress cash margins. These mechanisms are conditional and not forecasts for other companies. One counterargument is that certification and customer relationships can support durable economics, and current filings support this statement. The primary mechanism is explained in a filing with the Securities and Exchange Commission. There is no independent confirmation of customer volume or measured options expectancy, and no proposal from an outside fund.

A test was done on September 11, with the stock price of BlackBerry, or BB, at 7.54. By October 2, the price to reach 8.40 was 0.40, which requires an increase of 11.405836%. This covers 3 out of 12, or 25%, of the time. On the other hand, the price to reach 7.32 was 0.32, which requires a decrease of 11.405836%, covering 2 out of 12, or 16.67%, of the time. Over 12 quarters, from September 2023 to June 2026, the median absolute change in price was 9.987505%. The first 8 quarters had 2 calls and 1 put, while the last 4 quarters had 1 call and 1 put. Neither the 40% gate nor the 3 times expected option payout was passed. The October 2 expiry was 8 days after the September 24 event. The default exit was on September 23, but the hurdle exceeded the median by 1.418331 percentage points, so there was no exception to hold through. The sources of this information include event observations 24113-24126, as well as data from investors.blackberry.com and runway.adityaperswal.workers.dev. The issuer's Q1 baseline and Q2 guidance are recorded in observations 24176 and 24178, and can be found on sec.gov. The current quarterly revenue growth is real, but much of the growth in secure communications came from SecuSUITE licenses, while retention remains below 100%. Stronger sales and weaker recurring retention can both be true. The cash generation after property and intangible spending was approximately 0.1 million in Q1, and achieving the annual operating cash guidance of 100 million requires 95.4 million over three quarters, averaging 31.8 million. However, seasonality is a counterargument, as the fourth quarter of FY26 delivered 45.6 million in operating cash flow out of 50.3 million for the full year. Using the May 31 share and cash balances with the September 10 stock price gives an equity value of 4.4189 billion and an enterprise value of 4.2070 billion, excluding restricted cash. This is 6.925 times the revenue guide midpoint and 32.612 times the adjusted EBITDA midpoint. This is not a current reconciled valuation or an earnings surprise forecast. The QNX 950 million royalty backlog is not comparable to the 133.9 million contracted RPO, as usage royalty exclusions differ. Customers benefit from certified software, but shareholders need conversion into collected cash. Full new-quarter receipts and incremental paid volumes would strengthen business confidence, but no arbitrary probability upgrade overrides the measured option gate.

A test was done on September 11 to see how the value of BlackBerry stock would change. The stock price was $7.54 on September 11 and the price of a call option to buy the stock at $8 on October 2 was $0.40. This option would be profitable if the stock price increased by 11.405836% to cover the cost of the option for 3 out of 12 months, which is 25% of the time. The price of a put option to sell the stock at $7 on October 2 was $0.32, which would be profitable if the stock price decreased by 11.405836% to cover the cost of the option for 2 out of 12 months, which is 16.67% of the time. The test looked at 12 quarterly releases of news from September 2023 to June 2026 and found that the median absolute change in stock price was 9.987505%. The test also found that neither the 40% gate nor the documented 3x expected option payout was met. The option expires on October 2, which is 8 days after a major event on September 24. The test used a European Black-Scholes model to value the options, which is a way of calculating the value of an option, and assumed that the interest rate and dividend were 0. The test also assumed different levels of volatility, which is a measure of how much the stock price is expected to change, and found that the mean gross payout for the call option was 0.7999 to 1.4841 times the premium, and for the put option was 0.2878 to 0.8956 times the premium. However, no scenario gave a 3x mean or a 4x single-event gross target. The test used 20,000 paths and a maximum of 20 whole-contract tickets, and found that the median final cash was $106.6357 and the fund drawdown was 46.2201% for the call option, and $16.6977 and 90.6565% for the put option. The strongest countercase is persistent high post-event volatility and pre-event rally, but this is not measured with historical option executions. A cheap-threshold scenario is recorded separately, but this is not permission to substitute today's larger premium.

A news scan found the headline about a pre-earnings stock proxy for BB that failed training despite a positive holdout. The edge test was frozen on September 11, with the stock price at $7.54 for BB. The test looked at two scenarios: one where the stock price would need to increase by 11.405836% to reach $0.40 by October 2, and another where the stock price would need to decrease by 11.405836% to reach $0.32 by October 2. The test covered 3 out of 12 periods, or 25%, in the first scenario, and 2 out of 12 periods, or 16.67%, in the second scenario. The test looked at 12 consecutive quarterly releases from September 2023 to June 2026, and found that the median absolute change in stock price from before to after the earnings release was 9.987505%. The test also found that neither a 40% gate nor a documented 3x expected option payout passed. The option expiry date was October 2, which is 8 days after the September 24 event. The default exit date was September 23, but the hurdle exceeded the median by 1.418331 percentage points, so there was no hold-through exception. The sources for this information include issuer-dated event observations 24113-24126, and data from https://investors.blackberry.com/ and https://runway.adityaperswal.workers.dev/data/stock-bars and /data/option-bars. The test used a vectorbt system with T-3/T-5/T-10 open entries and T-1 close exits, with 5 basis point fees plus 10 basis point slippage on each side. The test also raised slippage to 30 basis points to stress the system. The full $100 stock allocation is a diagnostic, and is outside the fund's instrument mandate. The test found that the training returns were -10.3639%, -2.73664%, and -20.6952% respectively, and selected the T-5 option by training return only. The selected holdout return was +2.23248%, with a daily drawdown of 9.40210% and 2 wins out of 4. The stress train return was -5.79981%, with a holdout return of +0.609767% and a drawdown of 9.76377%. The unselected T-10 holdout return was +23.2792%, but this cannot replace the negative training return. A separate required CLI momentum diagnostic found a return of +261.0066%, with a drawdown of 36.1282% and a win rate of 40.7407%, but this did not include a chronological holdout, so it was not adopted. The validity of the test is that an unsplit momentum winner cannot establish an earnings option edge, and the soundness of the test is that a daily time-only stock proxy omits option theta, volatility crush, spreads, and mandated option stops/targets/15-minute checks. All training losses plus failed event coverage leave the size at zero.

A news scan found the report titled REV9110400 market exposure and matched selection attribution. The report is based on two premises. First, all 20 weekly costed cells are reconciled to vectorbt, which is a backtesting framework, with a holdout period from September 10, 2021, to September 10, 2026, excluding the full current week. Second, the historical window is not globally untouched, meaning it has been adjusted in some way. The primary baseline holdout had a return of 34.9556%, a compound annual growth rate of 5.43259%, a market beta of 0.453798, and an annual simple ordinary least squares intercept of -0.051691%. The stress beta was 0.453772, the annual intercept was -8.00789%, and the t-value was -1.27233. Fixed-beta ordinary least squares, a statistical method, cannot eliminate all conditional exposures, which are risks that are dependent on certain conditions. On 220 training weekly opportunities, the time-only loser mean was 0.038393% versus an equal 9 basket mean of 0.192879%, resulting in a selection delta of -0.154486 percentage points. On 255 holdout weeks, the loser mean was 0.128452% versus an equal 9 mean of 0.172813% versus an SPY mean of 0.212909%, resulting in a delta of -0.044361 percentage points. The costs were identical at 6 basis points per side, and there is no claim of significant negative delta, just an absence of demonstrated incremental selection. The full-window uncosted SPY buy-hold return was 103.0918% in the training period and 121.2061% in the holdout period, versus an equal 9 return of 82.8131% and 113.6551%, respectively. These returns have different exposures and cannot substitute for a matched control. The conclusion is that the positive baseline total is compatible with equity exposure, but the signal-selection evidence fails. A counterexample is that nonlinear crash recovery might escape fixed beta, and a pre-frozen high-vol model was still stressed -7.2856% in the holdout period. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the research results are available in research/results.json, inference.json, and matched_train/test.csv.

A question was raised about the REV9110400 mechanism and whether the institutional reversal premium may not transfer. The starting point was an account with $187.22 in cash and $850.99 in total, with no open or pending trades. The goal was to achieve a repeatable return while keeping the economic mechanism affordable. Several premises were considered: a study by Nagel in 2011 used daily stock and industry portfolios to remove market exposure; a study by Da, Liu, and Schaumburg in 2011 separated stock-specific reversal from industry and cash-flow news; a study by AQR in 2012 found that institutional reversal was not profitable after costs, but this did not necessarily apply to small ETF accounts; and the issuer of the XLI ETF stated that creation-unit redemption and potential stress discounts meant that retail shares were not a free option for redeeming at net asset value. The inference was that adapting the mechanism to a long-only sector approach would remove diversification and hedges, and change the source of expected returns, so it needed to be tested independently. The strongest counterargument was that low ETF spreads and small size could make liquidity provision viable, which was tested with different scenarios. The contradiction was resolved by considering different populations, time horizons, cost definitions, and conditioning. The conclusion was that high-volatility industry reversal could coexist with monthly industry momentum. The sources used included studies from the National Bureau of Economic Research, the New York Fed, AQR, and the State Street Industrial Select Sector SPDR ETF website.

A news scan found the primary supply indicator for R91104. The indicator is based on data from the International Aluminium Institute's public webpage chart API. The latest data is from July 2026. In June, the total production was 5939 kilotonnes over 30 days, which is 197.966667 kilotonnes per day. In July, the total production was 6160 kilotonnes over 31 days, which is 198.709677 kilotonnes per day. This means the monthly total increased by 3.721165% and the daily rate increased by 0.375321%. The daily rate is a more accurate measure because it removes the effect of different calendar lengths. The data also shows that the daily rate in the GCC region increased by 2.734923% from June to July. However, compared to July 2025, the GCC production decreased by 43.977055%. The primary definition of production excludes alloy additions and recycling. The conclusion is that the recovery is gradual and there is no evidence of a 3.72% increase in daily capacity. Other factors, such as producer inventories and export routes, may improve deliverability even if global output barely changes. A joint indicator that includes normalized production rate, deliverable inventory, and issuer net cash is required. The London Metal Exchange official stock summary did not provide current data, and secondary claims of 244525 tonnes were not verified as current deliverable inventory. This is a physical indicator, not a tested stock signal.

The current state of the aluminum market is characterized by a supply fund that is flat, impaired physical production in the Gulf, and some competitors restarting their operations. The goal is to own paid scarcity rents at a favorable stock payoff. The main constraint is the net realized margin after considering energy, input metal, working capital, and recovery, rather than the global metal headline. The first issue is that interrupted primary production and export routes are constraining deliverable metal. For example, EGA's sales decreased by 31.61% in the first half of 2026, while their cast production decreased by 29.15%, and their operating cash flow decreased by 56.87% despite an increase in EBITDA of 10.85%, according to their report. This shows that operational and cash channels differ. The second issue is that companies like CENX, which has exposure to the US regional premium, can retain value, while others like LME-linked Iceland power and alumina, and Sebree market power, share value with suppliers. Fabricators like KALU and CSTM can pass metal costs through, benefit from scrap spreads, and scarce qualified rolling capacity, but they have to fund higher inventories, as seen in their reports. End customers face higher invoice prices or delays, but allocation and pass-through timing are not measured here. The third issue is that scrap suppliers can renegotiate, customer substitution can lower demand, and restored primary and rolling capacity can compress premiums. For instance, Novelis reports that Oswego restarted in June and Bay Minette is commissioning, yet their operating cash flow was -455m and adjusted free cash flow was -1.1bn in the June quarter. However, insurance timing gave them an 18m net EBITDA benefit from the outage. These are conditional consequences, not probabilities. A strong counterargument to pessimism is that CSTM's EBITDA rose by 87.88% and generated 90m quarterly free cash flow. On the other hand, a strong counterargument to bullishness is that restarts and cash needs can reverse that rent. In conclusion, it is recommended to keep separate primary-metal and conversion-spread watches within supply, and no unrelated fund is warranted.

The fund started with $49.14 in cash and $850.99 in the account, with no holdings or pending orders. The goal was to make a decision based on the FDA and earnings run-up, with a plan to exit before the event. There was a binding constraint on the net executable event payoff due to uncertain overlapping CBP clearance, which was not a monthly shortfall. The premises were: 1) the exact FDA date of September 28 was verified by the issuer on May 14, and only the date was considered for early decisions; 2) the ITC lift published on September 4 superseded the partial suspension from May, and included a full LEO/CDO with an XL drug combination, but the legacy base was unaffected and not all cash was safe; 3) cash reconstruction showed $2,485 pre-WC burn and $3.889 million in the first half of the year, with significant potential claims; and 4) frozen momentum from 2477 to 2480 showed a train of -96.87% and a holdout of -52.21%, with a stress of -64.46% and a half Kelly of 0. This negative control was not a backtest of the FDA event. The conclusion was to make no order and not move any levels, as the candidate had no independent net asymmetric run-up edge. The validity was limited by unmet entry conditions, and the soundness was limited by missing current CBP/appeal outcomes and post-restriction paid reorders. A strong bull case could be made with a positive FDA outcome, cheap redesign, and legacy utilization, but this would not assign a trade probability from the clinical 65.5%. The sources stopped changing the core answer after the September 1 update, and the absence of new SEC information was not pending proof. The prior cash constraint view was strengthened by the actual enforcement update, not blanket bearishness. The existing monitor was replaced on September 25, and the tested collector preserved all coverage. The next check will be at 06 UTC, and the fund will preserve zero unless there is a new primary ruling. There is no separate FUND case, as there is no validated outside-mandate operation. The sources for this information include the SEC and federal register websites, as well as a stock bars dataset.

A news scan found the following information about four aluminum-chain stocks. The stocks are AA, CENX, CSTM, and KALU. Here are the reasons why they failed the frozen current entry filter: 1. On September 10, the adjusted closing prices were AA $48.30, CENX $44.36, CSTM $25.99, and KALU $158.42, which were obtained from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. 2. Each stock's price is below its prior 20-day high, which were $51.85, $48.17, $29.92, and $189.49, and also below its 50-day Simple Moving Average, which is a calculation of the average price over 50 days, and were $48.5086, $45.6681, $28.4026, and $166.195. 3. The Average Daily Range over 20 days, which is a measure of price volatility, is 1.972495, 2.333405, 0.97824, and 7.1416, and the 20-day returns are -5.70%, -7.91%, -11.90%, and -16.40%. The annualized RV20, which is a measure of risk, is 45.88%, 42.93%, 39.99%, and 60.78%. These measurements are based on adjusted daily prices and are not executable after-hours quotes or stop fills. The conclusion is that there will be no breakout entry tonight, even if operating margins are favorable. This screen does not test for value or reversal entries and cannot authorize one. Stops and targets require an actual future entry and independently justified payoff, and no anticipatory limit is queued. The source observations contain every symbol's numbers.

A study of aluminum stocks found that after an outage, the stocks of different companies moved in different directions. The study used a method called SIP all-adjusted daily bars, which means it looked at the daily prices of the stocks, adjusted for things like dividends and stock splits, from January 2020 to September 10, 2026. The data came from a website. The study compared the daily returns of the stocks from January 2025 to March 27, 2026, with the returns from March 30 to September 10. The returns after the study period were: AA -17.01%, CENX -10.65%, CSTM +11.21%, and KALU +39.53%. The study also looked at something called SPY betas, which measure how much the stocks move compared to the overall market. The betas were: 1.444, 1.361, 1.969, and 1.842. The study found that none of the stocks had returns that were significantly higher than the market as a whole. The correlations between the stocks, which measure how much they move together, were: AA and CENX 0.673, and CSTM and KALU 0.655. This means that buying two aluminum stocks is partly like making one bet. The study found that the outage did not necessarily mean that the stocks of companies that make aluminum would go up. In fact, the stocks of companies that use aluminum went up while the stocks of companies that make aluminum went down. The study concludes that the direction of the outage should not be used to predict the direction of the stocks. The study had some limitations, including only looking at a few companies and a short period of time, which means that the results may not be repeatable.

A diagnostic test for BFRI911 was run using a momentum strategy with specific parameters, including a 20-day lookback period, 1460 days of data, a 5% stop loss, and a 15% target profit. The test used 1003 daily price bars from IEX and resulted in 124 closed trades. However, the raw prices showed a discontinuity due to a July 2023 split, which made them not comparable with the 1176 SIP adjustment used in the study. A full sample diagnostic was performed without a costed holdout, but it was not an event study and was not adopted. Additionally, an independent fixed rule long-only test with a costed chronological holdout from 2477 to 2480 also failed. The data source was https://runway.adityaperswal.workers.dev/data/stock-bars?feed=iex and local cli.json.

A clinical trial, Registry 187, randomized 187 adults, with 145 receiving the active treatment and 42 receiving a vehicle, or placebo. The primary composite result showed 65.5% versus 4.8%, a difference of 60.7 percentage points. The approximate unadjusted Wald 95% confidence interval was 50.62 to 70.78 percentage points, based on rounded rates. This trial did not have a prespecified inference or approval odds, and the population excluded aggressive, recurrent, or high-risk H-zone lesions. Serious treatment-emergent adverse events occurred in 3 out of 145 patients versus 2 out of 42 patients, but this is not a causal comparison. A trading holdout of 74 fixed momentum trades had a mean return of -0.70697% per event, with a 95% mean confidence interval of -2.40017% to 1.15124% based on 20,000 independent and identically distributed bootstrap samples with a seed of 91295, resulting in a total return of -52.2123%. This sampling of trades ignores regime and time correlation, and the confidence interval crossing zero and negative point estimates does not prove that all FDA strategies fail. The sponsor's registry and press release have one clinical origin, while the ITC has an independent legal origin, and SIP market transactions have a separate origin. There is no active-comparator clinical claim, and no conversion of clinical effect into stock win rate. Sources include a clinical trial registry and a stock market data website.