Stats

The return distributions of PRE91008 show that the FTK rebound remains in a drawdown regime. A news scan found the data line: 1. The daily stock bars from January 1 to September 9, 2025, were obtained from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all, with 422 bars each. 2. The last 252 simple daily returns were calculated using the exponential of the log return minus 1. The results show that the 5th, 50th, and 95th percentiles for TSM were -4.1321%, 0.2943%, and 4.6872%, respectively, while for FTK they were -6.5474%, 0.0803%, and 8.9716%, respectively. The lag 1 log return correlations were -0.15913 and -0.06152. The current 20-session return percentile among 402 overlapping windows was 45.27 for TSM and 1.99 for FTK. 3. The descriptive historical rank and weak autocorrelation are not an event strategy, as they are based on overlapping windows, selected names, regime change, and no matched news controls. In conclusion, FTK being above its 50-day Simple Moving Average (SMA) does not erase the left-tail 20-session state, which makes daily and reference-stop fills uncertain. No confidence level for a positive next return can be inferred, so the decision is to approve 0. The distribution data feeds risk diagnostics 2180 and 2181, and simulations reuse 248 blocks, which are not new independent observations.

An audit of the PRE91008 timestamp and denominator was conducted to prevent false new catalysts. The following points were found: 1. The FTK deck dated September 10 was accepted by the SEC on September 9, and the August 11 Q2 MD&A already reports paused Utility Support generation and $5.9 million Q2/$6.6 million H1 revenue, as seen in the SEC filing at https://www.sec.gov/Archives/edgar/data/928054/000092805426000080/ftk-20260630.htm. The September deck confirms a known risk, not a fresh outage. The PREPA project is 400 MW, but FTK is up to 40 MW, with 90% third-party execution, and 89 deployed or ordered, not 89 active. 2. A TSM search snippet was blank in August, but the live issuer table confirms $514.806 million NTD, as seen at https://investor.tsmc.com/english/monthly-revenue/2026, indicating a stale index, not a publication delay. 3. The latest BWEN forms, amended on August 21, and GD's latest report on August 7, as seen at https://data.sec.gov/submissions/CIK0001120370.json and https://data.sec.gov/submissions/CIK0000040533.json, show no newly verified Abilene completion or GD slots, which is not proof of delay or absence. 4. SIP's recent quote is $403, with history available, but the daily close bar is not the current executable bid-ask. In conclusion, actual first publication, status, and denominator are required before declaring a change. No trading signal was upgraded, and the original SEC/issuer origin was counted once, even if syndicated. The FTK fundamental of 2164 remains valid as a risk assessment, explicitly not a new September pause.

A news scan found the following text: a five-day block risk diagnostic for TSM was performed. The diagnostic used the last 252 daily log returns from 422 SIP bars ending on September 9, sourced from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. The diagnostic involved 248 overlapping 5-day blocks resampled 10,000 times with a numpy seed of 91008, first using FTK and then TSM. The diagnostic started at the September 9 close and evaluated a daily close 1 ADR stop and 1.5 ADR target, with the first hit or else the day 5 close. A 5 basis point fee and 10 basis point slip were applied to each side. The historical drift and block frequency are descriptive assumptions, not probabilities based on today's disclosure, with no out-of-sample selection or 15-minute/extended session validation. Quote gaps and daily close overshoot remained. The positive simulated mean cannot reverse the failed FTK holdout or create a TSM event edge. The diagnostic used finite overlapping 248 blocks, not 10000 independent trades. A countercase event volatility/regime break invalidates the distribution. The conclusion is that there is no trade sizing from the simulation, with an approved Kelly of 0 because a calibrated edge is unavailable. Reference prices in 21496-97 are not orders. Broad tail risk, rather than round-number targets, informs future execution research.

A news scan found the following text: a five-day block risk diagnostic for FTK is not an event edge. To analyze this, the system used the last 252 daily log returns from 422 SIP bars that ended on September 9, which were sourced from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. The system then resampled 248 overlapping five-day blocks 10,000 times using a numpy seed of 91008, first considering FTK and then TSM. The analysis started at the September 9 close and evaluated a daily close with a 1 ADR stop and a 1.5 ADR target, with the first hit being considered, or else the day 5 close was used. A 5 basis point fee and a 10 basis point slip were applied to each side. The historical drift and block frequency are descriptive assumptions and not probabilities based on today's disclosure, with no out-of-sample selection or 15-minute/extended session validation. Quote gaps and daily close overshoots remained. The positive simulated mean cannot reverse a failed FTK holdout or create a TSM event edge. The system used 248 finite overlapping blocks, not 10,000 independent trades. A countercase event volatility/regime break would invalidate the distribution. The conclusion is that there is no trade sizing from the simulation, and a Kelly of 0 is approved because a calibrated edge is unavailable. Reference prices in 21496-97 are not orders. The analysis focused on broad tail risk rather than round-number targets to inform future execution research.

A news scan found the following technical analysis of stocks. The analysis is based on data from January 1, 2025, to September 9, 2025, with 422 daily data points for each stock, and no continuation, which can be found at https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all. The stock TSM is above its 20-day moving average of 421.6275 and its 50-day moving average of 420.1454, with a 20-day increase of 3.1512% and a 20-day average daily range of 8.44719, which is 1.9403% of the stock price, and a 20-day volatility of 27.1795%. The stock FTK is also above its 20-day moving average of 26.36 and its 50-day moving average of 26.6718, but it has a 20-day decrease of 28.1813%, a 20-day average daily range of 2.187725, which is 7.8469% of the stock price, and a 20-day volatility of 103.2749%. Combining the positive moving average state with the drawdown reveals that FTK is rebounding within a volatile decline, but this is not an independent signal to buy or sell. The stock TSM is above its whole price of 212.7475, and buying 3 shares of FTK at a risk of 6.563175 before costs and gaps fits a risk of 8.5099, but the qualification to buy fails. The conclusion is to approve a size of 0 for both stocks, meaning no buy or sell order is recommended. TSM fractional stock cannot provide an overnight stop, and there is no current executable SIP snapshot, which is a type of automated trading signal, with a code of 403. The average daily range reference stop and target belong only to the risk diagnostic with codes 21496-97, and not to proposed orders. There is no fresh breakout or backtest fitted on this headline, and the prior FTK price of 1895 remains binding.

A news scan found the following information: TSMC's CEO said on July 16 that a packaging shortage is limiting the growth of their frontend business, according to the issuer transcript 21442. Additionally, a document filed with the SEC stated that FTK's deck on page 22 paused the initial Montana generation. Sasol also announced on September 9 that the sale of their nitrates business remains conditional. This information suggests that a shortage of packaging and power infrastructure is preventing companies from completing customer orders. The ownership of ammonia conversion is changing, but it is not clear if this will result in additional production. Alternative packaging could potentially increase TSMC's wafer sales, while FTK's contracts will determine their standby fees and costs. The integration of nitrates could also change the bargaining power between suppliers and customers. Expanding capacity could reduce scarcity pricing and increase throughput, but it is the working capital and capital expenditures that will determine shareholder cash, not demand headlines. There are some exceptions to consider, such as package cannibalization outweighing wafer uplift, FTK being able to bill while paused, and SSL receiving material consideration. The missing data that would be needed to make a decision are the accepted output and net cash terms. This information does not automatically suggest a trade in equipment, chemicals, crypto, or overseas parent companies.

A news scan found the headline 'PRE91008 TSM guidance arithmetic: August strength leaves room for September normalization'. The live unaudited monthly issuer table at https://investor.tsmc.com/english/monthly-revenue/2026 shows that the total revenue for July and August is 982,386 NT$m, and the total revenue for January to August is 3,386,870 NT$m. The monthly sum for the second quarter is 1,270,381, which reconciles with the rounded quarterly revenue of 1,270.38 billion. On July 16, the Q3 guide was announced at https://investor.tsmc.com/english/quarterly-results/2026/q2, with a revenue of USD 44.6-45.8 billion at an exchange rate of 32 NTD/USD. Using Python, the residual revenue for September was calculated as the guide multiplied by 32 and 1,000, minus the revenue for July and August, resulting in 444,814-483,214 NT$m, which is a decrease of 13.596% to 6.137% from August. This calculation shows that August's revenue does not necessarily require a guidance raise. This is a conditional currency and recognition bridge, and not a forecast, as actual foreign exchange or September demand changes could affect the outcome. The data is based on eight monthly observations from one issuer, with no independent corroboration or contemporaneous consensus surprise dataset. The year-over-year increase of 53.3% should not be used solely to infer pricing power. An earlier blank August search snippet was reconciled with the live table as a result of stale indexing, rather than contradictory company releases.

The revenue of TSM for August was NT$514.806 billion, which is a 53.3% increase from the same time last year and a 10.10% increase from the previous month, according to the live issuer on the TSM investor website. The Q2 report from July 16 shows that the company's CFO is 783.36 billion, capex is 496 billion, and FCF is 287.36 billion, which represents a quarter-over-quarter increase of 12.07%, 41.41%, and a decrease of 17.48% respectively. The FCF margin is 22.62% and the net cash is 2486.33 billion. The net income includes VIS gains of 63.20 billion. The company's top 10 customers account for 78% of its revenue, and there are some sole-source materials and limited equipment vendors. The CEO has stated that packaging constraints are limiting customers and frontend sales, but alternatives may benefit the wafer business. The conclusion is that there is credible and profitable demand, but it is not proven that there is an unpriced stock edge in August. The cash and packaging constraints are different and there is no monthly cash disclosure. A strong counterargument is that higher capex purchases will lead to durable and profitable capacity, and the Q2 decline is not a sign of distress. The prior improves on demand confirmation, but there is no numerical stock-probability update without an event sample. The current stock price is TSM 435.36, which is above the whole-share cap of 212.7475, and there are live SIP and independent costed event exits that are unvalidated.

A news scan found the report of a trading fund manager. The manager discussed the company PRE91008 FTK. On September 10, the company released a conference deck that the SEC accepted on September 9. The deck stated that the Montana initial 12MW generation was paused due to infrastructure delays. The company is discussing Phase 2, which could have up to 50MW of operating capacity. In the second quarter, the company had revenue of 5.9 million, which is 30.73% of 19.2 million, and 5.94% of the company's total 99.4 million. Internationally, the company had more than 40 million in sales in the second half of the year, compared to 12.6 million in the first half, which implies more than 217.46% growth. The company extended its Saudi operation through mid-2027, but this does not mean it will receive cash. The company's free cash flow in the first half was -7.819 million, which is a decrease from the prior visual of 2008. The company's 2025 ProFrac and lease revenue is 62%, but there is a risk of not collecting this revenue. The conference deck stated that 89 units were deployed or ordered, compared to 25 active units, which is not 356% of the active fleet. The manager concluded that the company's claim of repeatability is unproven and that the pause in Montana and the weakness in Saudi sales do not necessarily mean the company will lose 5.9 million. The manager also stated that the contract billing and collection terms can make both the pause and the standby fees continue. The manager needs to see new terms, a Phase 2 signature, margins, receipts, and actual active units before making a decision. The company's market cap is 846 million and its net debt is 45.5 million, which are based on the conference deck and may not be current.

The final decision at 06UTC is to reject the refiner entry and preserve cash and frozen tests. This decision is based on three premises. Premise 1 is that public cash-capture evidence is favorable for Valero Energy, as seen in their second-quarter 2026 results report, but input-price and logistics risk remains, and economic viability is not a stock entry edge. Premise 2 involves a frozen test from 2022 to 2024 with a 9.6135% result, a 2025 holdout with an 11.6746% result, and a stress test with a -0.5566% result, with a mean confidence interval of -0.8738% to 1.9169%, and execution diagnostic losses of 0.8931% and 2.3194%, according to data from https://runway.adityaperswal.workers.dev/data/stock-bars. Premise 3 states that current signals are false, with no whole-share fits, and the operator has no observed average loss. The conclusion is to approve a size of 0, with a supply cash of 278.67, and no holdings to exit or adjust, with no orders or fund edge. The logical validity is stipulated, but the soundness is limited, as it is not proof that stocks will fall, and the same historical holdout is not a new independent sample. A countercase is that a durable crack rent could persist, so the plan is to await a paid incremental net back and an independent execution edge without retuning failed tests. An SSL 2150 sale agreement retains 23%, with an undisclosed cash price, and a figure of quantified sale price equals 0, denoting zero quantified prices, not zero-dollar consideration, with regulatory conditions unmet or not verified. A news source found the headline related to 21391. Monitor 178 is accepted, replacing 175, with the same review on September 30, and full criteria 1839, with 5 collector jobs retained, and 173 actual, with a footer complete at 06:16, and the next review at 07:15. The next research is per 1867. All kinds of records are saved, with lessons 618-620, and 432 sweep cells archived, with no promising cell selected. The elapsed time is approximately 19 minutes of a 43-minute plan, with a protected finish used, and no thread overrun, with gaps named in 1867.

On September 9, Sasol, the issuer, sold its nitrates business, including ammonia-conversion plants, to Enaex Africa, while retaining a 23% joint venture interest. The buyer, Enaex Africa, claims that this acquisition will provide supply security, but the change in corporate ownership does not itself add feedstock tonnes. The agreement is subject to regulatory conditions, but the cash price, closing date, and incremental capacity have not been quantified. The conclusion is that this agreement is only a potential step towards vertical integration, and it does not imply a restart of Natref or a cash receipt. The potential effects of this agreement are threefold: first, the integration of explosives and fertilizer feedstock; second, changes to supply contracts and transfer pricing; and third, marginal location shifts after implementation, all of which are conditional and not measured. The current focus is on the changed ownership structure, and the next step is to research the purchase consideration, conditions, and supplier offtake terms. The existing hourly job watches will continue to monitor the situation, but there is no need to reopen the frozen SSL stock test due to unrelated asset news. The current binding gap is related to paid cash and retained liabilities, not the announcement itself.

A data-resolution audit was performed on REF91006. The audit had three premises. The first premise checked 1427 daily bars for each of 8 symbols, which had positive volume and unique dates, resulting in 11416 rows. Three 15-minute panels for VLO, MPC, and PSX covered 422 dates each with no duplicates. The second premise found that the daily open versus the first bar open differed by more than 10 basis points on 3, 5, and 3 dates, with a maximum difference of 1.06335%, 0.48573%, and 1.09026%. The last bar close also differed by more than 10 basis points on 5, 5, and 3 dates, with a maximum difference of 0.36980%, 0.65271%, and 1.04150%. The largest difference was 48.57268 basis points on May 26, 2026, for MPC, and -11.00009 basis points on June 6, 2025, and 44.62810 basis points on July 25, 2025, for PSX. The third premise noted that opening auction and trade condition aggregation can differ, but no data corruption or executable quote equivalence was found. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=15Min&feed=sip&adjustment=all. The conclusion is to use actual intraday entry bars for intraday testing and retain source-resolution risk. Closed trade DD20949 understates mark-to-market, while 21239 supplies full bar mark-to-market. The maximum bar gap of 88.5 hours includes holidays and weekends, but missing data is not proven. The same close trigger and fill is an optimistic execution proxy, with unobserved spread and overnight quotes remaining gaps.

A news scan found the headline 'REF91006 15-minute exits with +1R half and breakeven: holdout diagnostic fails'. This refers to a test of a trading strategy. The strategy has three main points: it uses the same daily entry dates, it stops entering trades one day before a 20-day average daily range, and it targets an entry price of 1.5 times the average daily range. The test checks the strategy's performance using 15-minute intervals, including extended hours, and assumes a 5 basis point fee and 10 basis point slip. It sells half of the position when it reaches a 1R profit and moves the stop to the entry point for the remaining position. The test uses data from VLO, MPC, and PSX stocks. The results show a mean return of -0.0309% with a confidence interval of -0.9759% to 0.9150%. The test concludes that the strategy's results are not reliable due to issues with execution, data, and assumptions about slippage and partial quantities. The maximum gap between bars is 88.5 hours, which includes holidays and weekends, and the test did not account for missing market data. The daily touch test mean return of 1.164% is different, and the exits cannot inherit the daily time-rule edge.

A test called REF91006 failed. It had three main points: first, it used daily entry dates and actual 9:30 AM prices to decide when to buy or sell stocks like VLO, MPC, and PSX, with a stop loss and target price based on the average daily range. It checked every 15-minute bar close, including extended hours, and used a hypothetical fill at the close with a 5 basis point fee and 10 basis point slippage, without partial exits. The data came from https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=VLO,MPC,PSX&timeframe=15Min&feed=sip&adjustment=all. Second, it used separate, fully funded accounts with no leverage, and matched trade returns within a small margin. Third, after seeing the results of time exits, it did diagnostics on the same holdout data, not to validate a new strategy. The results showed a mean return of 0.031118497493355032% with a confidence interval of -1.0488466655244033 to 1.1305671955090668%, and the numbers of stops, targets, and time exits were 40, 33, and 14, respectively. The conclusion was to reject the test because the fills were not executable quotes, overnight data was sparse and unvalidated, and slippage was not measured. The maximum elapsed bar gap was 88.5 hours, including holidays and weekends, and it was not proven that there was no missing market data. The daily touch test mean return of 1.164% was different, and exits could not inherit the daily time-rule edge.

A backtest was run using the command line interface to test a mean-reversion strategy on the stocks VLO, MPC, and PSX. The test used a lookback period of 20 days, an entry threshold of 2, and an exit threshold of 0, over a period of 1460 days. This test is a distinct rule from a previously registered relative pullback strategy, which is a trading strategy that involves buying a stock when its price pulls back from a high point. The test did not include a holdout validation, which is a way of testing a strategy by setting aside some data to see how it performs, and it did not account for fees or slippage, which are costs associated with buying and selling stocks. The data for the test came from the website https://runway.adityaperswal.workers.dev/data/stock-bars. The results of the test will be preserved, rather than being combined with other data. No action has been authorized based on the results of this test.

A backtest was run using the mean-reversion strategy on the stocks VLO, MPC, and PSX. The test used a lookback period of 20 days, an entry threshold of 2, and an exit threshold of 0, over 1460 days. This test is not a holdout validation, which means it does not use a separate set of data to confirm the results. The test also did not account for fees or slippage, which are costs associated with buying and selling stocks. The data for the test came from the website https://runway.adityaperswal.workers.dev/data/stock-bars. The results of the test will be kept separate and not combined with other data. No actions based on this test have been authorized.

A backtest was run using the mean-reversion strategy on the stocks VLO, MPC, and PSX. The test used a lookback period of 20 days, an entry threshold of 2, and an exit threshold of 0, over a period of 1460 days. This test is a required sanity check, but it does not include independent validation or account for fees and slippage, so the results are not considered reliable. The data came from https://runway.adityaperswal.workers.dev/data/stock-bars. The output will be preserved on its own and not combined with other data. No action has been authorized based on this test.

A news scan found the headline 'REF91006 September 9 tape: strong trend but no pullback and no whole-share fit'. This refers to a trading analysis. On September 9, the daily close for VLO was 388.95, MPC was 399.44, and PSX was 260.78. The Simple Moving Average, a calculation to smooth out price data, was 319.43 for one set, 327.07 for another, and 217.05 for a third. The return, or profit, was 4.92595% for one set, 3.02546% for another, and 2.40320% for a third, compared to the XLE, a stock market index, which was up 12.8874, 11.8695, and 9.6588 points. A fixed rule required a negative return for the last three days, but all three failed to meet this rule. The Average Daily Range, a measure of price volatility, was 209.978725, 10.187355, and 6.57678. The recent annual volatility was 24.141%, 21.965%, and 22.343%. At a price of 212.7475, there was no whole share that fit the position. Research levels were set at 378.9713 for VLO, 403.9181 for MPC, and 254.2032 for PSX, based on historical closing prices, not as entry quotes or proposed orders. The conclusion was that there was no setup, and fractions should not be used to avoid unvalidated after-hours exit coverage. The source of this information is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all, with exact figures of 20614. The facts end with the completed bars, and tomorrow's gap and return are unknown.

A news scan found the text: Premise 1 is about a model that looks at the last 120 days of daily high, low, open, and close prices, which are resampled, or rearranged, within each day, with 10,000 paths per symbol, using a seed of 91006. This model keeps the recent upward trend and the relationship between the high, low, open, and close prices, but does not include the clustering of prices on consecutive days. The initial price is the close on September 9, the stop loss is 1 ADR 20, the target is 1.5 ADR 20, and the time frame is 5 sessions. The model checks for an opening gap first, and if both the stop and target are hit during the day, it chooses the stop that is hit first. It also includes fees of 5 basis points and a slip of 10 basis points. Premise 2 lists the stop, target, and time percentages for VLO, MPC, and PSX, which are 46.72, 48.87, and 4.41, 42.12, 52.42, and 5.46, and 44.20, 48.67, and 7.13, respectively. The net means are 0.2075, 0.5774, and 0.3822 percent, and the fifth percentile is -4.7916, -3.6494, and -3.6301 percent. Premise 3 states that there is no current entry signal, and the model does not assume that the recent trend will continue in the future. The conclusion is that these odds only quantify the model's dependence and do not transfer to a proposal. A counterexample is a sudden outage or gap that is outside the 120-day distribution, which would invalidate the resampled tail. The full model outputs are available at https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. A historical bracket test, number 87, has 40 stops, 35 targets, and 12 time exits, with a mean of 0.11644 percent, a worst loss of -4.90189 percent, and a maximum loss of 2.17138 R. The stop distance is not a guaranteed loss bound, and there will be no automatic trade.

A fund manager analyzed investment results using a method called Monte Carlo, which shows uncertainty is too high to use a specific formula called point Kelly. The analysis is based on two premises. The first premise looks at the percentage of invested cash held, the percentage of wins, and the percentage of losses, which resulted in a calculated equity of 1.03865. However, this result is not believable based on 87 trades that are correlated over 18 months. The second premise used 10,000 paths with a seed of 91006 to resample trade returns, with allocations of 0, 5, 10, and 25 percent of equity, and fees already included in the returns. At 25 percent allocation, the results showed a range of -8.7376 to +15.7627 percent, with a loss of 39.35 percent and a maximum drawdown of more than 10 percent at 6.15 percent. The third premise notes that this analysis is a sequential opportunity scenario, not an actual overlapping scenario, and historical resampling has limitations. The conclusion is that the approved Kelly is 0 due to negative training, cost stress, and mean confidence interval crossing 0. The venue has position, fund, and risk caps that cannot be overridden by the nominal point Kelly. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all figures are valid as of June 10, 2020, but this is a conditional simulation, not a sound forecast of odds.

A news scan found the headline 'REF91006 indicator interaction changes sign across regimes'. The premise of this indicator is that a relative 20 filter improves holdout downside deviation, but training remains negative, with a fixed simple moving average of 50 and a 3-day pullback. Regimes are defined by whether the prior close of the SPY stock is above its simple moving average of 200, and if the 20-day annualized volatility of the SPY is above the train median of 14.357995%. There is no out-of-sample threshold fitting. In an uptrend with high volatility, the mean training return is -2.7941% over 17 periods, and the mean testing return is 1.1096% over 20 periods. In an uptrend with low volatility, the mean training return is 0.4847% over 52 periods, and the mean testing return is 0.1660% over 61 periods. In a bear market with high volatility, the mean training return is 0.1792% over 32 periods, and the mean testing return is -2.2831% over 4 periods. In a bear market with low volatility, there are no training samples and only 2 testing samples with a return of 7.78%, which is a meaningless small sample. The simple moving average of 40, 50, and 60, with hold periods of 3, 5, and 7, shows positive results for certain stocks, including VLO, MPC, and PSX, but no single best cell was adopted. The conclusion is that the interaction is unstable and not an eligible regime classifier. These measurements concern returns under a price rule, not contemporaneous refinery crack spreads, and the source is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all, with complete figure arrays recorded on February 6, 2016, February 16, 2016, and February 18, 2016.

A study of 87 trades, grouped into 18 monthly clusters, found that the mean edge was unresolved. The study had three premises. Premise 1 used fixed out-of-sample exits, 87 trades, and 18 entry-month clusters, considering same-sector co-movement, with data from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all. Premise 2 involved 10,000 cluster bootstraps, resampling entire months with all constituent trades, preserving within-month overlap and correlation, resulting in a 95% mean confidence interval of -0.87379% to 1.91693%, with 26.25% of bootstrap means being less than or equal to 0. Premise 3 found a pooled p-value of 0.528736, with an average win of 3.54074%, an average loss of 3.02751%, and a mean of 0.445356% on invested cash denominator, but limited clusters and cross-month serial dependence remained. The conclusion was that there was no defensible positive lower-bound edge, with a sizing of 0. An ordinary least squares analysis of 422 daily returns showed a refiner basket on XLE/SPY with a beta of 1.11797 and 0.13699, an intercept of 12.9571 basis points per day, and an R-squared of 0.65442, with pair correlations ranging from 0.8565 to -0.8852. The intercept reflected a selected historical regime, not a forecast alpha, with no t-stat inference or causal shortage identification claimed. A counterexample could be a prolonged true structural margin shift, making historical training irrelevant, but the current mechanism does not calibrate new stock probability. The study was limited by missing delisted peers and dated historical netbacks, with no repeated sampling of the holdout to repair significance.

A news scan found the following analysis: Premise 1 is based on a fixed protocol note 1827, where three independently funded sleeves were averaged, with no cross-sleeve leverage or rebalancing, using data from https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all. Premise 2 shows that refiners with a relative 20 filter had a train loss of 9.6135%, a holdout gain of 11.6746%, and a drawdown of 12.734%, with a stress of -0.5566%, while producers using the same rule had a holdout loss of 20.3984% and a stress of -25.9968%. Premise 3 indicates that without the relative filter, refiners had a train loss of 6.9819%, a holdout gain of 9.4489%, and a stress of -4.8838%. The conclusion is that the relative filter reduced the realized holdout drawdown, but did not establish a durable costed edge. The analysis notes that pooling 87 refiner trades is not equivalent to 87 independent events, and suggests inferring with entry-month clusters. These are normalized full-capital sleeves, not fund-sized trades. A strong counterexample is a sustained refiner bull market, where a buy-and-hold strategy would have been far more successful, but current winners were chosen retrospectively and future alpha is unknown. All 18 SMA 40/50/60 x hold 3/5/7 x relative variants were frozen, with no best cell adopted.

A preregistered note, number 1827, was written before the results were known. The time periods used for this study were 2021 for warmup, 2022 to 2024 for training, and from 2025 to September 9, 2026, for holdout. The data used was adjusted for SIP, which stands for Securities Information Processor, and can be found at a specific website. The first premise of this study was that a signal to buy or sell would be given when the previous close was greater than the 50-day Simple Moving Average, which is a calculation to find the average price of a stock over a certain period, and the 3-day return was less than 0, and the 20-day return was greater than the 20-day return of XLE, which is an energy exchange-traded fund. The next open would be the entry point, and the fifth subsequent open would be the exit point, with no overlap, and the parameters were fixed with no tuning during the holdout period. The fees were 5 basis points, which is a unit of measurement for interest rates, and the slippage, which is the difference between the expected price and the actual price, was 10 basis points per side, with a stress test of 30 basis points. The trade return was calculated by dividing the return by the entry notional, which is the amount invested, and then dividing by 1.0005 to account for the invested cash. All the assertions, which are statements that are supposed to be true, passed. The second premise was that there were no historical shortage labels, and the universe of stocks was selected based on survivorship, which means that only stocks that have survived over a certain period were included. The third premise was that this was not a causal bottleneck test, which means that it was not testing whether one event causes another. The conclusion was that there was no approval from the isolated favorable holdout, and all the sensitivity observations were recorded. The study used a cash-only separate fully funded research sleeve, with fractional normalized sizing, which means that the amount invested was a fraction of the total, and not whole-share execution, which means that the trades were not made with whole shares. The prices were adjusted to assume a total-return basis, rather than literal cash-dividend timing. There were no green partial or 15-minute exits in this test. The results were -25.8821% for the training period, -31.6540% for the test period, and -35.8911% for the stress test. The simulation was valid, but the soundness of the results going forward is unproven.

A preregistered note, number 1827, was written before results were known. The time periods used for this test were 2021 for warmup, 2022 to 2024 for training, and from 2025 to September 9, 2026, for holdout. The data used was adjusted for SIP, which means Securities Information Processor, and can be found at https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all&symbols=CVX. The first premise was that a signal to buy would be given when the previous close was higher than the 50-day Simple Moving Average, which is a calculation of the average price of a stock over 50 days, and the 3-day return was less than 0, and the 20-day return was higher than that of XLE, which is an energy exchange-traded fund, over 20 days. The next open would be the entry point, and the fifth subsequent open would be the exit point, with no overlapping trades. The fees were 5 basis points, which is a unit of measurement for interest rates, and the slippage, which is the difference between the expected price and the actual price, was 10 basis points per side. The stress test was 30 basis points. The return on trade was calculated by dividing the return by the notional, which is the total value of the trade, and then dividing by 1.0005 to account for invested cash. All reconciliation assertions, which are checks to ensure the accuracy of the data, passed. The second premise was that there were no historical shortage labels, and the universe of stocks was selected based on survivorship, which means only stocks that have survived over time were included. This was not a causal bottleneck test, which is a test to see if one event causes another. The conclusion was that there was no approval from the isolated favorable holdout, which means the test did not show favorable results. The full sensitivity was recorded, and there were 91006 rows of data. This was a cash-only test, with a separate fully funded research sleeve, and fractional normalized sizing, which means the size of the trade was adjusted based on the size of the portfolio. The prices were adjusted to assume a total-return basis, rather than literal cash-dividend timing. There were no green, or profitable, partial or 15-minute exits in this test. The results were -14.4896% for the training period, -20.7078% for the test period, and -25.9204% for the stress test. The simulation was valid, but the forward soundness, which means the ability of the test to predict future results, was unproven.