Notes

A news scan found the following price levels: CELH last price was 30.62, previous close was 31.61. CMG last price was 36.95, previous close was 37.56. DUOL last price was 154.34, previous close was 158.82. ELF last price was 109.66, previous close was 107.4. WING last price was 109.24, previous close was 110.15. The ELF watch zone is between 90 and 95, which is a potential buy zone if the price pulls back. DUOL is being watched for momentum, and CMG, CELH, and WING are being considered in the same context.

A research project for a crypto fund found that funding and open interest quantities were not eligible for directional leverage. The project rejected a negative-funding rebound rule. It used data from OKX and Binance APIs. The project encountered errors when trying to access funding rate data for BTC, ETH, and SOL on Binance, with a 451 Client Error for each. The errors occurred for the URLs https://fapi.binance.com/fapi/v1/fundingRate?symbol=BTCUSDT&limit=100, https://fapi.binance.com/fapi/v1/fundingRate?symbol=ETHUSDT&limit=100, and https://fapi.binance.com/fapi/v1/fundingRate?symbol=SOLUSDT&limit=100, all with eligible set to 0.

This is a research alert only, meaning it is for informational purposes and no action will be taken. The data is based on hourly and cost-adjusted Out-of-Sample validation, which is a method of testing a model's performance on unseen data. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The following cryptocurrencies were watched on September 4, 2026: BTC/USD, which closed at 79676.747, ETH/USD, which closed at 2456.376, SOL/USD, which closed at 101.95, LTC/USD, which closed at 50.76, UNI/USD, which closed at 6.18484, AVAX/USD, which closed at 7.3885, DOGE/USD, which closed at 0.0848185, LINK/USD, which closed at 11.6349, XRP/USD, which closed at 1.398615, and DOT/USD, which closed at 0.90035. For each cryptocurrency, the prior 20-day average is listed, as well as a signal and next value, all of which are set to 0, indicating no action will be taken, and an eligible value, also set to 0.

A news scan found the headline that on 2026-09-04, Novartis’ Pelacarsen Phase III Lp(a)HORIZON Trial failed its primary cardiovascular endpoint, which means it did not meet its main goal of reducing cardiovascular events, although it did lower Lp(a) levels. On the same day, Needham reiterated, or repeated, its Buy rating on Ionis Pharmaceuticals and maintained its $86 price target. BMO Capital maintained its Market Perform rating on Ionis Pharmaceuticals and raised its price target to $61. An Ionis Pharmaceuticals executive said that Zanvastro will be available at a list price of $285,000 per dose. Also on the same day, Ionis Pharmaceuticals' drug won FDA approval as the first treatment for Alexander Disease. HC Wainwright & Co. reiterated its Buy rating on Ionis Pharmaceuticals and maintained its $115 price target. On 2026-09-03, the FDA approved Zanvastro for the treatment of Alexander Disease. On 2026-08-31, HC Wainwright & Co. reiterated its Buy rating on Ionis Pharmaceuticals and maintained its $115 price target. On 2026-08-27, Citizens reiterated its Market Outperform rating on Mirum Pharmaceuticals and maintained its $137 price target. On 2026-08-20, an analyst turned bearish on Allstate and listed the top 5 downgrades for Thursday, which included Ionis Pharmaceuticals. On the same day, an analyst turned bullish on Merck and listed the top 5 upgrades for Thursday, which included Ionis Pharmaceuticals. Also on the same day, an analyst began coverage on Sphere Entertainment with a bullish note and listed the top 5 initiations for Thursday, which included Ionis Pharmaceuticals. On 2026-08-19, an analyst turned bearish on Baidu and listed the top 5 downgrades for Wednesday, which included Ionis Pharmaceuticals. On the same day, an analyst began coverage on Alnylam Pharmaceuticals with a bullish note and listed the top 5 initiations for Wednesday, which included Ionis Pharmaceuticals. BMO Capital initiated coverage on Ionis Pharmaceuticals with a Market Perform rating.

A research project for a crypto fund found that negative funding rebound was rejected. The project looked at quantities, which are not related to directional leverage. It used data from OKX and Binance. A funding rate, which is a fee paid by traders, and open interest, which is the number of open contracts, were checked. Errors occurred when trying to get funding rates for BTC, ETH, and SOL from Binance, with a 451 Client Error, meaning the request was not allowed, and the eligible status was 0, meaning the request was not approved.

A research project for a crypto fund found that negative funding rebound rules were rejected. The project looked at quantities, which are not the same as directional leverage, meaning they do not indicate the direction of a trade. The project used data from OKX and Binance APIs, specifically the funding rate and open interest. However, the project encountered errors when trying to access funding rate data for BTC, ETH, and SOL on Binance, with a 451 Client Error for each, and eligible was set to 0 for each.

This is a research alert only, meaning it is for informational purposes and no actions will be taken. The data is based on hourly and cost-adjusted Out-Of-Sample validation, which is a method of testing a model's performance on unseen data. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The following cryptocurrencies were watched on September 4, 2026: BTC/USD, which closed at 79676.747, ETH/USD, which closed at 2456.376, SOL/USD, which closed at 101.95, LTC/USD, which closed at 50.76, UNI/USD, which closed at 6.18484, AVAX/USD, which closed at 7.3885, DOGE/USD, which closed at 0.0848185, LINK/USD, which closed at 11.6349, XRP/USD, which closed at 1.398615, and DOT/USD, which closed at 0.90035. For each cryptocurrency, the prior 20-day average is listed, as well as a signal and eligibility status, all of which are set to 0, indicating no action will be taken.

A news scan found the headline that on 2026-09-04T20:32:53Z, Novartis' Pelacarsen Phase III Lp(a)HORIZON Trial failed its primary cardiovascular endpoint, which means it did not meet its main goal of reducing cardiovascular events, although it did lower Lp(a) levels. On 2026-09-04T15:59:19Z, Needham reiterated, or repeated, its buy rating on Ionis Pharmaceuticals and maintained its price target of $86. On 2026-09-04T15:12:41Z, BMO Capital maintained its market perform rating on Ionis Pharmaceuticals and raised its price target to $61. An Ionis Pharmaceuticals executive said on 2026-09-04T14:23:20Z that Zanvastro will be available at a list price of $285,000 per dose. On 2026-09-04T12:23:39Z, the FDA, or Food and Drug Administration, approved an Ionis drug as the first treatment for Alexander disease. On 2026-09-04T10:29:48Z, HC Wainwright & Co. reiterated its buy rating on Ionis Pharmaceuticals and maintained its price target of $115. The FDA approved Zanvastro for the treatment of Alexander disease on 2026-09-03T19:56:03Z. On 2026-08-31T10:26:26Z, HC Wainwright & Co. reiterated its buy rating on Ionis Pharmaceuticals and maintained its price target of $115. Citizens reiterated its market outperform rating on Mirum Pharmaceuticals on 2026-08-27T14:01:27Z and maintained its price target of $137. A news scan found the headline that on 2026-08-20T16:01:25Z, an analyst turned bearish on Allstate and listed the top 5 downgrades for Thursday, which included Ionis Pharmaceuticals. On 2026-08-20T12:50:45Z, an analyst turned bullish on Merck and listed the top 5 upgrades for Thursday, which included Ionis Pharmaceuticals. A news scan found the headline that on 2026-08-20T11:51:49Z, an analyst began coverage on Sphere Entertainment with a bullish note and listed the top 5 initiations for Thursday, which included Ionis Pharmaceuticals. On 2026-08-19T12:52:50Z, an analyst turned bearish on Baidu and listed the top 5 downgrades for Wednesday, which included Ionis Pharmaceuticals. A news scan found the headline that on 2026-08-19T11:55:12Z, an analyst began coverage on Alnylam Pharmaceuticals with a bullish note and listed the top 5 initiations for Wednesday, which included Ionis Pharmaceuticals. On 2026-08-19T10:45:21Z, BMO Capital initiated coverage on Ionis Pharmaceuticals with a market perform rating.

A news scan found the following price levels: CELH last price was 30.62, previous close was 31.61. CMG last price was 36.95, previous close was 37.56. DUOL last price was 154.34, previous close was 158.82. ELF last price was 109.66, previous close was 107.4. WING last price was 109.24, previous close was 110.15. The ELF watch zone is between 90 and 95, which is a potential buy zone for a pullback. DUOL is being watched for momentum, and CMG, CELH, and WING are being considered in the same context.

A news scan found the headline 'Sep5runfinished~04:18UTC,about18min/43plan,hard45notapproached.' This means a run finished on September 5 at 04:18 UTC, taking about 18 minutes, which is less than the planned 43 minutes, and did not approach the hard limit of 45. The cash balance is 245.69, with no orders or proposals. The controlling strategy is based on the 04UTC deep-dive conclusion. There are frozen questions and protocols, including backtests, statistics, simulations, and technical and fundamental analyses. The results show a 5-session increase of 19.90% compared to a basket increase of 72.67%, with a range of -14.84% and 549 trades. The cost assumptions do not include measured venue fees, and 32.42% of the selections are repeats, which means netting is untested. The next research must identify temporary flow and remove announcement moves before creating a fresh protocol. A replay of 547 observations from 2280 to 2826 was verified, and metrics were retrieved with a fund quant and symbol filter. The results include all 26840 raw bars, main trade and equity ledgers, and 20 experiment metrics. The remaining gaps include raw cash dividends, 15-minute and whole-share fills, NAV and signed flow series, and netting and prospective validation.

At 04UTC, a deep dive finished at 04:15, which is about 15 minutes out of a planned 43 minutes and a maximum of 45 minutes. The final 5 minutes of reserve were entered at 04:12. There were no proposals and the cash amount was 235.86. A funding contrarian rule, which is a rule that goes against the usual funding strategy, rejected all three major tests, numbered 127 to 129, because the current entries were false. This should not be confused with a prior breakout watch for SOL and XRP or the return of historical subgroups. An archive of 229 observations, which is a collection of data points, was verified, with a metric named 'Sep 5 04UTC funding archive i/229' having a numeric value. A query of fund crypto observations, which is a request for data, currently contains all 229 verified exact observations, limited to 500. The results include code, exact trades, complete funding rates, daily bars, regression rows, and audits, and can be reproduced by running the scripts fetch.py, repair.py, test.py, stats.py, risk.py, audit.py, and audit2.py in sequence. The raw hourly data is not persisted in the journal and may be revised upon refetch. The artifacts are stored in local/tmp/codex-run-oafEEX/artifacts. There was a source bar inconsistency on December 2, 2025, and a missing BTC hour on March 12, 2026, which prevent exact execution claims, and there is no optimistic repair. A new job, job19, retains the OI contract, coin, and USD, with a forecast and settled Binance daily sum, and is eligible for 0. The local test is fresh for all three, and the first sandbox test is pending at 04:14, to be verified at 06UTC. This replaces the cancelled job 13 and retains the monitors 11/14 and 10/16. There is no need for a new dated monitor. The next useful question is to reconcile vendor daily and hourly aggregation or prospectively collect participant and flow data, and not to repeat the negative funding backtest as new evidence. The existing IBIT matched shares and UNI terminal audit remain, with prior gaps outside tonight's dive. No other fund lessons or monitors have changed.

A news scan found the headline about a job related to a USDA fresh-beef entry report. The report is for research only and does not include quota utilization. The data is based on product weight, not carcass equivalent. The report was fetched from the USDA website on 2026-09-05T04:15:16.955775+00:00. The report covers the period from 8/23/2026 to 8/29/2026. The fresh beef table includes rows for different countries and columns for various products, including carcasses, AMR/ground, boneless trimmings, and cuts/offals. The final four fields show the weekly total, year-to-date total, year-to-date prior total, and percentage change. The countries included in the report are Argentina, Australia, Brazil, Canada, Mexico, New Zealand, Nicaragua, and Uruguay. The data for each country is as follows: Argentina had 475 carcasses, 947 AMR/ground, 137 boneless trimmings, and a weekly total of 1,559, with a year-to-date total of 67,078 and a percentage change of 160%. Australia had 96 carcasses, 6,361 AMR/ground, 144 boneless trimmings, and a weekly total of 11,375, with a year-to-date total of 345,700 and a percentage change of 17%. Brazil had 3,766 carcasses, 1,776 AMR/ground, 26 boneless trimmings, and a weekly total of 5,849, with a year-to-date total of 229,294 and a percentage change of 18%. Canada had 24 carcasses, 2,214 AMR/ground, 42 boneless trimmings, and a weekly total of 6,675, with a year-to-date total of 182,112 and a percentage change of 2%. Mexico had 50 carcasses, 155 AMR/ground, 4,053 boneless trimmings, and a weekly total of 5,460, with a year-to-date total of 179,787 and a percentage change of 23%. New Zealand had 2,383 carcasses, 24 AMR/ground, 372 boneless trimmings, and a weekly total of 2,929, with a year-to-date total of 156,318 and a percentage change of 7%. Nicaragua had 808 carcasses, 222 AMR/ground, 84 boneless trimmings, and a weekly total of 1,393, with a year-to-date total of 46,219 and a percentage change of 20%. Uruguay had 1,512 carcasses, 530 AMR/ground, 102 boneless trimmings, and a weekly total of 2,378, with a year-to-date total of 80,388 and a percentage change of -6%. The total for all countries was 38,291, with a year-to-date total of 1,299,517 and a percentage change of 14%.

A research project for a crypto fund found that funding and open interest quantities were not approved for use. The project looked at negative funding rebound, but rejected it. The quantities were not used to increase or decrease investments. The project used data from OKX and Binance. There was an error when trying to get data for Bitcoin, Ethereum, and Solana from Binance, with an error code of 451, which means the client made an error. The error messages were for the following URLs: https://www.okx.com/api/v5/public/funding-rate and https://www.okx.com/api/v5/public/open-interest, and https://fapi.binance.com/fapi/v1/fundingRate for the symbols BTCUSDT, ETHUSDT, and SOLUSDT with a limit of 100, all with eligible set to 0.

A 04UTC deep dive was completed in about 15 minutes, which is under the planned 42 minutes and hard limit of 45 minutes. The final recording and report phase was entered early. The catalyst remains at $49.14 in cash, with no orders or holdings. The controlling strategy is titled '04UTC deep-dive decision', following generic rule 132, which is to reject, and the scientific outcome is to watch and not score. A new monitor was set for a September 25, 2014, 15:30 UTC conference, and an existing review from September 23, 2018, before the September 23 panel, and a daily job from September 17, remain active. The first job in the sandbox was verified, not just scheduled. The next step at 06UTC is to check the changed primary evidence only. Future priorities include the original pre-result SAP and registry history, absolute stage III/IV counts, conditional false-positive workup denominators, FDA updated-version bridging, and earliest briefing and voting questions. There is no estimated release date from the registry. The missing approval database completeness and exact 15-minute plus one R implementation remain explicit, and no new entry will be made until an independent net event payoff qualifies. The data and code archive, which has 91 parts, was verified, and note 169 reconstruction was noted. There were 126 clinical, 131 economic, 132 fixed test, 151 denominator audit, 154 technical, 155 statistical, 156 simulation, 157 CLI, and 158 consequence analyses, and the final strategy was saved. Raw source and date conflicts remain in note 165. Other funds' PATH lessons and social monitors are outside this run's authority. No action is required before 06UTC while the situation is flat, and scheduled research remains informational.

A news scan found the headline '04UTC exact archive and scope handoff'. This is a report from a trading fund manager. The report is about a full archive of 91 observation parts, measured on September 5 at 04UTC, with global IDs from 1986 to 2097, but not all IDs are consecutive. The data is in a specific order and includes text and numbers. The data was compressed and encoded, and its integrity was verified with a unique code, SHA256 7a72ef82dcbdcc6babe79305877bfee2e2c0ab40e118d2d27f9bd1f09845bb44, which is a type of digital fingerprint. The archive is 493342 bytes in size and includes 1232 raw data points for GRAL and SPY, 560 rows of metrics, and results from 6 tests and trades. It also includes math and statistics, as well as executable code to fetch and test data. The report mentions that the data was verified to be exact. The report also mentions running three scripts: fetch.py, test.py, and derive.py. The current rule, number 158, was not changed, and a simulation was run to fix a gap in the data, but it did not change the results. There is no validation of actual 15-minute executions. There are no quote responses, so tape prices are not executable. There is no actual approval probability. A PDF file failed to load on the web, but a direct request was successful, and the PDF text was archived. The report mentions that one note was initially rejected but then validated. Other sources had gaps in their data. The report also mentions monitoring a new conference on September 14 and retaining existing panels and jobs. No other tasks were modified.

About 11 minutes have passed, and the plan is on track. A question about the trading strategy was answered within 8 minutes, and tests were done within budget. The main rule is to always sell and buy again at a certain point: 177 out of 546 possible times (32.42%) will repeat previously chosen stocks. This means the cost model includes potentially unnecessary buying and selling; however, the effect of keeping the same stocks is not fully tested, and the rule's performance is not proven to be optimal. The range of possible costs is uncertain, and even the most optimistic costs are not reliable. The trading limit is based on actual trading, not the volume of Exchange-Traded Funds (ETFs), which are a type of investment fund. No new monitoring is needed for the rejected rule, and no action is required before 06UTC. Existing jobs, including a country job and a PEAD job, which is a type of financial analysis, are being taken care of. Other responsibilities, such as monitoring social media, remain with the owner. The current ranking is for description only and does not allow for special entry outside of the normal cycle.

A news scan found the report of the automated trading fund manager. No proposals were made, with 278.67 in cash, which is flat. The strategy controlled at 04UTC, with tests showing 133 and 134 TSN and PPC reject, and 135 JBS watch, which had no training, and all signals were false. This is a type of trading strategy, TSN means Trading System Number. New fundamentals were 124, with consequences at 125, and statistics at 136, simulation at 137, technical at 138, and Command Line Interface at 139. Existing watches were unchanged. Job 18 had daily 30 runs, testing the AMS report period, with 9 new rows of Fresh Beef data. A software tool called pypdf was installed and pinned at version 6.17.0, which replaces a previous note about version 5.9.0. The first sandbox output is pending at 04:11, to be checked at the next 06UTC. New monitors were added for the September 26 USDA report, and a review on September 29 before the quota end on September 30. An existing monitor, 11 FRO, was retained. The archive was recovered using a query, which is a way to search for data, with a filter metric of 'Sep 5 04UTC protein reproducibility archive', and the results were sorted and decoded. The SHA2, a type of digital fingerprint, was 21a98926f09130fb3317e26ba56d88210f071d27382bb795f324aa663ee7549, and all 146 items were read back with an exact match. The archive contains 4584 raw bars of data, code, trades, sweeps, statistics, and primary filing sections, as well as collector output. The reproduction used the fetch.py and then test.py, with no re-tuning of the holdout, which is a type of testing data. The local file may expire. There were gaps in the actual CBP quota fill, underlying 403, current plant eligibility, landed cost, contract pass-through, post September 1 shipments, and actual venue exits. The first research branch was closed within 8 minutes, with a final recording phase at 04:10, and finished at approximately 04:12 under plan 43. Prior commentary elapsed estimates were corrected by the UTC clock. No other fund monitor or lesson was modified.

At 04:12, 12 minutes were used out of a planned 43 minutes. The system completed three primary venues and daily/hourly fixed tests, with no source thread exceeding the 9-minute allowance. A new research job, number 19, replaces the cancelled job 13, which runs every 120 minutes for 22 runs. The local test checked all three rows and preserved OI contracts, coin, and USD, as well as settled and forecast data, plus three previous settlement days from Binance. The initial local freshness used the pre-request clock and gave an age of -1, but this was corrected with a retest using the receipt clock, which gave an age of 0 and a freshness of true. The first sandbox has not yet been verified. Existing jobs 11 and 14 were verified and their complete outputs were retained, as were the ETH monitor 10 and PAXG 16. No new dated catalyst requires monitoring. Source and API errors did not block approvals. Oversized observation writes failed validation due to exceeding the 2000 character limit and wrote nothing, but 21 summaries were retried successfully, with exact trades archived. The stats boolean object mask was repaired before the final output. A requested period of 978 days was used, split from October 14, 2025, with training through October 10 and purge from October 11 to 13. The inspected holdout should not be retuned. A new funding rule to reject has been implemented for BTC, ETH, and SOL, distinct from the prior SOL/XRP breakout watch. Incomplete historical SOL data starts from August 26, 2024, with a missing hour audit saved, and there are no new risk or short proposals.

A news scan found the headline about an addendum to the American Society of Clinical Oncology, or ASCO, analysis 126. Later primary slides add more details to analysis 126. The NHS 197146 pooled round observations are not unique people. The safety claim is limited because it is based on a collection of referrals to the NHS. The first round of 303 false positives includes 54 later diagnoses, but these cannot retroactively change the 12 month classification. On PF2 slide 17, 91% of invasive cancer cases are conditional, compared to 50% of no-cancer cases, but the denominators are not specified in the text. The headline states that less than 1% of all participants had a conditional workup burden. Release one post lock has serious adverse event controls, but there is no date on slide No, and there are no serious adverse events. There is a conflict with primary p. Slides 6234 and 6324 have not been resolved, and both have failed. The failure and incidence rate ratio, or IRR, is 1.03 with a confidence interval, or CI, of 0.92-1.14, but this has a precision caveat. The latest ASCO titles have URLs. There are two independent study programs, but not multiple independent copies per release. The full absolute counts of grade III and IV events, the original statistical analysis plan history, the updated version bridge, and FDA questions remain missing. No approval or payoff probability has been assigned.

A test was done on the AMS import collector recipe and execution repair on September 5 at 04:08. The test used the AMS PDF from https://www.ams.usda.gov/mnreports/lswimpe.pdf, which covers the period from August 23 to 29. The test selected 9 rows related to Fresh Beef. The test required a specific version of pypdf, version 5.9.0, to be installed. The test extracted the final 4 fields, which are weekly total, year-to-date total, year-to-date total for the previous year, and percentage. The test did not try to guess missing product columns. Initially, the test had trouble converting the PDF to text, but this was fixed by replacing the initial method with pypdf. The test passed on the local machine. The script, called collector.py, will be archived and will install the pinned pypdf version in a temporary folder in a sandbox environment, keeping the output compact. This test tracks changes to the report, not how much of the quota is being used. The last 5 minutes of recording and reporting starts around 04:10, which is long before the hard stop time. No more research is planned.

At 04:07:22, about 7 minutes had passed, which is less than the initially estimated 2 to 6 minutes. The least predictable part of the process, which happened within 8 minutes, involved an official instrument from August 26 and a notice from August 31 from the Customs and Border Protection, or CBP, which confirmed that the policy quote does not refer to shipment. The CBP provided details in section 403, but the actual quota fill and plant contracts were not available, and these gaps were not filled from other sources. The weekly report from the United States Department of Agriculture, or USDA, will be available on September 4 and covers the period up to August 29, so it cannot be used to explain changes on September 1. New primary sources are now being used to refine the accounting without changing the overall financial outcome. Tests of the TSN and PPC networks were negative, and JBS lacks training, with all current signals being false. The remaining work includes testing the details and code, testing the AMS collector, researching USDA and quota monitors, and making a final recording. No authority to place orders was used.

The automated trading fund manager reviewed the cost attribution at 04UTC. Before calculating additional costs, the manager retained the frozen note 161 signals, weights, exits, and windows, and computed zero-cost and 1bp fee + 1bp slippage per side to decompose turnover drag. The manager did not select a new adoption threshold from these inspected windows. Only 3 out of 10 lookbacks were completed for training, and all holdout uses 5. The internal temporary filename'monthly' was renamed 'fixed', which always meant 5 sessions, with no rule or metric change. Simulations and means were clustered by calendar, and the benchmark buy-hold may remain exposed through the final incomplete rotation when the strategy is cash. The manager referred to the Nagel full paper, which featured original 5 daily-lag subportfolios, proportional negative market-adjusted returns, $1 long/$1 short, 48 industries, from 1998-2010. The manager's bottom 3/9 ETF/5 session/long only is an explicit adaptation. A news scan found the headline was not needed, instead, a PDF retrieval succeeded via requests despite a web 403 error, and pypdf was installed locally after a missing pdftotext and managed global pip failure.

A news scan found the headline '04UTC frozen experiment: five-session sector reversal'. This refers to an automated trading fund manager's notes. The manager registered an experiment before fetching sector history results on September 5 at 04:04 UTC. The main question is about low-frequency ETF sector reversal, inspired by Nagel's liquidity-provision and industry reversal strategy, but not the original long-short stock strategy. The manager used nine original sector ETFs: XLB, XLE, XLF, XLI, XLK, XLP, XLU, XLV, and XLY, with SPY as a benchmark. The training period was from 2017 to 2022, and the diagnostic chronological holdout was from 2023 to September 4, 2026, with a warm-up period in 2016. The sector holdout was not inspected in this run, but the dates overlap with prior different strategies. Every fifth common session was anchored to the first session of each partition. The manager ranked the previous completed 5-session close returns minus the equal-sector mean, bought the bottom 3 with negative residual, and allocated one-third of the initial sleeve to each, with empty slots in cash. The next open was bought, and the position was exited after 5 sessions at the next scheduled open before new selection. The baseline held for 5 sessions. A separate range variant froze the previous completed ADR 20 stop 1 ADR and target 2 ADR. The manager used a daily synthetic event order via vectorbt, with a fee of 5 basis points plus a slip of 5 basis points on each side, and stress of 25 plus 25 basis points. The training-only lookback was 3, 5, and 10, with 5 remaining fixed, and no holdout selection. The benchmark was equal 9 buy-hold and SPY, with net returns, drawdown, profit, loss, and other metrics. If the result was positive, the manager would still watch pending 15-minute, raw dividend, whole-share, and PDT execution, with a size of zero if the edge was uncertain. The manager noted that AQR's 2012 institutional cost evidence is not proof that a tiny ETF account would fail, and that original papers differ in population and horizon, so contradictions are resolved by definition. The ETF overnight 5% study is a mechanism counterexample, not a rule substitution.

The primary test for the funding experiment will use Binance's BTC/ETH/SOL USDT settled funding from 2024-01-01 through 2026-09-04. This will be compared to OKX and Bybit's recent and historical data to check for consistency across different venues, without choosing the venue with the best profit. The test period will be split into two parts: the first two-thirds of the 977 calendar-day period and the last one-third. The exact date for the split will be calculated. The data will be frozen before it is fully retrieved. The test will use UTC day, which excludes the following day's settlement, and the previous day's funding will be available by the next day's entry. The daily range tests will stop if both the high and low prices are touched. If there is a dispute about the order of events, an hourly replay can be used to resolve it. If a trade makes a profit of +1R at the end of a bar, the stop loss will be reduced by one-third at the next bar's open, and the remaining stop loss will be raised to the entry price. This management rule is an expansion of the registration 154 stop/target/time rule and is required before the results are released. The daily and hourly management frequencies are used for diagnostics and are not certified for 15-minute intervals. The assumptions about fees and slippage have not been verified with the account's fee tier, and the test is sensitive to higher costs.