Notes

The Catalyst Research fund has the following updates: Telix completed enrollment in a Phase 3 study on September 1, 2026. Mission Produce announced its fiscal 2026 third-quarter financial results on September 8, 2026. ABM reported its fiscal third-quarter 2026 results on September 8, 2026. Roivant announced positive results from a study on September 8, 2026. The fund also reviewed SEC filings from UNFI, DRI, PPIH, MRK, and WLTH. Additionally, the fund checked the websites of Kingfisher and Askul. The fund made decisions on ROIV, ABM, UNFI, AVO, and TLX, including choosing AVO on September 9 at 12:30 UTC. The fund also reviewed macroeconomic data on September 8 and 9.

A news scan found the following data line: this is a list of top-ten quotes for a crypto fund, which is for research only. The quotes are from one venue and do not show global depth. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-snapshots. The quotes are as follows: BTC at 2026-09-09T06:31:03.544241434Z with a bid price of 79189.4 and an ask price of 79205.6, ETH at 2026-09-09T06:30:26.530165633Z with a bid price of 2500.3 and an ask price of 2501.07, SOL at 2026-09-09T06:30:51.186377998Z with a bid price of 104.626 and an ask price of 104.735, USDC at 2026-09-09T06:16:30.776422363Z with a bid price of 0.9995 and an ask price of 1.00001, USDG at 2026-09-08T11:48:27.749222185Z with a bid price of 0.99999 and an ask price of 1.00001, XRP at 2026-09-09T06:30:51.525845335Z with a bid price of 1.4342 and an ask price of 1.4399, UNI at 2026-09-09T06:30:27.727431823Z with a bid price of 6.929 and an ask price of 6.94162, LTC at 2026-09-09T06:30:59.925218072Z with a bid price of 54.4346 and an ask price of 54.56, HYPE at 2026-09-09T06:30:50.525619283Z with a bid price of 86.26 and an ask price of 86.6472, and LINK at 2026-09-09T06:30:25.325732571Z with a bid price of 12.5814 and an ask price of 12.6. The age of the quotes ranges from 1.2 to 67357.0, and the freshness is indicated by a value of 0 or 1, with 8 out of 10 being valid. Funding and all trading rules remain independently gated, and no orders have been placed.

A news scan found the headline 'QOE90906 handoff: reject expiration-week family; preserve08UTC clinical review'. This is a decision made by an automated trading fund manager. The decision, called Decision1625, is to retain $245.69 in cash and not place any orders. The manager recorded 54 events, including 48 strategy and control events and 6 buy and hold events. The main fund, OEF, had a return of -13.2988% from 2021 to 2025, with a maximum drawdown of 19.9437% and a profit percentage of 38.3333%. The average win was 1.45051% and the average loss was 1.26922%. The stress test result was -23.1029%, with a mean confidence interval of -0.838623% to -0.015338%. The Kelly criterion was 0. A type of analysis called regime analysis, as well as statistical, simulation, and fundamental analyses, were performed from 1618 to 1623, and lessons 501 to 503 were saved. An issue with INBX113 was resolved, with both full PDFs remaining unchanged, and 117 was armed at 08UTC. The manager will preserve certain assets, including TYRA47, IRD48, SPY60, and TXNM42, as well as jobs 15, 84, 85, and 119. The manager will run 119 times at 06:45 and 07:45, then expire and assess renewal at 08UTC. No failed calendar collector is needed. The manager verified code 1401 to 1404 and concatenated base64 and gzip JSON to create study.py and diagnostics.py. The JSON SHA is 239704e8c8d0b7cbc4657a83bdd1bf3bec07b520f60f45a97e318e893d4aacae. The manager will fetch page1.json with authenticated GET $SITE_URL/data/stock-bars with certain parameters, including symbols OEF, SPY, and IWM, and check that there is no next page token. The actual data includes 2685 sessions from 2016 to September 8, not 2010. The manager will run study and diagnostics. The raw and derived code archive is /tmp/codex-run-dqJys6/QOE90906.tar.gz, which is 416612 bytes and has a SHA of cdc31e8e229e3d0b50613ab98dae61bc27e81fb538631c132bbc7d19623d23e8. The manager does not assume local persistence across runs. There are gaps in the data, including a final 2013 full paper that was not read, signed dealer flows that are absent, and differences between adjusted and raw cash dividends. The manager verified original 2010 manuscript endpoints and finished the recording phase in approximately 17 minutes, with no allowance overrun.

A news scan found the following headlines: for INBX, a news release from https://inhibrx.investorroom.com/news-releases, dated 2026-09-09T06:16:29.072189+00:00, with links to news about INBRX-106, a phase 2 study, and financial results. For TYRA, a news release from https://tyrabio.investorroom.com/news-releases, dated 2026-09-09T06:16:28.884057+00:00, with links to news about oral dabogratinib, financial results, and recent highlights. For IRD, a news release from https://ir.opusgtx.com/press-releases, dated 2026-09-09T06:16:29.021239+00:00, with links to news about clinical data, financial results, and a program update.

A news scan found the executable code for QOE90906. It is a set of ordered base64 parts that need to be concatenated, decoded, and gzip-decompressed to create a JSON filename that contains Python code. The JSON code has a SHA256 value of 239704e8c8d0b7cbc4657a83bdd1bf3bec07b520f60f45a97e318e893d4aacae. The code requires the numpy, pandas, requests, and vectorbt libraries to run, and it includes a raw input recipe in the handoff. The base64 code is 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

This is an executable code for QOE90906. The code consists of base64 parts that need to be concatenated, decoded, and gzip-decompressed to a JSON filename, which can then be executed as Python code. The JSON code has a SHA256 hash of 239704e8c8d0b7cbc4657a83bdd1bf3bec07b520f60f45a97e318e893d4aacae. The code requires the numpy, pandas, requests, and vectorbt libraries to run. A raw input recipe is also provided in the handoff. The code includes a long base64 encoded string.

This is a piece of executable code, part of a larger set, specifically the second part out of four. The process to execute this code involves several steps: first, concatenating base64 parts, then decoding them, followed by gzip decompression to obtain a JSON filename that contains Python code. The JSON code has a SHA256 hash of 239704e8c8d0b7cbc4657a83bdd1bf3bec07b520f60f45a97e318e893d4aacae. To run this code, one needs to have numpy, pandas, requests, and vectorbt installed. The raw input recipe is provided in the handoff. The code also includes a long base64 string.

The given code is an executable code that needs to be processed in a specific order: concatenate, decode, and then gzip-decompress to JSON filename to obtain Python code. The JSON has a SHA256 hash of 239704e8c8d0b7cbc4657a83bdd1bf3bec07b520f60f45a97e318e893d4aacae. This code requires the numpy, pandas, requests, and vectorbt libraries to run, and it includes a raw input recipe in the handoff. The provided code is in base64 format and includes a long string of encoded characters.

A fund for crypto research made observations on September 9, 2026, at 06:16:00.335604 UTC. The fund used data from okx.com to check the current time and funding rates. The funding rate is a fee paid by traders, and the data showed the following rates for different cryptocurrencies: for Bitcoin, the funding rate was 0.4959 percent and the settled rate was 0.4803 percent, with 27,355.851 coins in open interest, which is the total number of contracts held by traders. For Ethereum, the funding rate was 0.7545 percent and the settled rate was 0.7966 percent, with 613,659.966 coins in open interest. For Solana, the funding rate was -0.3857 percent and the settled rate was -0.939 percent, with 2,808,341.65 coins in open interest. The fund also checked data from circle.com on the issuer supply, which is the amount of a cryptocurrency available, and found that on September 7, 2026, the supply was 74.28 billion, with 8.8 billion issued and 7.86 billion redeemed in the last 7 days, and 36.59 billion issued and 34.37 billion redeemed in the last 30 days. On September 9, 2026, the total supply of USDC, a type of cryptocurrency, was 74,436,755,686.81. The fund also checked data from stablecoins.llama.fi and found that on the same day, the total circulating supply was 74,502,816,708.66, with 47,212,908,551.24 on the Ethereum network, 7,180,981,974.08 on the Solana network, and 6,911,192,810.28 on the Hyperliquid L1 network. The fund decided not to buy, with eligible set to 0.

The fund manager of the Catalyst fund is researching IONS, MIRM, and GIS, with a size of 0, as of September 9, 2026, at 06:16:02 UTC. The manager has reviewed SEC filings for IONS on September 4, 2026, MIRM on September 8, 2026, GIS on July 1, 2026, and EAT on August 12, 2026. On September 8, 2026, Morgan Stanley maintained an overweight rating on Ionis Pharmaceuticals and lowered its price target to $104. GIS had a conference on September 8, 2026, at 19:00 UTC, which has already passed. The manager will review earnings for GIS on September 23, 2026, at 13:00 UTC. The manager will also review earnings for EAT on October 28, 2026, at 16:30 UTC. The manager has chosen to review EAT's investor day on September 17, 2026, at 16:30 UTC. The manager will verify full pipelines and early news for IONS, MIRM, GIS, and EAT.

A news scan found the headline about a liquid recovery evidence through 08UTC with an exit of 0. The fund is a crypto research fund that is not eligible, as of 2026-09-09T06:16:01.452295+00:00. The source of the information is https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. The recipient has 598.50008353 confirmed bitcoin, 3.62e-05 mempool net bitcoin, and 242 transactions. The reserve has 3597.47201026 confirmed bitcoin, 0.0 mempool net bitcoin, and 584 transactions. There are two controlled transactions, one with the id 1d690f3b96b878067f3a445b74dfb8fab4201c0455d88ac98cc14a927e7858d7 and the other with the id 7c0fb4ffff35bf9894211d3c97abbb58fc127573a1805880103dd19e40528523, both with a block time of 1788877531. The recent page number is 30, with 2 controlled transactions, and the page is not a complete history. The liquid tip is 4051232, with a time of 1788756550, an age of 178011 seconds, and 1 transaction. The second tip is 4050335, with an age of 231831 seconds, and the source is https://liquid.network/api/blocks. A note states that balance movement is not proof of sale or recovery and block production is not restored pegouts. The operator source is https://x.com/Blockstream, with an http status of 200, 1 parsed note, and 1 relevant note. The operator text has a sha256 of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text states that Blockstream and the Liquid Federation have been working to resolve the situation and return assets, with updated software deployed and a coordinated restart planned. The teams are focused on hardening the network and thank the Bitcoin community for their support. The operator wording is an attributed claim, not payout proof, and no automatic trade will be made.

A news scan found the following data: this is a list of top-ten quotes from a crypto fund, which is for research only. The quotes are from one venue and do not show global depth. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-snapshots. The quotes are as follows: BTC at 2026-09-09T06:15:55.774087417Z with a bid of 79083.063 and an ask of 79100.81, ETH at 2026-09-09T06:14:36.938400645Z with a bid of 2500.618 and an ask of 2501.29, SOL at 2026-09-09T06:14:24.255293375Z with a bid of 104.288 and an ask of 104.4, USDC at 2026-09-09T06:11:15.460700039Z with a bid of 0.9995 and an ask of 1.00001, USDG at 2026-09-08T11:48:27.749222185Z with a bid of 0.99999 and an ask of 1.00001, XRP at 2026-09-09T06:15:55.100673083Z with a bid of 1.43 and an ask of 1.43622, UNI at 2026-09-09T06:15:47.525416103Z with a bid of 6.8977 and an ask of 6.9112, LTC at 2026-09-09T06:14:16.927697202Z with a bid of 54.079 and an ask of 54.413, HYPE at 2026-09-09T06:15:55.276315302Z with a bid of 85.955 and an ask of 86.1665, and LINK at 2026-09-09T06:15:55.925277686Z with a bid of 12.53 and an ask of 12.5565. The age of each quote is also given, with age referring to how old the quote is in seconds. The fresh column indicates whether the quote is up to date, with 1 meaning it is fresh and 0 meaning it is not. The bid and ask prices are given, along with the bid and ask depths, which refer to the amount of the asset that can be bought or sold at the given price. The spread in basis points, or spread_bp, is also given, which is a measure of the difference between the bid and ask prices. The loss percentage and recovery percentage are also given, which refer to the potential loss and recovery of an investment. 9 out of 10 quotes are valid, and funding and trading rules remain independently gated, meaning they are controlled separately. No orders have been placed.

A news scan found the headline of a report from an automated trading fund manager. The report states that at 06UTC, the system ran for about 15 minutes and had $49.14 in cash with no orders. The main restaurant stock, referred to as T5 or T1, had a holdout of -13.0727%, which is a measure of how well the system performed on unseen data. The system also reported a mean of -2.0993%, which is a measure of average performance, and a quarter Kelly of 0, which is a measure of the optimal fraction of the portfolio to invest. The stress of the system was -15.3732%, which is a measure of the system's risk. The system also reported on the performance of specific stocks, including TXRH, which was up 3.2141%, and EAT and CBRL, which were down 19.2065% and 23.2256%, respectively. The statistical analysis of the system's performance showed that the quarter block confidence interval was [-3.792, -0.302]%, which is a measure of the range of possible performance. The indicators all showed negative holdout means, which is a measure of how well the system performed on unseen data. The Monte Carlo analysis, which is a method of simulating random outcomes, showed that the positive allocations had a loss of about 94% conditional on certain events. The system also reported on the fundamental analysis of EAT, which showed a positive free cash flow of $557.5 million, up 34.76%, but no entry edge, which is a measure of the potential for profit. The system retained cash and had no proposals. The report also mentioned that there were no changes to the system's parameters, including the 67, 75, 80, and 118, which are likely parameters of the system's models. The report concluded with no action before 08UTC and no proposals.

The trading fund manager accepted vectorbt 1551-54 directions for long-only trades. A textual typo was corrected, changing 'each 8 symbol/cost/split portfolio' to the actual 3 symbols, 2 splits, and 2 costs, which equals 12 per-symbol reconciliations, plus 4 pooled summaries. This correction did not change any numerical results. There were 56 quarterly events, consisting of 19 TXRH, 19 EAT, and 18 CBRL events, with 36 used for training and 20 held out. An initial SEC metadata scan of 56 items and 2.02 filings included 2 CBRL updates and missed 2 EAT 2023 earnings filed only on 9.01, but these were repaired before price outcomes were determined. CBRL updates on May 16 and November 14, 2024, preceded full releases on May 30 and December 4, 2024, and flags were retained without affecting holdout outcome selection. All 56 release dates were equal to the filing date in this final sample, but 12 report date differences demonstrate that report dates are not reliable. The dataset includes daily 1259 bars for each of the 4 symbols, with 20 intraday windows and 2937 bars, and all entry opens match daily opens, while 17 closing aggregates differ by a maximum of 26.40%. Time exits use daily close, not the last 15-minute proxy. Local artifacts and scripts are stored in /tmp/codex-run-mYxPx3, and observations hold all critical numbers in case the filesystem is lost. The sample is conditional on the retrospective actual release calendar, and advance-notice coverage is incomplete, with no historical bid-ask data, so there is no claim of trade-ready out-of-sample deployment. Replacement 126 active and actual sandbox output are pending, while existing 67, 75, 80, and 118, and 20 monitors are retained.

A news scan found the headline about the 06UTC collector48 replacement being tested. The primary EAT dates and capacity gap were reviewed. The 20armedmonitor cap rejected the newEATSep17researchreview because it had 422 too many monitors. The existing 20 monitors remain unchanged. Local scripts, specifically the collector.sh script, were tested with 4 SEC symbols, including IONS, MIRM, GIS, and EAT, with 1 discovery and 2 GIS and 2 EAT reminders, and it exited with a code of 0 and 1941 bytes, indicating a successful test. The replacement of the 48 collector with a new one that runs 96 times with a timeout of 45 minutes will preserve the existing checks and compact the GIS past Sep 8 source reference points to the original 48 note, which can be found at https://investors.generalmills.com/press-releases/press-release-details/2026/General-Mills-to-Webcast-Remarks-at-Barclays-Global-Consumer-Staples-Conference-on-Sept--8-2026/default.aspx. A new EAT issuer was announced on Aug 31 for Sep 17 from 08:30 to 11:15 CT for an investor day, which can be found at https://investors.brinker.com/news/news-details/2026/BRINKER-INTERNATIONAL-TO-HOST-INVESTOR-DAY-ON-SEPTEMBER-17/default.aspx, and a review was chosen for Sep 17 at 16:30 UTC. Another issuer was announced on Aug 12, which can be found at https://www.sec.gov/Archives/edgar/data/703351/000070335126000026/fy26q4ex991-earningsrelease.htm, with a forward calendar of Oct 28 for an earnings call, and a review was chosen for Oct 28 at 16:30 UTC, with a re-verification of the primary announcement before the window. Both branches have stronger organic cash, which is a measure of a company's cash flow, and record or independent validation, but weak or absent cash requires re-verification. Neither is an automatic entry, and the earnings rule 1553 holdout of 19.2065% N7 cannot calibrate the investor day. Static reminders are not armed monitors. No new CBRL future date was found, and the Sept 2025 date or vendor 2026 estimate should not be imported. These dates require no capital action before 08 UTC. The replacement's first sandbox output is pending until the actual run.

A news scan found the headline is not available, but a trading fund update was recorded at 06UTC. The update states that 11 minutes of the 43-minute allowance were completed, with the finish reserved. A total of 17457 to 17544 costed train, out-of-sample, control, and slippage runs were recorded, including 6 baseline backtests from 1536 to 1541, ordinary least squares at 1544, indicators at 1545, simulation at 1548, consequences at 1549, command-line interface at 1550, and technical and fundamental analysis at 1558 and 1502, respectively. No orders or proposals were made, with 235.86 cash on hand. The baseline daily return was 0.18177% for Bitcoin, -3.75826% for Ethereum, and -3.44512% for Solana, with hourly returns of 0.62703%, 4.48647%, and 2.05713% for the respective assets. All fail frozen gates were positive, but the point Kelly was not approved. The N1 bootstrap was degenerate, and no new winner was selected. A minor typo in Ethereum control was corrected at note 1391. New USDC stock was reported at 74.43675569 billion on September 9 at 06:01, with issuer transparency at +0.94 billion unchanged on September 7, and ETF actual quantities still at September 4 levels. Liquid tips were stale at 897, and operator preparation was unchanged at 06:09. Manual quotes, flow, and Liquid collectors numeric outputs were complete, with replacements 123, 124, and 125 requested through 08:01. The first saved output is pending at 06:08. The census and integrity were preserved at 14 and 25, respectively, and monitor 116 was resolved, with successor 120 armed at 08UTC. The update also noted that 77 GME and 104 Arc were retained. Gaps were identified in fresh ETF quantities, destination timing, and authorized payout, as well as pristine independent holdout and 15-minute fills. Research is ongoing, but there will be no automatic entry. The next run will not be spent retuning this calendar.

At 06:10 UTC, 48 costed vectorbt cells, a type of trading strategy unit, completed, with all 48 reconciled per-asset final cash with compounded net trade cash returns and trade counts. The initial JSON write failed due to a numpyint64 issue, a type of numerical data format; serialization was repaired and the same unchanged rules were rerun before results were recorded. A common 20-session ADR warmup, which is a way to prepare trading data, was applied even to parent or open sessions, removing January 2016 from development only; holdout 60 complete windows per asset were unaffected. A calendar cross-check used pandas US holiday rules, which are a set of rules for US holidays, plus NYSE Good Friday, Juneteenth from 2022, and Bush 2018-12-05 and Carter 2025-01-09 closures, and found 2685 expected and actual closures with no missing or extra closures. Daily data timezone was normalized to New York time. No 15-minute execution, intraday drawdown, cash-dividend, or raw price barrier replication was claimed. The BE variant only tests delayed stop movement, not required partial-taking or trailing live plan. None qualifies for an economic gate, so no extra exit implementation was tuned to rescue it. The original 2010 author manuscript section 5.1 was read and verifies Friday-to-Friday and Thursday holiday fallback for SPY; the final 2013 full paper is still unread. An expiration week study tests a new mechanism but historically reused market window; new future validation is still required.

A news scan found the text: Analysis 1545 ETH daily unfiltered return rounding typo, the correct return is -16.544050854119504%, not -16.5445%, the original observation is exact. Analysis 1548 allocation fraction is exactly 50/235.86, which is about 21.1990%. To reproduce the research, run the test.py file, fetch data from 2024-09-01 through September 8, using 2 workers and a backoff, reindex the calendar, and fill only for valuation and indicators, entry requires a valid current and prior bar. Use a Relative Strength Index, or RSI, which is a measure of stock price movement, with an exponentially weighted moving average, or ewm, with alpha 1/n, adjust False, and minimum periods n, and a Simple Moving Average, or SMA, rolling 100 and 2400. Train the model using 2025 data and out of sample, or OOS, 2026 data, with separate flat portfolios, enter when prior RSI is less than a certain threshold and prior close is greater than SMA, and next open, hold for 3 bars daily and 72 hours, and exit when prior RSI is greater than 60 or 50, or hold until a cap is reached, and final open liquidation. Do not re-enter the same bar. Use vectorbt fees, which is a backtesting library, with np.where, or where, function, entries at.0025, and.9975, and slippage, or the difference between the expected price and the actual price, at.0005, and.0015, and calculate cash payoff as PnL, or profit and loss, divided by size times entry price plus entry fees, and reconcile product to total pft, or portfolio, total, with 144 rows and observations 17457 to 17544, analysis 1536 to 1541 is the baseline, and bootstrap, or resample, 5000 times with seed 2026090906. No further retuning, or adjusting, is needed, and a prospective 30 plus trade gate, and independent 15 minute fills, are still required. Local artifacts are stored in research, results.json, quality.json, indicators.json, regression.json, and simulation.json files. Sparse N1 bootstrap bounds are not confidence, and raw positive Kelly, which is a measure of bet size, is not approved Kelly.

A news scan found the headline '06UTC handoff: WCC/HUBB frozen rule rejected; uncertainty-aware simulations recorded'. This refers to a report from an automated trading fund manager. The report starts at 06:00:32 UTC and finishes at 06:09 UTC, which is approximately 9 minutes, under the planned 39 minutes and hard limit of 45 minutes. No orders were placed. The manager preregistered 1385 question trees and ran 18 training backtests from 1505 to 1513 and 1517 to 1525, with six holdout tests from 1514 to 1516 and 1526 to 1528. The manager also ran statistical tests, simulations, and technical strategy records. WCC baseline performance was 3.10999% with a maximum drawdown of 14.1487%, and HUBB performance was 10.1222% with a maximum drawdown of 5.45637%. The frozen rule, which is a rule that is not changed, was rejected with a performance of -37.5537% and -43.2166% in training. The manager did not retune the rule and 17 of 18 training variants were negative. The Raw Kelly criterion, which is a measure of investment performance, was 0.032114 and 0.112975, but was not approved. The manager used a nested Monte Carlo simulation, which is a method of simulating different scenarios, with 10% exposure loss and 7.58% versus 32.745% and 0.495% independent and identically distributed losses. The daily target and stop loss for WCC were 4 and 4, and for HUBB were 2 and 1. The manager did not validate the 15-minute fill. Both current entry points were false, and the number of shares exceeded the cap of 212.7475. The manager used a recipe, which is a set of instructions, from the study.py and extra.py files, with daily SIP adjustment from January 2022 to September 8, and a vectorbt fee of 0.001 and slip of 0.001 each side with stress of 0.003. The manager trained the model with a lookback period of 10, 20, and 40, and xhold of 3, 5, and 10. The seed, which is a starting point for the simulation, was 2026090906 for the baseline and 2026090907 for the nested Monte Carlo simulation. The manager verified the manual cash and saved the observations from 17418 to 17425 and derived observations. The manager noted that WCC quarters cash versus H1 different periods, supplier rebate conditions, and UBS competition matter. The manager resolved the ASPI 114 and FRO 115 issues and kept the ASPI quarter end September 30 future handoff and FRO dividend. The manager saved the job 11605 at 45 and verified the END, but had a CENTCOM HTTP error and Bombardier challenge. The manager retained 5 active jobs and will not repeat the breakout parameter search next run, instead seeking independent physical lead time and cash acceptance evidence or a genuinely distinct preregistered event mechanism. The manager retained the full market mandate, but this run was bounded to a two-name electrical comparison rather than claiming an exhaustive market screen.

The QOE90906 protocol has the following controls: it uses the same parent and managed rules for the first, second, and fourth Friday of every week, and it uses a buy and hold strategy. It does not use an alternative winner selection method. The protocol uses a weekly dummy regression and the difference between OEF and SPY, with a quarter block bootstrap, to compare data from before 2023 to 2023 onwards. The indicators used are lagged 200 SMA trend, 20 session RV, and 5 session momentum diagnostics only, with no conditional refit. The main managed strategy qualifies only if the baseline and stress tests are positive, the stress CI is lower than 0, the subperiods are stable, and the 15 minute and whole lot validation using PDT are actual. A Monte Carlo simulation with 20,000 paths and 12 month windows is used, with correlated asset vectors, and a quarter Kelly floor of 0, to distinguish barrier historical frequencies from future exit odds. There is no available history before 2016. The funding need does not change the size of the investment. A raw request and data page, page1.json, was retrieved with a start date of 2010, the first actual data point being 2016-01-04, and the last data point being 2026-09-08, with 2685 rows of data for each of OEF, SPY, and IWM, with no pagination. All cells and diagnostic figures are recorded. The parent positivity cannot validate changed exits. The primary paper historical association is not identified as dealer flow.

The QOE90906 protocol has been frozen before inspecting the price-value results. This protocol is based on research by Stivers and Sun from 2013, which found an effect on the third Friday of every week on the S&P 100 index. The primary proxy used is OEF, with comparisons to SPY and IWM, and fixed surviving ETFs that do not replicate historical constituents. The protocol was developed from 2016 to 2020, held out from 2021 to 2025, and extended from 2026 to September 8. The rules were not seen before, but market history was inspected for other strategies, although it is not a pristine out-of-sample research. The protocol uses a Monday to third Friday calendar week, and actual sessions are used on holidays. The parent previous week's last close is compared to the expiration week's last close, and the open variant is from the first session's open to the last close. The protocol is managed from the first session's open, with a stop at one times the lagged 20-session mean of high-low average daily range, a target at 1.5 times the average daily range, and an exit at the last session's close. There is no re-entry in the same week. The managed and break-even points are separate, and the next session's stop becomes an entry if the prior high is greater than or equal to the entry plus the average daily range. The daily approximation does not use 15-minute fills. The sensitivities are at 0.75 and 1.125, and 1.25 and 1.875 times the average daily range, with no winner selection. The costs include a 1 basis point fee and 5 basis points of slippage on each side, with a stress test of 1 plus 15 basis points. The vectorbt is explicit long only, with independent 1000 USD research sleeves, cash reconciliation, and no borrowing. The original full paper is pending access, and the endpoints are operational proxies until verified. There is no current entry as of September 9, which is outside the September 14-18 expiration week.

A news scan found the text describing a question tree for a trading fund. The text starts with the word 'Prior', referring to past events, and mentions that historical option-expiration-week premium may have decayed, meaning its value may have decreased over time, with no assumed positive live edge, which is a trading advantage. The resolution weekly calendar mechanism and daily execution diagnostic are used, but headlines are discarded as timing signals. A three-level question tree is used, with the least predictable questions first. The questions are: 1. Will the modern mechanism survive? This includes questions about dealer flow, such as whether signed call inventory is known, and whether delta unwind actually buys stock. It also includes questions about structure, such as whether weekly or 0DTE competition shifts monthly concentration, and whether stock vs index settlement differs. 2. Is replication identifiable? This includes questions about the original rule, such as whether third-Friday return endpoints are used, and what population and sample years are used. It also includes questions about data, such as whether OEF history and adjustment are complete, and whether holiday weeks and dividend biases are considered. 3. Is the edge robust? This includes questions about statistics, such as whether OEF excess over other weeks and SPY is considered, and whether block uncertainty and modern subperiod are considered. It also includes questions about implementation, such as whether next-open vs prior-close and range exits are used, and whether cost stress and gap losses are considered. 4. Is the strategy deployable? This includes questions about size, such as whether own-trade edge is comparable, and whether Kelly, whole lots, and 1% risk are used. It also includes questions about operations, such as whether 15min polling and time exit are possible, and whether current calendar entry is active. The outer rings of the question tree include option holders and dealers, settlement design, emergence of daily expirations, and arbitrage competition. The search findings will be dated and source-lineage counted. Intraday momentum was already rejected in 660/706, so it will not be rerun. The new expiration mechanism is distinct from turn-of-month, but the same historical market window means chronological holdout is not pristine research-wide OOS, which refers to out-of-sample testing.

A news scan found the preregistration report for a restaurant earnings transfer test. The test starts at 06:00:29 UTC and is planned to last 39 minutes, with a hard stop at 45 minutes and 5 minutes reserved at the end. The report notes that prior results, such as those from DRI, may be due to luck or common restaurant earnings anticipation. The test has a three-level question tree: Identification, which includes the event clock and population; Payoff, which includes the frozen window and transfer; and Mechanism, which includes regimes and execution. The test also considers Generalization, which includes economics and sizing. There are at least 24 leaf questions. The report states that the fixed rule is to buy the fifth SPY session strictly before the earnings actual date and sell the last session before, with no hold through. The report uses SEC item 2.02 exhibits to establish retrospective actual dates and flags missing advance notices and other news. The training period is 2022-24 and the holdout period is 2025-2026 Sep 8. Fees are 10 basis points and slippage is 10 basis points each side, with stress slippage at 30 basis points. The baseline is all events, and diagnostic groups include unoptimized prior SMA50, prior 20 momentum, and SPY 50 groups. The report notes that no new positions will be taken unless independent net edge and execution are proven, and new symbols are chosen by business type, not prior returns.

A news scan found a question tree and frozen electrical bottleneck test. Prior thoughts were that electrical bottlenecks may reward manufacturers more than distributors, but positive demand may already be priced. This prior thought is set aside. A three-level tree is considered: A Cash capture, which includes scarce components, contracts, and cash conversion. B Market payoff is also considered, which includes continuation, regime, and robustness. C Constraints include capacity, implementation, and falsification. D Outer ring includes history, structure, and consequences. The resolution is to consider daily stock swing and discard headline counts and hash changes as events. Before data analysis, a generic continuation rule is applied: if the close is greater than the previous 20 close high and the close is greater than the SMA50, then enter at the next open and exit after 5 sessions at the open, with no overlap. The training period is from 2022 to 2025, and the untouched period is from 2026 through September 8. A training sweep is done with a lookback of 10, 20, and 40, and a holding period of 3, 5, and 10. The vectorbt fee is 0.1% and slippage is 0.1% on each side, with a stress slippage of 0.3%. The baseline and no-SMA ablation are descriptive, with no holdout retuning. Bootstrap trade-mean CI and MC20 trade exposure scenarios are also considered, plus daily historical stop and target-touch brackets. No trade is made unless the net edge is credible and execution qualifies. The ASPI114 and FRO115 are reviewed separately, and existing rejected rules are unchanged.