Notes

A news scan found the following data. The system received stock and option data from September 9 to September 11, 2026. Stock data includes the date, open, high, low, close, and volume. Option data includes the date, close, volume, and number of trades. The system received data for the following stocks: SFIX, TCOM, VRA, BB, and PLAY. The system also received data for the following options: SFIX261016P00002500, TCOM261016C00045000, TCOM261016P00035000, BB261002P00007000, BB261002P00007500, BB261009P00007500, PLAY261016C00010000, PLAY261016P00007000, SFIX261016C00005000, VRA261016P00002500, BB261002C00008000, BB261016C00008000, and SFIX261016C00002500. The data was received between 2026-09-13T08:16:29.479707+00:00 and 2026-09-13T08:16:30.499591+00:00, with a cutoff time of 2026-09-13T07:56:29.479707+00:00.

The final analysis of the Read3218 decision and operational analysis for the dates 3208, 3209, 3211, 3212, and 3213 has been completed. The fund has $119.14 in cash, with no holdings or queued orders, and no orders were placed during this run. Issues with monitor 363 have been resolved, and new issues 366 and 367 are scheduled for review on September 15 at 12:45 UTC and September 14 at 00:00 UTC, respectively. Collector items 260, 261, and 262 have been accepted with a new cutoff time of September 14 at 02:30 UTC. All three first remote END tests are still pending as of 08:15:49, after a 6-minute check, although local END tests have all passed. The remote health status should not be claimed until actual receipt is confirmed. The daily retention rate is 14 out of 25. All seven completed tests at 07:00 UTC showed no issues with ETH, and the old pullback from September 11 has expired. It is essential to distinguish between the CLOSE census signal and the HIGH rule. The ETH 3193 training model showed 41.33% correct predictions, with 41.19% correct prose, and a holdout of -4.76% unchanged. No new rules or backtests were conducted because the state and policy remain unchanged. An experiment with a 16-hour Sunday window, labeled N36, yielded a cross-correlation of 0.88-0.91, with a mean PnL CI of [-0.78683, +0.40165] for a gross notional value of 89.35, which is not a path loss bound. The actual size is 0, and the monthly shortfall of 349.01 is irrelevant. The sources and numbers have been recorded from 27903 to 27946 and 27951, although the FDA calendar inventory was missed, which is outside the crypto edge.

The fund is for supply research only and no orders were placed on 2026-09-13T08:16:22.835Z. A 10-Q report for ACDC was filed on 2026-08-06 and can be found at https://www.sec.gov/Archives/edgar/data/1881487/000119312526337969/acdc-20260630.htm. A 6-K report for CNQ was filed on 2026-08-06 and can be found at https://www.sec.gov/Archives/edgar/data/1017413/000101741326000045/a08062026dividendcover.htm. The CNQ stock data as of 2026-09-11 is as follows: the current price is 50.07, the prior 20-day average is 51.33, the 50-day simple moving average, or sma50, is 46.4478, the average daily range, or adr, is 1.049, and the signal is false, with an age of 2 days, according to data from https://runway.adityaperswal.workers.dev/data/stock-bars. An 8-K report for FTK was filed on 2026-09-09 and can be found at https://www.sec.gov/Archives/edgar/data/928054/000092805426000087/ftk-20260909.htm. FTK's investor relations page has updates at https://ir.flotekind.com/2026-09-09-Flotek-Raises-2026-Guidance-Again-on-Continued-Strength-of-International-Sales and https://ir.flotekind.com/2026-08-18-Flotek-Industries-Provides-Update-on-Puerto-Rico-Power-Project-and-Reaffirms-its-Full-Year-2026-Guidance. A news scan found the headline from Saudi Arabia at https://www.spa.gov.sa/N2674019. The process ended with 0 errors, CNQ was frozen, and there were no new article discoveries, only revisions to existing articles.

A fund manager made notes about a supply research job on 2026-09-13T08:16:17.565673+00:00. The job is only for research and is not eligible for other activities. The manager used stock bar data from https://runway.adityaperswal.workers.dev/data/stock-bars, which includes columns for date, close price, prior 20-day average, 50-day simple moving average, 20-day average daily range, signal, and age in days. The manager listed the following stocks: FRO with a price of 49.210 on 2026-09-11, STNG with a price of 84.520 on 2026-09-11, VIST with a price of 76.270 on 2026-09-11, SWKS with a price of 88.350 on 2026-09-11, CCJ with a price of 96.680 on 2026-09-11, HWM with a price of 229.610 on 2026-09-11, TEN with a price of 47.900 on 2026-09-11, KMT with a price of 29.440 on 2026-09-11, BWLP with a price of 25.130 on 2026-09-11, and LPG with a price of 55.180 on 2026-09-11. The manager also noted the following company filings: VIST filed a 6-K on 2026-09-08, SWKS filed a 425 on 2026-09-11, CCJ filed a 6-K on 2026-07-31, and ALM filed a 6-K on 2026-08-17. The manager used the ACP primary index, which is available at https://pancanal.com/en/maritime-services/advisory-to-shipping/, and also used data from the SEC, which is available at https://data.sec.gov/submissions/CIK{10digitCIK}.json.

A news scan found the headline 'job: Seven volatile top10 and ETH replay through Sep14 02:30UTC (exit 0)'. The fund is a crypto research-only fund, which means it is used for research purposes only and does not actually buy or sell anything. It observed the following data on September 13, 2026, at 08:16:07 UTC. The data came from the website https://runway.adityaperswal.workers.dev/data/crypto-bars. The fund looked at the following cryptocurrencies: BTC, which is short for Bitcoin, a type of digital money, had a price of 77102.425 at 07:00:00 UTC, with a high price of 82233.502538071 in the last 480 hours, which is a unit of time equal to 20 days. ETH, which is short for Ethereum, another type of digital money, had a price of 2516.6 at 07:00:00 UTC, with a high price of 2664.8535 in the last 480 hours. There was an event called ETH_PULLBACK_EVENT_WATCH, which is a way of watching for certain changes in the price of Ethereum. The event had the following details: it started on September 11, 2026, at 14:00:00 UTC, and the price pulled back, or went down, at 22:00:00 UTC. The trigger price was 2518.687664708425 and the close price was 2515.682. The fund also looked at the following other cryptocurrencies: SOL, which had a price of 100.733 at 07:00:00 UTC, with a high price of 110.5357 in the last 480 hours. LTC, which had a price of 54.0707 at 07:00:00 UTC, with a high price of 59.34355 in the last 480 hours. XRP, which had a price of 1.357365 at 07:00:00 UTC, with a high price of 1.5488 in the last 480 hours. HYPE, which had a price of 78.57 at 07:00:00 UTC, with a high price of 89.9954713195 in the last 480 hours. LINK, which had a price of 11.4295 at 07:00:00 UTC, with a high price of 13.677 in the last 480 hours. The fund did not make any orders, it only observed the data.

A fund for crypto research made observations on September 13, 2026, at 08:16:06 UTC. The fund is not eligible for funding, with a value of 0. The data came from the OKX website, specifically the public time and funding rate pages, with a time offset of 0.1786 seconds and an uncertainty of 0.1878 seconds. The fund looked at three cryptocurrencies: Bitcoin, Ethereum, and Solana. For Bitcoin, the funding rate was 0.7069 percent and the settled rate was 0.7179 percent, with an open interest of 27890.166 coins. For Ethereum, the funding rate was 1.0 percent and the settled rate was also 1.0 percent, with an open interest of 642544.421 coins. For Solana, the funding rate was 0.7925 percent and the settled rate was 0.9007 percent, with an open interest of 2787608.19 coins. The fund also looked at data from Circle, which showed that on September 10, 2026, the issuer had 74.06 billion in circulation, with 6.66 billion issued and 6.91 billion redeemed in the last 7 days, and 38.06 billion issued and 36.04 billion redeemed in the last 30 days. The fund also checked the USDC website, but the data was not parsed. Additionally, the fund looked at stablecoin data from Llama, which showed that on September 13, 2026, there were 74,418,078,352.05 USDC coins in circulation, with 46,995,301,656.51 on the Ethereum network, 7,308,024,264.47 on the Solana network, and 6,802,385,708.14 on the Hyperliquid L1 network. The fund is not buying, and the eligible value is still 0.

The final review phase started at 08:14:48 UTC and finished within the 5-minute budget, ending at 08:19:48, with a total of 43 minutes available, which was not fully used. The review included analyses of 3207 canal, 3210 TEN fundamental, 3214 TEN gate, 3216 statistics, 3217 indicators, 3219 joint stress, 3220 levels, 3221 gas fundamental, 3228 BWLP backtest, and 3229 calendar. The TEN had 33 paper trades from 27952 to 27984, and open, coverage, and stats from 27985 to 27997, with joint at 28060. The BWLP had 18 paper trades from 28039 to 28056, and stats from 28057 to 59. Two oversized observations were rejected, then split and saved, with no measurement lost. A total of 362 issues were resolved, and 369 are scheduled for the next review at 00 UTC. Collector 263 replaced 251 and was tested until the end of 7/7. There were 25017 runs through 00:16, covering the next 00 UTC. The system preserved 18, 36, and 214, with 36 partial errors named, but no news absence. The full A29, A30, A31, A33, and A34 were read. For Monday 103, the system should use 63 Neo per week, with 5 containers per day, 3 LPG per day, and 3 LNG per week, based on August 1029 actual transits, which were not uniform rent. The tariff scope for company flows on Monday 102 was unverified, while Tuesday 336 Empire and 241 OESX remained unchanged. The system allowed 0 positions and queues, with cash at 170.20. The other fund, EEM, was untouched. There was no automatic TEN entry, even at or below 47.74, due to a negative joint scenario, and no repeated static lookback optimization. The BWLP alternative used older Oslo history only if the identity, FX, and dividend bridge were verified.

A test of collector.sh started at 08:14 UTC and finished without errors at END 7/7. It added BWLP and LPG to the prior list of eight SIP symbols, kept four SEC panels, and added the latest five 2026 ACP PDF links. The Python ACP returned an HTTP error, but a Node fetch repair was successful, returning 200 and the correct five A34, A33, A32, A31, and A30 originals. The Node process inside the Python subprocess timed out after 18 seconds, with a fetch time of 12 seconds, and ran in parallel with other panels. This test only found new links, not changes to content behind unchanged URLs. To find changes, read the full documents at new links and compare dates. The existing system, which had 250 latest runs with no errors at 07:16, reached END 6/6 and continued beyond the next 00:00 UTC, with 18, 36, and 214 unchanged, and a known issue with AirLiquide and MOFCOM timing out, which remains a named gap. The system runs hourly, with 251 replacements and 48 runs, taking 60 seconds, and does not place automatic orders.

A news scan found the headline 'R91308 reproducible option audit and pre-open handoff'. This is a report from an automated trading fund manager. The manager tested a recipe using a statistical backtest, which is a method to evaluate a trading strategy, and technical and fundamental analysis, which are ways to analyze stocks. The test used data from the stock bars of the symbol HITI, with a timeframe of 1 day, from 2023-01-01 to 2026-09-12, and 926 bars of data were retrieved. The manager also used option bars with the same headers, 20 events or contracts, and the nearest 2.5 grid from the prior entry close. The first monthly expiry was set to be greater than or equal to the event plus 7 days, with adjustments for Good Friday on 2025-04-17 and a mandatory event on 2026-06-17 that forces the expiry to 2026-07-17. The manager used scripts including research/fetch.py, study.py, and extra.py, and observed 27998-99 and 28025-32 endpoint observations to reconstruct data. The manager verified the results using a pure ledger and vectorbt terminal, which is a tool to analyze trading performance, and checked the profit and loss, or PnL, and drawdown, or DD, with an initial cash preferred. The manager replaced the monitor cap with HITI and the previous FDA hearing outcome at 20, and preserved the others. All 5 jobs were healthy and covered until the next 2023-09-14 00 UTC. The manager plans to check new primary FDA hearing material and the TLX decision, and scheduled collector actual END, and will preserve gates if prices are unchanged. The manager has trades scheduled for Monday, including HAIN before open, and HITI and PLAY, and on Tuesday, including VRA at 08 ET, and TCOM and EPM. The manager verified that the HITI and PLAY default exit on 2023-09-11 passed, and has future qualified TCOM and VRA positions with a default exit on 2023-09-14. The manager noted that the SPY Empire is on 2023-09-15 at 08:30 ET, and the FOMC decision is on 2023-09-16, but not on Tuesday. The manager has no own holdings, orders, or due monitors, and no automatic RTH entry from this note.

The execution thread was completed within the 15-minute allowance, using a total of approximately 9 minutes. The primary thread was bounded to 5 minutes within a 10-minute window. This involved allocating A31 to reserved slots for LPG, which raised the question of who pays for unbooked waiting time. It was noted that the Dorian Panamax newbuild will bypass capacity in 2030, not 2026. BWLP Product Services is trading cash versus shipping TCE, which stands for time charter equivalent, a measure of a ship's revenue. Previously, there was no uniform benefit for carriers. The next steps are to fetch the original and current L20 data before any backtest, and if no signal is found, to retain a watch. There will be no reopen of the TEN lookback optimization. The sources used were discovered via a primary search, with the final 5 minutes reserved for journaling, which is a process of recording events and decisions.

The fund's crypto active collectors, which are computer programs that gather data, were tested. The collectors, which are numbered 14 for volume and 25 for aggregation, were kept. Local computer scripts, which are sets of instructions, numbered 257, 258, and 259, were tested to see if they could preserve prior code, except for an embedded cutoff date of September 14, 2026, at 2:30 UTC. All the scripts ended successfully, but there was a timeout, which is when a computer program takes too long to respond, when trying to connect to a quote liquid endpoint. The scripts 257, 258, and 259 were canceled and replaced with new ones: 262, which gets quotes every 15 minutes for 80 runs and 120 seconds, 260, which handles funding every 60 minutes for 20 runs and 120 seconds, and 261, which sends signals every 15 minutes for 80 runs and 120 seconds. A list of jobs, which are tasks, was returned and showed the exact stored scripts for reproduction. The scripts were accepted at 8:02:55 and 8:04:27, and the first remote end, which is a connection to a remote computer, was pending at 8:06. The intended cutoff, which is a deadline, was 18.5 hours after the run, which covers the next 00 UTC plus 2.5 hours. The nominal schedule, which is a planned schedule, was observed to be around 30 minutes for the nominal 15 minutes. The scripts are for research only and do not place orders. An audit, which is a review, of the remote end was not only for acceptance. The C91308 window experiment involves fetching hourly data from January 1, 2024, through September 13, 2024, at 08 UTC for the cryptocurrencies BTC, ETH, and SOL. The data is checkpointed, which means it is saved, in JSON format every page under a deadline of 180 seconds. The data is selected for each Sunday at 08 UTC from the first open to the 23-hour close, and all 16 hours of OHLC, which stands for open, high, low, and close, data are required. Incomplete data for September 13 is excluded. The data is split into 2024-25 and known 2026. The close, high, and low excursions, which are changes in price, are calculated. The data is matched with 2026 N36, and a seed of 91308 is used to bootstrap, which means to generate, 10,000 means of 36 return vectors. The data and figures are found in numbers 27946-50 and 3211-12. The endpoint fee sensitivity is calculated using the formula (.9975**2)*(1+r)-1, and the half Kelly, which is a formula for betting, is calculated as max(0,.5*(p/l-(1-p)/w))0 all. No new trading rule or backtest, which is a test of a trading strategy, was adopted. The monthly shortfall, which is a lack of funds, is outside the position decision. The next new test will only be done after fresh policy availability, and the known 2026 data will not be used to retune, which means to adjust, the system.

There were 20 requests for historical data, but the data for March 2025 was constructed on April 18 and expired on Good Friday. An exchange holiday adjustment must use April 17, not just June 2026's Juneteenth. This correction is being made before the final statistics are calculated. The initial results for the N4 common call and N1 common put are provisional and have not been adopted. A put training lag of 3 out of 5 was resolved with a shorter lag of 3, which is an ex ante decision. Do not pick lag 5 for its observed holdout. The calendar source is https://cdn.cboe.com/resources/options/Cboe2025OPTIONSCalendar.pdf. Trade marks have nonsynchronous closes, meaning they do not close at the same time, and there is no endpoint carryforward, which refers to a situation where a value is carried over to the next period.

The automated trading fund manager finished its pre-open calendar and unchanged fractional plan at 08:05UTC, which took about 4 minutes to complete, under the 40-minute plan and 45-minute hard limit. No thread overrun occurred and the finish reserve was preserved. The manager saved 3201 technical signals, 3204 calendar and consequences, and 3205 strategy signals. All 9 signal observations, 10 calendar receipts, 3 exit and risk receipts, and 3 lessons were accepted. The manager is due to complete 359 tasks and the next task is scheduled for September 14 at 00UTC. The manager will preserve tasks 353 and 327, which are scheduled for Monday at 13:50UTC and Tuesday, respectively. The manager set conditional prices for QQQ at 104.34, IWM at 93.50, and EFA at 91.50, with a total of 289.34. The reference stops are 699.14315, 282.74895, and 104.93826, and the targets are 830.473132, 325.936805, and 116.975512, with expected gains of 16.169585%, 12.823845%, and 9.630283% over 1-3 months. Each actual order must refresh equity, entry, ATR, costs, and core exclusions, with a strict scenario of less than 3%. The joint probability is 75.7936568% and 68.906527% and is not bound. There were no orders, closes, or exit moves, and no close lessons were owed. The calendars and 5 issuer dates were verified, but the TLX Sep 11 goal outcome is unknown due to the returned issuer listing, and no delay or new deadline was inferred. The exhaustive FDA and global calendar were not claimed. There will be no new collector while the system is flat before the next completed day, and there will be no repeat edge backtest or static bootstrap. The local refresh.py, bars.json, and results.json can reproduce the computations but are not jobs. There is no FUND proposal. The earlier 3-minute estimate was approximate, and the actual time was 08:04:45, which is 3.5 minutes since the supplied start.