Notes

A news scan found the text: Source https://runway.adityaperswal.workers.dev/data/crypto-bars; local /tmp/codex-run-Yk0cow/fetch.py fully paginates 200 page guard, 3 workers; study.py vectorbt cash reconciled 48 cells; simulate.py seed 91106. Matrix observations 24627–24674, backtests 2602/2604/2605/2606/2608/2609; statistics/simulation 24796–24828; regime cells 24852–24875; analyses 2669–2671/2675/2676. Daily and hourly 2026 cutoff 00 UTC, ETH hourly 17.2404388093 reproduces original 6 ledger (note 2200). Native SMA warmup unknown excluded only from attribution, all backtest trades retained. HAC, which means Heteroskedasticity and Autocorrelation Consistent, 7 regression is a separate daily-outcome specification on the same prices, not statistically independent validation; wording independent daily outcome in 2669 means separately computed, not new holdout. Hourly lag 1 diagnostic uses dropna, so a few missing hour boundaries can join; not actionable autocorrelation edge. Download counts BTC 985 daily/23619 hour, ETH 985/23618, SOL 747/17888 includes Sep 11 partial end, accepted study 984/984/746 completed day rows. All source hashes recorded 24955–24967. API observation note limit 2000; two oversized simulation group writes rejected before persistence, retried as individual verified records. No orders. Monitor handoff pending final receipt check.

A news scan found the headline about a crypto funding and supply update to 08:00 UTC, which replaces a previous update from 20:08 with an exit code of 0. The crypto research fund observed data on September 11, 2026, at 06:15:23.895412 UTC. The fund used data from https://www.okx.com/api/v5/public/time with an offset of 0.0635 seconds and an uncertainty of 0.153 seconds. The fund also used data from https://www.okx.com/api/v5/public/funding-rate and https://www.okx.com/api/v5/public/open-interest. The data included the following information for three cryptocurrencies: BTC, ETH, and SOL. For BTC, the funding rate was 0.7572 basis points and the settled rate was 0.876 basis points, with an open interest of 29,470.944 coins. For ETH, the funding rate was 0.3269 basis points and the settled rate was 0.3053 basis points, with an open interest of 598,355.535 coins. For SOL, the funding rate was 0.7221 basis points and the settled rate was -0.0697 basis points, with an open interest of 2,743,277.11 coins. The data also included information from https://www.circle.com/transparency, which showed the following values for a specific issuer on September 7, 2026: 74.28 for circulation, 8.8 for 7-day issues, 7.86 for 7-day redemptions, 36.59 for 30-day issues, 34.37 for 30-day redemptions, 333.52 for 365-day issues, and 331.64 for 365-day redemptions. Additionally, the fund used data from https://www.usdc.com/ and https://stablecoins.llama.fi/stablecoin/2, which showed a raw CMS value of 74,160,792,259.19 and a circulating value of 74,212,069,483.13, respectively. The fund is not buying and has an eligibility status of 0.

A news scan found the following data line: fund=crypto research-only eligible=0 observed=2026-09-11T06:15:21.667219+00:00 src=https://runway.adityaperswal.workers.dev/data/crypto-bars. The data shows observations for three cryptocurrencies: BTC, ETH, and SOL. For each, the data includes the symbol, hour of observation, closing price, prior 480-hour high, and whether the 480-hour period is complete. The results are as follows: for BTC, the close was 77229.185, prior 480-hour high was 82233.502538071, complete480 was true, signal was false, and eligible was false, with a note that the frozen rolling threshold was the same and a quote/cost/exit and sizing review was required. For ETH, the close was 2467.145, prior 480-hour high was 2565.9, complete480 was true, signal was false, and eligible was false, with the same note. For SOL, the close was 99.8787, prior 480-hour high was 110.5357, complete480 was true, signal was false, and eligible was false, with the same note. The data ends with the statement 'END fund=crypto entry watch; observations only, no orders'.

An audit found that early 2024 and first SOL training entries happened before the native 50-day Simple Moving Average and 55-day 5-day slope warmup. When comparing price to Not a Number, it yields False, which should not be called a downtrend. All baseline trades and returns will be preserved, but unavailable feature labels will be excluded from descriptive attribution, including trend for the first 50 days, slope 55, and RV21, and the excluded counts will be recorded. Regression already requires a full 50-day Simple Moving Average, which is a calculation of the average price over 50 days, and 20-day RV. This is a matter of accounting for feature availability, and it does not mean that early trades are being dropped from the backtest. An unknown regime cannot be used as evidence for or against a filter, and no new entry rule has been selected. The source of this information is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the local study in the file localstudy.py has been corrected before accepting regime observations.

The C91106 prespecified controlled execution and regime experiment implements a strategy with a frozen rolling 480-hour high and 240-hour low, completed 24-hour reconstructed daily Average True Range (ATR) of 14 days multiplied by 2. This is compared to a native daily comparator of 20-day high and 10-day low and native ATR. A signal is generated after the completion of the next exact open, but if the exact open is missing, the signal is skipped, with a 20-day timeout. The experiment also includes 25 basis points credited-asset fees and 10 basis points lip baseline, with a 20 basis points stress target fixed at baseline net 3R formula k = 0.9975^2 * 0.999 / 1.001. The trailing arms are raw price plus 2 times initial risk, raising the stop to cost break-even and sampled peak minus 2 entry ATR, with the target retained. A static control separates the trail effect, without new selection. The data is trained and reused from 2024 to 2025 and held out from January 1, 2026, to September 11, 2026, with 10 and 20 basis points lip and static/trail, resulting in 48 cells. The entry regimes use prior native UTC day price greater than Simple Moving Average (SMA) 50, slope 5 days, and 20-day annualized Realized Volatility (RV) relative to the 2024-2025 median, which are descriptive partitions only, with no selection of filters from 2026. The reconciliation is explicit, using vectorbt with initial cash of 1.0, cash-only, long-only, buy fee of 0.0025, and sell fee of 0.0025, with raw wealth and closed returns multiplied by 0.9975, which must match 1e-10. The data source is https://runway.adityaperswal.workers.dev/data/crypto-bars. A separate 730-day breakout is run using the command-line interface, which is a gross same-close diagnostic, not a replication of net 3R.

The C91106 experiment has a 43-minute plan with a hard cutoff at 06:45:29 UTC, and the last 5 minutes are reserved. The prior experiment on Ethereum, which was watched for 480 hours, showed positive results on only 6 reused holdout trades, with no new entries. The least predictable aspects of the experiment include measurement of execution and asset fees, exact entry and exit times, and coverage of missing hours. The selection process also has uncertainties, such as validation of unused data from 2027 and parameter multiplicity. The experiment is divided into four parts: regime and trend analysis, including price and SMA50 daily trends, volatility distribution, and payoff interactions; allocation and payoff analysis, including win/loss distribution and correlation between Bitcoin, Ethereum, and Solana; continuity and catalysts, including numeric endpoints and primary catalysts like CPI; and decision premises, including current entry and net target. The experiment is resolved daily and hourly, with 15-minute sampling, and does not use tick alpha or headlines as entry inputs. The diagnostic regime uses fixed SMA50 and training median realized volatility, and does not optimize on 2026 data. No trade is made unless a frozen setup and executable payoff exist.

A news scan found the headline is not available, but a data line from a fund manager says the crypto research-only fund made observations on September 11, 2026, at 05:45:27.355279 UTC. The fund watched three symbols: BTC, ETH, and SOL. The data for each symbol is as follows: BTC closed at 77090.0985 at 04:00:00 on September 11, 2026, with a prior 480-hour high of 82233.502538071, and the signal was false. ETH closed at 2459.12 at the same time, with a prior 480-hour high of 2565.9, and the signal was also false. SOL closed at 99.67, with a prior 480-hour high of 110.5357, and the signal was false. All symbols had a complete 480-hour period, but were not eligible for entry, with a note saying the frozen rolling threshold was the same and a quote, cost, exit, and sizing review was required. The fund made no orders, only observations.

A news scan found the following data: the fund is for crypto research only and was observed on 2026-09-11T05:45:26.786528+00:00. The data sources are https://runway.adityaperswal.workers.dev/data/crypto-snapshots and https://blockstream.info/liquid/api/blocks. The fund has the following quotes: BTC bid/ask price is 77125.112/77149.75 with an age of 2 and a depth in USD of 77.7/77.8 with a hurdle percentage of 0.534, ETH bid/ask price is 2462.98/2463.33 with an age of 94 and a depth in USD of 82.0/82.3 with a hurdle percentage of 0.516, SOL bid/ask price is 99.74/99.819 with an age of 5 and a depth in USD of 94.9/95.0 with a hurdle percentage of 0.581, USDC bid/ask price is 0.9995/1 with an age of 3938 and a depth in USD of 9995000.0/45999.0 with a hurdle percentage of 0.552, USDG bid/ask price is 0.99999/1.00001 with an age of 93205 and a depth in USD of 9999900.0/10000100.0 with a hurdle percentage of 0.504, XRP bid/ask price is 1.346/1.3516 with an age of 238 and a depth in USD of 5572.7/5621.1 with a hurdle percentage of 0.920, UNI bid/ask price is 6.119/6.12819 with an age of 2 and a depth in USD of 4281.0/4329.3 with a hurdle percentage of 0.653, LTC bid/ask price is 52.64/52.939 with an age of 5 and a depth in USD of 3531.1/3615.7 with a hurdle percentage of 1.073, HYPE bid/ask price is 79.58/79.982 with an age of 39 and a depth in USD of 61455.4/61826.1 with a hurdle percentage of 1.010, and LINK bid/ask price is 11.5284/11.55 with an age of 4 and a depth in USD of 13404.8/13493.1 with a hurdle percentage of 0.690. There are 8 out of 10 fresh quotes. The recipient BTC is 598.50048115 with a mempool net of 0.00015061 and a transaction number of 296. The reserve BTC is 3601.47203772 with a mempool net of 3.384e-05 and a transaction number of 590. The recent page is 57, controlled is 0, and not full history is 1. The tip is 4052278 with an age in seconds of 17 and a time of 1789105510. There are 3 operator notes and the operator hash is 1e913157cc811e629b1b86d4db894412f924f07ea0928c7aa1a3e183681de0ca with a claim to those responsible for the theft of bitcoin from the Liquid Network. The fund has 7 out of 7 panels attempted and 0 eligible, with balances not sales or recovery.

A news scan found the following data line: fund=crypto research-only eligible=0 observed=2026-09-11T05:30:20.758022+00:00 src=https://runway.adityaperswal.workers.dev/data/crypto-bars. The data shows observations for three cryptocurrencies: BTC, ETH, and SOL. For each, the data includes the symbol, hour of observation, closing price, prior 480 hour high, whether the 480 hour period is complete, a signal, eligibility, and a note. The notes for all three cryptocurrencies mention a frozen rolling threshold and the need for a review of quote, cost, exit, and sizing. The observations are as follows: BTC closed at 77090.0985 at 2026-09-11T04:00:00+00:00 with a prior 480 hour high of 82233.502538071, ETH closed at 2459.12 with a prior 480 hour high of 2565.9, and SOL closed at 99.67 with a prior 480 hour high of 110.5357. All three had a complete 480 hour period, but no signal and were not eligible. The data ends with the statement END fund=crypto entry watch, indicating that these are observations only and no orders were placed.

A fund for crypto research was observed on 2026-09-11T05:30:23.057137+00:00. The data came from two sources: https://runway.adityaperswal.workers.dev/data/crypto-snapshots and https://blockstream.info/liquid/api/blocks. The fund looked at the following cryptocurrencies: BTC, ETH, SOL, USDC, USDG, XRP, UNI, LTC, HYPE, and LINK. For each cryptocurrency, the fund noted the bid and ask prices, the age of the data in seconds, and whether the data was okay. The bid and ask prices are the prices at which someone is willing to buy or sell a cryptocurrency, which is also known as a market quote. The age of the data refers to how old the information is. The fund also calculated a hurdle percentage, which is a measure of how much the price of a cryptocurrency would need to change in order to reach a certain level. The hurdle percentage is calculated as a percentage of the current price. The results were as follows: BTC had a bid price of 77044.883 and an ask price of 77070.01, with an age of 12 seconds and a hurdle percentage of 0.535. ETH had a bid price of 2458 and an ask price of 2458.33, with an age of 167 seconds and a hurdle percentage of 0.515. SOL had a bid price of 99.55 and an ask price of 99.686, with an age of 168 seconds and a hurdle percentage of 0.639. USDC had a bid price of 0.9995 and an ask price of 1, with an age of 3035 seconds and a hurdle percentage of 0.552. USDG had a bid price of 0.99999 and an ask price of 1.00001, with an age of 92301 seconds and a hurdle percentage of 0.504. XRP had a bid price of 1.34448 and an ask price of 1.35, with an age of 102 seconds and a hurdle percentage of 0.915. UNI had a bid price of 6.139 and an ask price of 6.15184, with an age of 9 seconds and a hurdle percentage of 0.712. LTC had a bid price of 52.6576 and an ask price of 52.9773, with an age of 159 seconds and a hurdle percentage of 1.112. HYPE had a bid price of 79.3 and an ask price of 79.69, with an age of 35 seconds and a hurdle percentage of 0.996. LINK had a bid price of 11.51 and an ask price of 11.53, with an age of 28 seconds and a hurdle percentage of 0.677. The fund received 8 out of 10 fresh quotes. The recipient of a BTC transaction had 598.50048115 BTC, with a mempool net of 0.00014572 and a transaction number of 296. The reserve had 3601.47203772 BTC, with a mempool net of 3.384e-05 and a transaction number of 590. The recent page was 56, and it was not a full history. The tip was 4052263, with an age of 13 seconds and a time of 1789104610. There were 3 operator notes and 1 operator hash, which is a unique code used to identify a transaction. The operator hash was 1e913157cc811e629b1b86d4db894412f924f07ea0928c7aa1a3e183681de0ca, and it included a claim about a theft of bitcoin from the Liquid Network. The message claimed that Blockstream will not pay a ransom. The fund attempted 7 panels and found 0 eligible, with balances but no sales or recovery.

A news scan found the headline about a crypto funding and supply update to 06:00 UTC, which replaces the previous update 202 with an exit code of 0. The fund is for crypto research only and was observed on September 11, 2026, at 05:15:20.572418 UTC. The source of the information is https://www.okx.com/api/v5/public/time with an offset of 0.1068 seconds and an uncertainty of 0.0978 seconds. Another source is https://www.okx.com/api/v5/public/funding-rate and https://www.okx.com/api/v5/public/open-interest. The data shows the following information for Bitcoin, or BTC, which is a type of cryptocurrency: the timestamp is 1789103710592 and 1789103721293, the age is 10.8 and 0.1, the data is fresh, the funding rate is 0.8158 basis points, the settled rate is 0.876 basis points, and the open interest in coins is 29634.173. For Ethereum, or ETH, which is another type of cryptocurrency, the timestamp is 1789103710642 and 1789103721286, the age is 10.7 and 0.1, the data is fresh, the funding rate is 0.4023 basis points, the settled rate is 0.3053 basis points, and the open interest in coins is 595428.015. For Solana, or SOL, which is a blockchain platform, the timestamp is 1789103711578 and 1789103721288, the age is 9.8 and 0.1, the data is fresh, the funding rate is 0.6357 basis points, the settled rate is -0.0697 basis points, and the open interest in coins is 2734171.44. The source https://www.circle.com/transparency provides information on the issuer date of September 07, 2026, and the following data: bn circ is 74.28, issue7 is 8.8, redeem7 is 7.86, issue30 is 36.59, redeem30 is 34.37, issue365 is 333.52, and redeem365 is 331.64. The source https://www.usdc.com/ provides a raw cms value that is unparsed and an issuer date that is also unparsed due to locale and template ambiguity. The source https://stablecoins.llama.fi/stablecoin/2 provides data on the circulating supply of a stablecoin, which is a type of cryptocurrency pegged to a stable asset, with a timestamp of 1789084800, a circulating supply of 74212069483.12572, and the following data for Ethereum, Solana, and Hyperliquid L1: the timestamp is 1789084800, the value for Ethereum is 47184685905.99, the value for Solana is 6987674198.11, and the value for Hyperliquid L1 is 6839110360.24. The crypto collection has ended, and the definitions and cutoffs differ, with the flow not buying and the eligible value being 0.

A news scan found the headline 'job: Daily aggregation integrity; research only (exit 0)'. The fund is for crypto research only and is not eligible. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The data includes information for three symbols: BTC, ETH, and SOL, all for the day 2026-09-10. For BTC, the daily close was 76535.875, the daily high was 78530.3175, the daily low was 76450.69, and the volume was 2.618561963. For ETH, the daily close was 2436.8075, the daily high was 2483.6235, the daily low was 2405.3715, and the volume was 21.67071483. For SOL, the daily close was 98.6025, the daily high was 102.145, the daily low was 98.5760837435, and the volume was 295.652469985. The hour aggregate and difference percentage are also provided for each symbol. The quality pass is true for all symbols, and the eligible status is 0. The coverage is 3/3 for quality pass, and the eligible status is 0.

A fund for crypto research was observed on 2026-09-11T05:00:50.820350+00:00. The data source is https://runway.adityaperswal.workers.dev/data/crypto-snapshots and https://blockstream.info/liquid/api/blocks. The fund had the following quotes: BTC had a bid of 77078.264 and an ask of 77098.37, with an age of 32, and was considered okay with a depth in USD of 76.7/76.7 and a hurdle percentage of 0.528. ETH had a bid of 2458.85 and an ask of 2459.39, with an age of 212, and was considered okay with a depth in USD of 82.7/81.7 and a hurdle percentage of 0.524. SOL had a bid of 99.64 and an ask of 99.7, with an age of 148, and was considered okay with a depth in USD of 94.7/94.4 and a hurdle percentage of 0.562. USDC had a bid of 0.9995 and an ask of 1, with an age of 1263, and was not considered okay with a depth in USD of 9995000.0/45999.0 and a hurdle percentage of 0.552. USDG had a bid of 0.99999 and an ask of 1.00001, with an age of 90529, and was not considered okay with a depth in USD of 9999900.0/10000100.0 and a hurdle percentage of 0.504. XRP had a bid of 1.345 and an ask of 1.35062, with an age of 141, and was considered okay with a depth in USD of 5584.0/5636.5 and a hurdle percentage of 0.922. UNI had a bid of 6.1579 and an ask of 6.1735, with an age of 30, and was considered okay with a depth in USD of 4323.0/4391.2 and a hurdle percentage of 0.756. LTC had a bid of 52.658 and an ask of 52.972, with an age of 39, and was considered okay with a depth in USD of 3555.4/3554.5 and a hurdle percentage of 1.101. HYPE had a bid of 79.12 and an ask of 79.4847, with an age of 124, and was considered okay with a depth in USD of 60677.1/61756.1 and a hurdle percentage of 0.965. LINK had a bid of 11.5254 and an ask of 11.5485, with an age of 104, and was considered okay with a depth in USD of 13472.3/13340.5 and a hurdle percentage of 0.703. There were 8 fresh quotes out of 10. The recipient BTC was 598.50048115, with a mempool net of 0.00014572 and a transaction number of 296. The reserve BTC was 3601.47203772, with a mempool net of 3.384e-05 and a transaction number of 590. The recent page was 56, controlled 0, and not full history was 1. The tip was 4052233, with an age in seconds of 41 and a time of 1789102810. The second was also 4052233, with an age in seconds of 41 and a time of 1789102810. There were 2 operator notes and 1 operator hash, which was 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b, with a claim about an incident affecting the Liquid Network and the movement of funds. The fund had 7 out of 7 panels attempted, but was not eligible, and the balances were not sales or recovery.

A news scan found the data line: fund=crypto research-only eligible=0 observed=2026-09-11T05:00:20.481264+00:00 src=https://runway.adityaperswal.workers.dev/data/crypto-bars. The data shows observations for three cryptocurrencies: BTC, ETH, and SOL. For each, the data includes the symbol, hour, close price, prior 480 hour high, whether the 480 hour period is complete, a signal, eligibility, and a note. The notes for all three mention a frozen rolling threshold and the need for a quote, cost, exit, and sizing review. The data is as follows: BTC closed at 77090.0985 at 2026-09-11T04:00:00+00:00 with a prior 480 hour high of 82233.502538071, ETH closed at 2459.12 with a prior 480 hour high of 2565.9, and SOL closed at 99.67 with a prior 480 hour high of 110.5357. All three had complete 480 hour periods, no signal, and were not eligible. The data ends with the note: END fund=crypto entry watch; observations only, no orders.

A fund for crypto research was observed on 2026-09-11T04:45:59.959391+00:00. The data came from two sources: https://runway.adityaperswal.workers.dev/data/crypto-snapshots and https://blockstream.info/liquid/api/blocks. The fund looked at the following cryptocurrencies: BTC, ETH, SOL, USDC, USDG, XRP, UNI, LTC, HYPE, and LINK. For each cryptocurrency, the fund noted the current buy and sell prices, the age of the data in seconds, whether the data was okay, the depth of the market in US dollars, and a hurdle percentage. The hurdle percentage is a measure of how difficult it is to buy or sell a cryptocurrency. The results were: BTC buy/sell=77056.35/77080.7 age=119 ok=1 depthUSD=76.7/77.8 hurdle%=0.534, ETH buy/sell=2455.77/2456.7 age=119 ok=1 depthUSD=82.1/81.4 hurdle%=0.540, SOL buy/sell=99.646/99.7 age=162 ok=1 depthUSD=95.0/93.9 hurdle%=0.556, USDC buy/sell=0.9995/1 age=372 ok=0 depthUSD=9995000.0/45999.0 hurdle%=0.552, USDG buy/sell=0.99999/1.00001 age=89638 ok=0 depthUSD=9999900.0/10000100.0 hurdle%=0.504, XRP buy/sell=1.34432/1.3494 age=165 ok=1 depthUSD=5568.6/5591.2 hurdle%=0.882, UNI buy/sell=6.129/6.1394 age=46 ok=1 depthUSD=4285.6/4349.4 hurdle%=0.672, LTC buy/sell=52.749/53.0673 age=54 ok=1 depthUSD=3552.2/3564.8 hurdle%=1.108, HYPE buy/sell=79.2136/79.58 age=119 ok=1 depthUSD=60877.6/60889.8 hurdle%=0.967, and LINK buy/sell=11.507/11.5263 age=94 ok=1 depthUSD=13381.5/13376.3 hurdle%=0.670. The fund received 8 out of 10 fresh quotes. The recipient had 598.50048115 BTC and the reserve had 3601.47203772 BTC. The recent page was 56 and the tip was 4052218. There was an error with a connection timeout. The operator notes had 2 entries and the operator hash was 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The operator noted that following a recent incident affecting the Liquid Network and the movement of funds, Blockstream made a statement. The fund attempted 7 out of 7 panels and had 0 eligible results, with balances but no sales or recovery.

A news scan found the headline about a job for a frozen rolling 480-hour entry watch until 06:00 UTC for research purposes only, with an exit value of 0. The fund in question is a crypto research-only fund, which was eligible 0 times and was observed on 2026-09-11T04:45:19.545544+00:00, with data from https://runway.adityaperswal.workers.dev/data/crypto-bars. The data includes the following observations: for Bitcoin, the symbol is BTC, the hour is 2026-09-11T03:00:00+00:00, the close price is 76864.9, the prior 480-hour high is 82233.502538071, the 480-hour period is complete, there is no signal, and it is not eligible, with a note that the frozen rolling threshold is the same and a review of quote, cost, exit, and sizing is required. For Ethereum, the symbol is ETH, the hour is 2026-09-11T03:00:00+00:00, the close price is 2445.725, the prior 480-hour high is 2565.9, the 480-hour period is complete, there is no signal, and it is not eligible, with the same note as Bitcoin. For Solana, the symbol is SOL, the hour is 2026-09-11T03:00:00+00:00, the close price is 99.29365, the prior 480-hour high is 110.5357, the 480-hour period is complete, there is no signal, and it is not eligible, with the same note as Bitcoin. The data ends with a note that it is the end of the crypto entry watch, with observations only and no orders.

A news scan found the text: 207 quotes/Liquid replaces cancelled 201; 208 funding/supply replaces done 202; 213 rolling 480 hour watch new, all cutoff Sep 11 06:01 UTC. The first remote actual END was verified at 207 04:16:05.657, 208 04:15:20.535, and 213 04:15:58.800, which is not merely scheduler acceptance. The next steps for 207 and 213 are at 04:30 and then every 15 minutes for 7 left, and for 208 at 05:15 with 1 left. There are 14 daily 36 pair census and 25 daily aggregation still active with 23 left. Prior to 201, there was a 03:31:02 to 04:01:01.750 receipt gap of 29.9952 minutes on a 15-minute schedule, which the collector END does not prove cadence or execution. The latest Liquid quote is 4052188, aged 56 seconds, with second timeout and unchanged balances or hash. The quote for Bitcoin is 76954.146/76974.439, for Ethereum is 2449.2/2450.079, and for Solana is 99.43/99.5329. Eight out of ten quotes are fresh, but USDC and USD are stale. The forecast for 208 is Bitcoin at 0.9721, Ethereum at 0.4557, and Solana at 0.4298 basis points, with open interest at 29805.812, 589029.432, and 2728351.59 coins. Circle on September 7 was unchanged. The latest completed signal for 213 at 03 UTC was all false, with 255 resolved at 26806 UTC review armed, and 181 CPI, 104 Arc, 152 GME, and other future crypto monitors preserved. A local test of the script entry_watch.sh paginates 23 days, with a maximum of 12 pages, and suppresses incomplete 480 bars for research only, with no orders. The gaps are due to the original 2025 bar construction cause, independent forward outcomes, executable quotes, and exact 15-minute trail implementation, with full Liquid recovery evidence. The full primary funding and supply notes context is not alpha. The source URLs are 24539-41 and 2534-41. The research finished in approximately 15 minutes, with 5 minutes reserved for writing, under a total allowance of 43 minutes.

A news scan found the trade ledger for an automated trading fund manager. The source of the data is /data/crypto-bars with a baseline credited fees of 25 basis points and a slip of 10. The seed is 91104. The trading strategy uses a rolling 480 hour and 240 low, 2 ATR 14, net 3R, and trail, with 6 closed trades in 2026. The trades are as follows: from January 5 at 01 UTC to January 20 at 15 UTC, the entry mark was 3168.99 and the exit mark was 3041.30, with a net loss of 4.699389 percent; from March 4 at 16 UTC to March 24 at 16 UTC, the entry mark was 2149.74 and the exit mark was 2133.58, with a net loss of 1.444648 percent; from April 8 at 00 UTC to April 28 at 00 UTC, the entry mark was 2241.05 and the exit mark was 2303.85, with a net gain of 2.084737 percent; from July 6 at 22 UTC to July 26 at 22 UTC, the entry mark was 1814.18 and the exit mark was 1925.45, with a net gain of 5.392441 percent; from August 19 at 15 UTC to August 19 at 22 UTC, the entry mark was 1970.45 and the exit mark was 2286.72, with a net gain of 15.240409 percent; and from August 20 at 17 UTC to September 2 at 10 UTC, the entry mark was 2342.92 and the exit mark was 2375.34, with a net gain of 0.676120 percent. The full daily trades are also listed: from January 6 to January 21, the entry mark was 3228.26 and the exit mark was 2935.59, with a net loss of 9.700616 percent; from March 17 to March 22, the entry mark was 2351.34 and the exit mark was 2083.85, with a net loss of 11.994781 percent; from April 12 to April 30, the entry mark was 2285.40 and the exit mark was 2252.88, with a net loss of 2.110865 percent; from July 15 to August 2, the entry mark was 1891.35 and the exit mark was 1841.80, with a net loss of 3.299708 percent; and from August 20 to September 9, the entry mark was 2252.83 and the exit mark was 2484.12, with a net gain of 9.496585 percent. The window ends on September 11 at 00 UTC. The fees and slip result in a flat price exact loss of 0.69817744755 percent and a breakeven of 0.70308623710 percent. The rounding improves in 2540. The bootstrap uncertainty fields have a p half and half terminal mean fixed at 50 percent notional stress, but do not use the Kelly size for negative policies. There are no orders and the actual notional is 0.

A fund for crypto research was observed on 2026-09-11T04:16:05.657465+00:00. The data came from two sources: https://runway.adityaperswal.workers.dev/data/crypto-snapshots and https://blockstream.info/liquid/api/blocks. The fund looked at the following cryptocurrencies: BTC, ETH, SOL, USDC, USDG, XRP, UNI, LTC, HYPE, and LINK. For each cryptocurrency, the fund noted the current buy and sell prices, the age of the data in seconds, whether the data was okay, the depth of the market in US dollars, and a hurdle percentage. The hurdle percentage is a measure of how much the price of a cryptocurrency would need to change to reach a certain level. The results were as follows: BTC had a buy price of 76954.146 and a sell price of 76974.439, with an age of 42 seconds and a hurdle percentage of 0.528. ETH had a buy price of 2449.2 and a sell price of 2450.079, with an age of 109 seconds and a hurdle percentage of 0.538. SOL had a buy price of 99.43 and a sell price of 99.5329, with an age of 132 seconds and a hurdle percentage of 0.606. USDC had a buy price of 0.9995 and a sell price of 1, with an age of 659 seconds and a hurdle percentage of 0.552. USDG had a buy price of 0.99999 and a sell price of 1.00001, with an age of 87844 seconds and a hurdle percentage of 0.504. XRP had a buy price of 1.34253 and a sell price of 1.347, with an age of 132 seconds and a hurdle percentage of 0.837. UNI had a buy price of 6.06207 and a sell price of 6.07327, with an age of 43 seconds and a hurdle percentage of 0.688. LTC had a buy price of 52.8987 and a sell price of 53.2034, with an age of 75 seconds and a hurdle percentage of 1.081. HYPE had a buy price of 78.779 and a sell price of 79.15, with an age of 64 seconds and a hurdle percentage of 0.975. LINK had a buy price of 11.4845 and a sell price of 11.513, with an age of 139 seconds and a hurdle percentage of 0.751. The fund received 8 out of 10 fresh quotes. The recipient of a Bitcoin transaction had 598.50048115 BTC, with a mempool net of 0.00014572 and a transaction number of 296. The reserve had 3601.47203772 BTC, with a mempool net of 3.384e-05 and a transaction number of 590. The recent page was 56, with a controlled value of 0 and a not full history value of 1. The tip was 4052188, with an age in seconds of 56 and a timestamp of 1789100110. There was a connection timeout error. The operator notes were 2, with an operator hash of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The operator noted that following a recent incident affecting the Liquid Network and the movement of funds, Blockstream made a statement. The fund attempted 7 out of 7 panels and had 0 eligible results, with balances but no sales or recovery.

A news scan found the following data line: fund=crypto research-only eligible=0 observed=2026-09-11T04:15:58.800041+00:00 src=https://runway.adityaperswal.workers.dev/data/crypto-bars. The data shows observations for three cryptocurrencies: BTC, ETH, and SOL. For each, the data includes the symbol, hour, close price, prior 480 hour high, whether the 480 hour period is complete, a signal, eligibility, and a note. The notes for all three mention a frozen rolling threshold and the need for a review of quote, cost, exit, and sizing. The data is as follows: BTC closed at 76864.9 at 2026-09-11T03:00:00+00:00, with a prior 480 hour high of 82233.502538071, complete480 is true, signal is false, and eligible is false. ETH closed at 2445.725 at the same time, with a prior 480 hour high of 2565.9, complete480 is true, signal is false, and eligible is false. SOL closed at 99.29365, also at the same time, with a prior 480 hour high of 110.5357, complete480 is true, signal is false, and eligible is false. The data ends with the note that these are observations only, with no orders.

A news scan found the headline about a crypto funding and supply update as of 06UTC, which replaces the previous update 202 with an exit code of 0. The fund in question is for crypto research only and was observed on 2026-09-11T04:15:20.534830+00:00. The source of this information is https://www.okx.com/api/v5/public/time with an offset of 0.0566 seconds and an uncertainty of 0.1416 seconds. Additional information was found at https://www.okx.com/api/v5/public/funding-rate and https://www.okx.com/api/v5/public/open-interest. The data shows the following information for Bitcoin, or BTC, which is a type of cryptocurrency: the timestamps are 1789100108592 and 1789100121383, the age is 12.9 and 0.1, the data is fresh, the funding rate is 0.9721 basis points, the settled rate is 0.876 basis points, and the open interest in coins is 29805.812. For Ethereum, or ETH, which is another type of cryptocurrency, the timestamps are 1789100108634 and 1789100121387, the age is 12.9 and 0.1, the data is fresh, the funding rate is 0.4557 basis points, the settled rate is 0.3053 basis points, and the open interest in coins is 589029.432. For Solana, or SOL, which is a type of blockchain, the timestamps are 1789100109694 and 1789100121384, the age is 11.8 and 0.1, the data is fresh, the funding rate is 0.4298 basis points, the settled rate is -0.0697 basis points, and the open interest in coins is 2728351.59. The source https://www.circle.com/transparency provides information on the issuer date of September 07, 2026, and the following data: bn, which refers to the balance, is [74.28, 8.8, 7.86, 36.59, 34.37, 333.52, 331.64] for circ, issue7, redeem7, issue30, redeem30, issue365, and redeem365. The source https://www.usdc.com/ provides a raw_cms_value of UNPARSED and an issuer_date of UNPARSED with a locale/template ambiguity and no flow inference. The source https://stablecoins.llama.fi/stablecoin/2 provides the following data: the timestamp is 1789084800, the circulating value is 74212069483.12572, and the Ethereum, Solana, and Hyperliquid L1 values are [1789084800, 47184685905.99], [1789084800, 6987674198.11], and [1789084800, 6839110360.24] respectively. The crypto collection has ended, and the definitions and cutoffs differ, with the flow not buying and the eligible value being 0.

A preliminary result removed daily data from March 12, 2026, because the hourly history only had 23 hours. This affected the selection of trades and the 20-day entry windows. To understand the impact of timeframe, controls were added with the same settings, including 2ATR, 3R, costs, timeout, and trail toggles. A missing hour in the data is considered a data availability issue, not a filter, unless it is validated. The controls retained full daily data and skipped missing hourly bars without creating fake quotes. The goal is to understand how different factors, such as bar construction and entry time, affect the results. The original study, number 2469, and its initial results were preserved unchanged.

An audit of the C91104 implementation found that two validation runs failed to produce accepted results. The vectorbt init_cash setting was changed from an integer to an explicit 1.0 cash value to ensure an unlevered portfolio. The raw roundtrip wealth return calculation includes the credited buy fee, but the vectorbt closed trade returns denominator does not include the entry fee, so the results are compared against closed returns multiplied by 0.9975. The portfolio total return already uses the whole cash and remains unchanged. The system checks for trade count equality and per trade differences less than or equal to 1e-10 before saving any experiment. This change puts into practice an existing lesson about equity-normalized fees rather than relying on attractive headline output. No orders were placed.

A controlled study uses matched full 24-hour UTC days for native and reconstructed daily inputs, with missing days remaining as Not a Number (NaN), never filled in for signal or risk. The study uses 20 high and 10 low, and a simple 14-day True Range (TR) with a 2 Average True Range (ATR) stop and a net 3R target at 25 basis points (bp) credited asset fees and 10 basis points (bps) slippage. There is a 20-day timeout and a 20 bp stress that leaves the target fixed. Daily signals become known at day + 1, and hourly signals at hour + 1, with the exact next existing bar timestamp required, and gaps drop pending signals rather than using fictitious fills. The calendar-hourly method uses the previous 20 completed UTC days, while the rolling-hourly method uses the prior 480 hours high and 240 hours low with full minimum bars, which is a different rule from the calendar method. Execution checks only open prices, and target or stop crossing gets the actual sampled open, not a guaranteed level. A trail variant has arms at + 2R and moves to a break-even point (BE) and peak - 2 entry ATR, with the target retained. The study includes both static and trail controls, with no parameter selection. The study uses data from 2024-25 for training and from 2026 January 1 to September 11 00 UTC for the holdout period, with flat resets and last available open liquidation. This is an attribution experiment, not an exact replication of a previous 15-minute study. Native matched bars mask incomplete reconstructed days to avoid comparing coverage regimes. The API hourly pages yield around 180-200 bars, despite a limit of 10,000, and the original 30-page guard was aborted rather than accepting truncated data, with a revised 100-page guard and 3 workers with backoff. Collection attempts and completed files are retained, but no conclusions are drawn from incomplete panels.

A news scan found the text: Prior to this, the positive intraday result of Ethereum, or ETH, may be due to sensitivity in sampling or execution, not a robust edge, with no presumption either way. The process started with 235.86 in cash and no holdings, with the goal of distinguishing timing from data-definition artifact, and a binding constraint of matched histories with causal signals. The resolution was completed in UTC, using daily and hourly bars from 2024-25 for development and 2026 as a holdout, which was explicitly not untouched. The tree had three levels and 16 leaves, including: 1 Measurement, which included construction, coverage, and consistency, such as whether to use native UTC boundary or OHLC derived trades or quotes, and whether daily OHLC matches hourly aggregation. 2 Mechanism, which included entry, exits, and risk, such as whether to use daily close or intraday completed-hour trigger, and whether to use ATR native or reconstructed. 3 Generalization, which included regimes and robustness, such as whether to use 2024-25 or 2026, and BTC or SOL as controls. 4 Decision, which included implementation and payoff, such as whether to use 15-minute venue or modeled checks, and whether to use positive net expectancy or uncertainty and cash-only sizing. The freeze was set at 20-day high and 10-day low, with ATR14 risk at 2ATR and a 20-day timeout. The goal was to replicate static net 3R first to isolate entry and data, with no tuning. The original dynamic trail study remains distinct, and the results here cannot automatically authorize it. The sources planned to use Alpaca crypto docs and /data/crypto-bars. The strongest counterargument is that cash loses rare tails, and the response must turn on actual signal or net payoff, not financial shortfall. The budget was 43 minutes from 04:00:23, with a data cutoff at 04:15, analysis at 04:30, decision at 04:38, and journal or report by 04:43.