Notes

A data scan found the following information about BTC/USD trading. The data includes rows labeled with the following information: L, trainT/known2026H, slip, return percentage, drawdown percentage, N, win percentage, openN, average win percentage, average loss percentage, and half Kelly. The training period is from 2024 to 2025, and the data for SOL starts from August 26. The prior day's high is used to determine the long position, with the condition to long if the close is above the prior day's high, and to exit otherwise at the next open. The fees for buying and selling are 0.0025 and 0.9975, respectively. The data is based on full calendar windows and 24-hour coherent days, with no unfilled data. The cash is reset on January 1, and there is no 3R or partial policy. All cells are included, with no selection. The data is as follows: [[15, T, 0.001, -60.003912, 60.003912, 136, 15.441176, 0, 0.761144, 0.931141, -47.255664], [15, T, 0.003, -76.785449, 76.785449, 136, 9.558824, 0, 0.695077, 1.253636, -61.245986], [15, H, 0.001, -12.189031, 12.189031, 28, 14.285714, 0, 1.235892, 0.743085, -25.064649], [15, H, 0.003, -21.493145, 21.493145, 28, 7.142857, 0, 1.862002, 1.068027, -21.59081], [20, T, 0.001, -56.967337, 56.967337, 120, 14.166667, 0, 0.77841, 0.942272, -47.616453], [20, T, 0.003, -73.372158, 73.372158, 120, 8.333333, 0, 0.768761, 1.26544, -56.327121], [20, H, 0.001, -12.189031, 12.189031, 28, 14.285714, 0, 1.235892, 0.743085, -25.064649], [20, H, 0.003, -21.493145, 21.493145, 28, 7.142857, 0, 1.862002, 1.068027, -21.59081], [30, T, 0.001, -49.779496, 49.779496, 94, 14.893617, 0, 0.856123, 1.005043, -42.29509], [30, T, 0.003, -65.518541, 65.518541, 94, 9.574468, 0, 0.811758, 1.33023, -52.098513], [30, H, 0.001, -8.841297, 8.841297, 18, 11.111111, 0, 0.863212, 0.684205, -43.367597], [30, H, 0.003, -15.174037, 15.174037, 18, 5.555556, 0, 0.986495, 1.022012, -45.150748]]. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees

A data scan found the following information about BTC/USD trading. The data includes rows labeled with the following categories: L, trainT/known2026H, slip, return percentage, drawdown percentage, N, win percentage, openN, average win percentage, average loss percentage, and half Kelly. The training period is from 2024 to 2025, and the data for SOL starts from August 26. The prior day's close is used to determine the maximum cost, with a long position taken if the close is above the prior day's close, and an exit otherwise, with the next open being used as the exit point. The fees for buying and selling are 0.0025 and 0.9975, respectively. The data includes full calendar windows, 24-hour coherent days, and no unfilled data points. The cash is reset on January 1, and there is no 3R or partial policy. All cells are included, with no selection. The data points are as follows: [[15, T, 0.001, -82.032146, 82.033768, 235, 14.468085, 0, 0.859503, 0.993115, -42.472421], [15, T, 0.003, -92.981285, 92.981285, 235, 7.234043, 0, 1.12555, 1.297683, -38.42187], [15, H, 0.001, -24.742329, 25.484212, 58, 13.793103, 0, 1.527586, 0.807802, -19.679271], [15, H, 0.003, -40.324782, 40.324782, 58, 10.344828, 0, 1.588426, 1.169217, -23.797569], [20, T, 0.001, -78.978048, 78.978048, 208, 14.423077, 0, 0.880681, 1.01782, -41.500358], [20, T, 0.003, -90.851741, 90.851741, 208, 6.730769, 0, 1.275319, 1.315621, -34.008996], [20, H, 0.001, -25.349461, 26.444395, 57, 12.280702, 0, 1.465549, 0.785363, -22.108611], [20, H, 0.003, -40.568953, 40.73947, 57, 8.77193, 0, 1.595349, 1.147727, -24.770455], [30, T, 0.001, -71.170205, 71.170205, 162, 15.432099, 0, 0.995476, 1.081263, -35.339974], [30, T, 0.003, -84.919438, 84.919438, 162, 8.024691, 0, 1.320053, 1.374146, -31.91784], [30, H, 0.001, -12.677638, 12.677638, 31, 12.903226, 0, 1.352345, 0.698734, -22.968843], [30, H, 0.003, -22.861228, 22.861228, 31, 9.677419, 0, 1.324951, 1.06371, -29.536358]]. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees

A data scan found the following daily trading costs for BTC/USD. The data includes rows labeled as L, trainT, known 2026 H, slip, return percentage, drawdown percentage, N, win percentage, open N, average win percentage, average loss percentage, and half Kelly. The training period is from 2024 to 2025, and the data for SOL starts from August 26. The prior day's high maximum is used to determine if a long position should be taken, with the condition that if the close is above the prior day's high, a long position is taken, otherwise, the position is exited at the next open. The fees for buying and selling are 0.0025 and 0.9975, respectively. The data uses full calendar windows and 24-hour coherent days, with no unfilled data points. The cash is reset on January 1, and there is no 3R or partial policy. All cells are included, with no selection. The data points are as follows: [[15, T, 0.001, -6.458, 26.711681, 40, 35, 0, 3.501374, 2.045374, -0.726179], [15, T, 0.003, -20.288821, 36.359468, 40, 30, 0, 3.641182, 2.280222, -3.033954], [15, H, 0.001, 5.123294, 7.55752, 8, 25, 0, 6.064044, 1.090934, 5.27408], [15, H, 0.003, 1.812588, 9.569417, 8, 25, 0, 5.639633, 1.486782, 1.758049], [20, T, 0.001, -6.489969, 26.674139, 36, 33.333333, 0, 3.349448, 1.866834, -1.024115], [20, T, 0.003, -19.030861, 35.355126, 36, 25, 0, 4.009024, 2.040028, -3.226531], [20, H, 0.001, 5.123294, 7.55752, 8, 25, 0, 6.064044, 1.090934, 5.27408], [20, H, 0.003, 1.812588, 9.569417, 8, 25, 0, 5.639633, 1.486782, 1.758049], [30, T, 0.001, -7.606561, 22.490641, 27, 33.333333, 0, 3.517953, 2.098575, -1.53331], [30, T, 0.003, -17.065138, 29.444878, 27, 25.925926, 0, 4.055555, 2.264137, -3.407077], [30, H, 0.001, 5.688955, 5.468737, 5, 20, 0, 10.411554, 1.079523, 5.421466], [30, H, 0.003, 3.596165, 6.783332, 5, 20, 0, 9.969788, 1.475417, 2.765625]. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees

A data scan found the following daily costed cells for BTC/USD: Rows are labeled with the following information: L, trainT or known 2026H, slip, return percentage, drawdown percentage, N, win percentage, open N, average win percentage, average loss percentage, and half Kelly. The training period is from 2024 to 2025, and SOL started on August 26. The prior day's close is used to determine the maximum, with a long position taken if the close is above the prior day's close, and an exit otherwise, with the next open being used. Fees for buying are 0.0025 and for selling are 0.0025, with 0.9975 being the selling price multiplier. The data uses full calendar windows and 24-hour coherent days, with no unfilled data points. The cash is reset on January 1, and there is no 3R or partial policy. All cells are included, with no selection. The data points are as follows: [[15, T, 0.001, -22.384756, 39.586954, 56, 25, 0, 4.921861, 2.142737, -1.785408], [15, T, 0.003, -37.961021, 50.637047, 56, 23.214286, 0, 4.855396, 2.477587, -3.222398], [15, H, 0.001, -11.013344, 19.835354, 20, 20, 0, 4.115376, 1.702129, -3.84465], [15, H, 0.003, -17.854992, 24.049762, 20, 20, 0, 3.698744, 2.095537, -6.042435], [20, T, 0.001, -2.113871, 26.18931, 45, 28.888889, 0, 5.217158, 2.064022, 0.183084], [20, T, 0.003, -18.238696, 36.977054, 45, 26.666667, 0, 5.203473, 2.38435, -1.454556], [20, H, 0.001, -10.194852, 17.371016, 18, 22.222222, 0, 3.565994, 1.730163, -4.483482], [20, H, 0.003, -16.433561, 21.086185, 18, 22.222222, 0, 3.151555, 2.12346, -7.107035], [30, T, 0.001, 3.913754, 18.339965, 37, 29.72973, 0, 5.351273, 1.992334, 0.895279], [30, T, 0.003, -10.381603, 26.536882, 37, 27.027027, 0, 5.431879, 2.299673, -0.840827], [30, H, 0.001, -3.4658, 8.372547, 13, 23.076923, 0, 3.717138, 1.405304, -2.136431], [30, H, 0.003, -8.357318, 10.530829, 13, 15.384615, 0, 5.107739, 1.664378, -3.661326]. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees

A news scan found the headline 'Localnew247quotes00:10:44UTC six/10fresh'. This is a data line showing prices for several cryptocurrencies: Bitcoin (BTC) was priced between 77262.2 and 77283.7, with a 41-second age; Ethereum (ETH) was priced between 2526.03 and 2526.567, also with a 41-second age; Solana (SOL) was priced between 101.7 and 101.774, with a 448-second age, marked as stale; and USDT was priced between 0.99895 and 1.001, with a 10740-second age, also marked as stale. The data line also mentions watched BTC balances, with values 598.50192119 and 3601.47207156. Additionally, it notes a Liquid tip with an age of 95 seconds, 72 recent transactions with 0 controlled inputs, and an independent explorer timeout. Funding data is also provided, with prices for BTC, ETH, and SOL at 00:10:03 UTC: 27479.803 for BTC, 639913.07 for ETH, and 2740849.85 for SOL, along with coin open interest and forecast values. The forecast values are 0.4962 for one, 0.9968 for another, and 0.2396 basis points, with settled values of 0.497, 1, and 0.2761 basis points. The issuer supply remains unchanged. A new volatile watch was set on September 12 at 23 UTC, with all values set to false, except for an old pullback for ETH. The local system ended all three tasks and accepted jobs 247, 248, and 249 without remote proof. https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://www.okx.com/api/v5/public/open-interest

A news scan found the headline 'RunSep13~00:04UTC'. The results of a diagnostic test on September 13 at 00:04 UTC are as follows: for Bitcoin, also known as BTC, the test analyzed 982 bars of data and found a return of 46.4607511532 percent, with a buy and hold return of 75.6331040807 percent, a drawdown of 14.6924283918 percent, a win rate of 34.328358209 percent, and an average trade return of 0.6443234598 percent, with 67 trades. For Ethereum, also known as ETH, the test analyzed 982 bars of data and found a return of 41.5219262747 percent, with a buy and hold return of 12.6846910117 percent, a drawdown of 24.9432564043 percent, a win rate of 31.4814814815 percent, and an average trade return of 0.7762240372 percent, with 54 trades. For Solana, also known as SOL, the test analyzed 749 bars of data and found a return of 72.151727352 percent, with a buy and hold return of -35.3023847377 percent, a drawdown of 18.7953943766 percent, a win rate of 52.7777777778 percent, and an average trade return of 1.6935012342 percent, with 36 trades. The test used the same closing prices, did not hold out any data, and used partial latest daily data, but the results are not executable as a live policy. https://runway.adityaperswal.workers.dev/data/crypto-bars ; node /app/agent/dist/backtest.js --symbols BTC/USD,ETH/USD,SOL/USD --strategy breakout --lookback 20 --days 981

A news scan found the following data: there were 25 quotes from 2026-09-12T13:02:10.758Z to 2026-09-12T23:46:57.326Z, with a minimum price of 15.001616666666667, a maximum price of 30.570033333333335, and a mean of 26.865672222222226, where 19 were over 20. There were also 25 signals from 2026-09-12T13:01:50.642Z to 2026-09-12T23:46:27.132Z, with a minimum of 15.051633333333333, a maximum of 30.26225, and a mean of 26.85867361111111, where 19 were over 20. Additionally, there were 9 funding events from 2026-09-12T14:01:22.725Z to 2026-09-12T23:01:33.870Z, with a minimum of 59.7445, a maximum of 75.2958, and a mean of 67.52321875, where 8 were over 20. A Last60note sample was queried on September 13 at 00:09 UTC. All latest receipts were present, but a new run at 00:01 was not yet returned, and the expected cadence differed from the realized delivery. The maximum values were 30.57 for quotes, 30.26 for signals, and 75.30 for funding. There was no claim of exact 15-minute order coverage. New local tests are distinct from remote proof. query_research fund=crypto notes2798–2694; list_jobs fund=crypto jobs244–246

On September 13, an article from the issuer dated August 5 was observed, which schedules the public mainnet launch for September 16. This launch will have more than 100 builders on a private network, where USDC will pay for gas, and validators will be permissioned. The schedule does not mean the actual activation or evidence of incremental native USDC paid fees. The existing 104 out of 307 monitors were retained, with no inference of price capture. https://www.arc.io/blog/arc-mainnet-goes-live-on-september-16-2026

A fund that only researches cryptocurrency made observations on September 13, 2026, at 00:03:52.958751 UTC. The fund used data from okx.com, which is a website that provides information about cryptocurrency, to look at the current derivatives and liquidity of the issuer. The data included information about Bitcoin, Ethereum, and Solana, which are types of cryptocurrency. For Bitcoin, the funding rate was 0.4941 and the settled rate was 0.497, with an open interest of 27484.289 coins. For Ethereum, the funding rate was 1.0 and the settled rate was 1.0, with an open interest of 639751.978 coins. For Solana, the funding rate was 0.2573 and the settled rate was 0.2761, with an open interest of 2741317.07 coins. The fund also looked at data from circle.com, which is a website that provides information about the USDC cryptocurrency, and found that the issuer's liquidity was unchanged as of September 10, 2026, with a 7-day net change of -0.25 billion and a 30-day net change of 2.02 billion. Additionally, the fund looked at data from stablecoins.llama.fi, which is a website that aggregates information about stablecoins, and found that the circulating supply of USDC was 74462450366.86803 as of September 12, 2026. The fund did not make any buying decisions based on this data, and the eligibility for buying was 0. The open interest for each coin was compared to the previous day, and the percentage changes were computed separately. The Circle issuer's data was unchanged, and the Llama data had different definitions and cutoffs, and did not provide any inference about net buying. https://www.circle.com/transparency ; https://www.okx.com/api/v5/public/funding-rate ; https://www.okx.com/api/v5/public/open-interest ; https://stablecoins.llama.fi/stablecoin/2

A fund for crypto research made observations on September 13, 2026, at 00:04:34.079935 UTC. The data came from two sources: https://runway.adityaperswal.workers.dev/data/crypto-snapshots and https://blockstream.info/liquid/api/blocks. The fund observed the following prices for various cryptocurrencies: BTC had a bid of 77267.9 and an ask of 77281.3, with an age of 58 and a depth of 77.2 USD, and a hurdle percentage of 0.519. ETH had a bid of 2525.3 and an ask of 2525.7, with an age of 164 and a depth of 83.5 USD, and a hurdle percentage of 0.518. SOL had a bid of 101.7 and an ask of 101.774, with an age of 78 and a depth of 96.1 USD, and a hurdle percentage of 0.575. USDC had a bid of 0.9995 and an ask of 1.00001, with an age of 8560 and a depth of 9995000.0 USD, and a hurdle percentage of 0.553. USDG had a bid of 0.99999 and an ask of 1.00001, with an age of 62294 and a depth of 9999900.0 USD, and a hurdle percentage of 0.504. XRP had a bid of 1.363 and an ask of 1.3669832, with an age of 100 and a depth of 5670.4 USD, and a hurdle percentage of 0.796. UNI had a bid of 6.37723 and an ask of 6.39198, with an age of 83 and a depth of 4473.3 USD, and a hurdle percentage of 0.734. LTC had a bid of 53.519 and an ask of 53.83, with an age of 115 and a depth of 3596.3 USD, and a hurdle percentage of 1.086. HYPE had a bid of 79.4485 and an ask of 79.81, with an age of 82 and a depth of 61686.8 USD, and a hurdle percentage of 0.959. LINK had a bid of 11.500747 and an ask of 11.513, with an age of 190 and a depth of 100.0 USD, and a hurdle percentage of 0.609. There were 8 fresh quotes out of 10. The recipient BTC was 598.50165719, with a mempool net of 0.00032899 and a transaction number of 639. The reserve BTC was 3601.47207156, with a mempool net of 1.092e-05 and a transaction number of 597. The recent page was 75, controlled 0, and not full history was 1. The tip was 4054816, with an age of 84 seconds and a time of 1789257790. There was an error, a ConnectTimeout, for the second item. The operator notes were 3, and the operator hash was 1e913157cc811e629b1b86d4db894412f924f07ea0928c7aa1a3e183681de0ca, with a claim regarding the theft of bitcoin from the Liquid Network. A news scan found the headline 'To those responsible for the theft of bitcoin from the Liquid Network: Blockstream will not pay a r'. The fund had 7 out of 7 panels attempted, but 0 were eligible. The direct collection was on September 13 at 00:04 UTC, with labels inherited from prior incident research, and ownership not independently established. The operator's latest text hash was unchanged, and the posting date was unresolved. The recent 75 transaction page was not full history, and the increasing balance may be dust, with no sale or recovery inference. An independent explorer timed out. https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte ; https://blockstream.info/api/address/bc1qdlld6antmv4xug242ed83q7k4rqw50cwfns38szx4qu2f4jwaxxsuhwxxr ; https://blockstream.info/liquid/api/blocks ; https://x.com/Blockstream

A news scan found the data line for SOL/USD C91300 100 settled OKX funding distribution. The data shows the following information: the symbol is SOL, the number is 100, the first timestamp is 1786406400000, and the last timestamp is 1789257600000. The latest basis point, or bp, is 0.276108461242, which is a measure of the difference between two prices. The strict percentile is 55.55555555555556, and the weak percentile is also 55.55555555555556. The median bp is 0.121511314192, and the 95th percentile bp is 1. The sum of bp is 22.055246562602. This data was collected on September 13 at 00:05 UTC, and it covers the period from August 11 at 00:00 UTC to September 13 at 00:00 UTC. The latest data point is ranked against 99 prior data points. The strict count only includes prior data points that are less than the latest, while the weak count includes prior data points that are less than or equal to the latest. The data is based on single venue 8-hour settlements, and repeated 1 bp observations can cause ties. No extreme definition or trading signal has been adopted. The period sum is a constant notional funding, which means it is not a leveraged return. https://www.okx.com/api/v5/public/funding-rate-history

A news scan found the data line: ETH/USD C91300 100 settled OKX funding distribution. The data shows the following information about Ethereum funding: the symbol is ETH, the amount is 100, the first timestamp is 1786406400000, and the last timestamp is 1789257600000. The latest basis point is 1, the strict percentile is 76.76767676767676, the weak percentile is 100, the median basis point is 0.666945112963, and the 95th percentile basis point is 1. The sum of basis points is 62.343908509609. This data was collected on September 13 at 00:05 UTC, and it covers the period from August 11 at 00:00 UTC to September 13 at 00:00 UTC. The latest data point is ranked against 99 prior data points. A basis point is a unit of measurement, equal to 0.01%. The strict count includes prior data points that are less than the latest data point, while the weak count includes prior data points that are less than or equal to the latest data point. The data is based on single venue 8-hour settlements, and repeated observations of 1 basis point can cause ties. No extreme definition or trading signal has been adopted. The period sum represents constant notional funding, not leveraged return. https://www.okx.com/api/v5/public/funding-rate-history

A news scan found the data line for BTC/USD C91300 100 settled OKX funding distribution. The data shows the following information: the symbol is BTC, the number is 100, the first timestamp is 1786406400000, and the last timestamp is 1789257600000. The latest basis point, or bp, is 0.4970346915, the strict percentile is 35.35353535353536, and the weak percentile is also 35.35353535353536. The median bp is 0.6465002350419999, the 95th percentile bp is 1, and the sum of bp is 61.96421760265598. This data was collected on September 13 at 00:05 UTC, covering the period from August 11 at 00:00 UTC to September 13 at 00:00 UTC, with the latest data ranked against 99 prior observations. The strict count includes prior observations that are less than the latest, while the weak count includes prior observations that are less than or equal to the latest. The data is based on single venue 8-hour settlements, and repeated 1 bp observations can cause ties. No extreme definition or trading signal was adopted, and the period sum represents constant notional funding, not leveraged return. https://www.okx.com/api/v5/public/funding-rate-history

The fund manager's book for C91300 shows it owns $119.14 in cash and has no open or pending cryptocurrency trades. One closed trade was for Bitcoin, which was entered at $81614.8 and exited at $81548.7 with a quantity of 0.01681857 on September 3 from 19:42:24.677 to 20:30:20.345 UTC, and this was due to an operator-initiated liquidation. The gross loss was -0.08099%, which was not due to a strategy stop, target, or calibrated Kelly, a method to determine the optimal bet size. The account balance is $850.99, with $29.785 allocated to the quarter fund and $212.7475 to the position account cap, which is 3% of the normal equity of $25.5297. No holdings need to be reset, and another fund, EEM, is in a queue but outside of authority. query_trades fund=crypto trade10; user account context Sep13 00:01:47UTC

A news scan found the following data: Collected on September 13 at 00:03:52 UTC, the prior 480-hour high for several cryptocurrencies was completed on September 12 at 23 UTC. The high prices were: Bitcoin at 77,266.385 to 82,233.502538071, Ethereum at 2,525.045 to 2,664.8535, Solana at 101.7655 to 110.5357, Litecoin at 53.6502 to 59.34355, XRP at 1.366315 to 1.5488, and Hype at 79.685 to 89.9954713195, with Link at 11.51335 to 13.677 and legacy Uniswap at 6.3709 to 7.4795. All windows are complete and all are false. The current top 10 includes 7 volatile cryptocurrencies and 3 stablecoins. An Ethereum 72-hour detector showed a breakout available on September 11 at 14 UTC, with a pullback at 22 UTC, triggering a price of 2,518.687664708425 and closing at 2,515.682. The detector is set to ignore portfolio occupancy and earlier arms, and there is no permission for late entry. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'C91300 completed Sep12 universe and aggregation audit'. This refers to an automated trading fund manager's notes. The manager collected data on September 13 at 00:03 UTC from 36 published USD pairs, covering seven complete UTC days from September 6 to 12. The top 10 assets by value in USD were USDC with $10,350,216, USDG with $1,899,983, BTC with $676,586, ETH with $383,393, SOL with $196,860, XRP with $185,076, HYPE with $150,323, LTC with $144,752, LINK with $141,496, and USDT with $102,726. The values are the sum of the value of each asset multiplied by its volume-weighted value, rounded. USDG only reported data for 3 out of 7 days, and its inventory is not live, meaning it is not currently being traded. The assets were confirmed, and USDT replaced UNI. The daily close screen for ETH showed a price of $2,525.045, which is higher than $2,516.066, and this is true. XTZ, which is not in the top 10, also showed a true value. The daily open, high, low, and close prices and volume for BTC, ETH, and SOL on September 12 were reconstructed exactly from 24-hour bars, with a small rounding error of 2.2e-14% in SOL's volume. No trades were inferred from the volume rank or close screen. https://runway.adityaperswal.workers.dev/data/crypto-bars

A trading fund manager closed one lot of Bitcoin on September 3. The lot was bought at $81,614.8 at 19:42:24 UTC and sold at $81,548.7 at 20:30:20 UTC. This was done by the operator, not because of the strategy, which is a set of rules for buying and selling. The quantity was 0.01681857, the total value was $137.26, the stop loss was $73,440, and the target price was $97,900. The trade was expected to last 3 to 4 weeks. The return on the trade was approximately -0.08099%. This trade was done manually and does not follow the normal strategy, so it cannot be used to measure how well the strategy is working. The fund currently has no Bitcoin holdings or pending orders, and its current capital is $119.14. query_trades fund=crypto trade10

A news scan found the note: Remove the December 31, 2025 row so the training response cannot cross January 1. The comparison between N496 and N497 shows a correlation of -0.026123675. The mean of the next return is 18.4456816% for positive flow and 18.6261395% for non-positive flow. The known values for 2026 remain unchanged. This update affects 2974 training values and resolves a one-session split issue. The availability of the publication is still unknown, and no trade or backtest is planned. https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf

A news scan found the following data line for SOL/USD: the ATR20, which is a measure of price volatility over 20 days, is 5.676981067074999. The last complete day was 2026-09-11 00:00:00+00:00. The hypothetical entry price was 101.8017. The stop, which is the price at which to sell to limit losses, is 90.44773786585. The partial 2R, which is a target price for a partial sale, is 124.50962426829999. The target net 3R, which is another target price, is 142.91938019717904. This represents a target gain percentage of 40.38997403499061. The modeled stop loss fraction, which is the fraction of the investment that could be lost, is 0.11685101122665387. The quarter notional, which is a quarter of the investment amount, is 29.785. The quarter stop loss in USD, which is a quarter of the potential loss in US dollars, is 3.4804073693858855. The full stop loss in USD, which is the total potential loss in US dollars, is 13.921629477543542. The MFE20, which is the maximum favorable excursion over 20 days, has a value of 213. The median percentage of MFE20 is 8.897815440488642, and the 90th percentile is 23.71503978146556. The target coverage percentage of MFE20 is 6.103286384976526. The actual size of the investment is 0. This data was computed on September 12 from completed daily OHLC, which is a type of financial data, reconstructed only for 24-hour days. The hypothetical entry was made at 08:03 with a 10 basis point slip, and the stop was set at 2 ATR20 with a 25 basis point fee on each side, credited to the coin, and a 10 basis point sell slip. The formula used to calculate the target price is k = 0.9975^2 * 0.999 / E, risk = 1 - k * S, and T = ((1 + 3 * risk) / k - 0.25 * (E + 2 * (E - S))) / 0.75. The full net 3R includes a quarter at 2R. The MFE 213 overlapping 20-day windows in 2026 is an unconditional high excursion, not an executable target with odds. The SOL quote is stale, and all actual sizes are 0. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://docs.alpaca.markets/us/docs/crypto-fees

A news scan found the following data line for ETH/USD: the ATR20, which is a measure of price volatility, is 98.98817058315004. The last complete day of trading was 2026-09-11 00:00:00+00:00. The hypothetical entry price was 2526.6240999999995. The stop, which is the price at which a trade is automatically stopped to limit losses, is 2328.6477588336993. The partial 2R, which refers to a profit target, is 2922.5767823326. The target net 3R, another profit target, is 3267.731902343976. The target gain percentage is 29.331937518682594. The modeled stop loss fraction, which is the fraction of the investment that could be lost, is 0.08387557333362894. The quarter notional, which is a measure of the investment size, is 29.785. The quarter stop loss in USD is 2.498233951742138. The full stop loss in USD is 9.992935806968552. The MFE20_n, which is a measure of the maximum favorable excursion, is 213. The MFE20_median_pct, which is the median percentage of the maximum favorable excursion, is 7.1107744590251265. The MFE20_P90_pct, which is the 90th percentile of the maximum favorable excursion, is 23.87248064917241. The MFE20_target_coverage_pct, which is the percentage of the target that is covered by the maximum favorable excursion, is 8.450704225352112. The actual size of the investment is 0. This data was computed on September 12 from completed daily OHLC, which is a type of financial data that includes open, high, low, and close prices, reconstructed only for 24-hour days. The hypothetical trade was based on an ask price plus a 10 basis point slip, and a stop price minus 2 ATR20. There were 25 basis point fees on each side, credited to the coin, and a 10 basis point sell slip. The formula used to calculate the trade was k = 0.9975^2 * 0.999 / E, risk = 1 - k * S, and T = ((1 + 3 * risk) / k - 0.25 * (E + 2 * (E - S))) / 0.75. The full net 3R includes the quarter at 2R. The MFE 213 overlapping 20-day windows in 2026 is an unconditional high excursion, not an executable target odds. The SOL quote is stale, and all actual sizes are 0. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://docs.alpaca.markets/us/docs/crypto-fees

A news scan found the headline is not available, but a data line shows information about a hypothetical trading scenario for Bitcoin, or BTC, in US dollars, as of September 11, 2026. The average true range, or ATR, over 20 days is 2437.141232487298. The last complete day of trading was September 11, 2026. The hypothetical entry price is 77340.627364. The stop loss price is 72466.3448990254. The partial 2R price is 87089.1922939492. The target net 3R price is 96073.38072512562. This represents a target gain percentage of 24.221103447946966. The modeled stop loss fraction is 0.06863489373987353. The quarter notional is 29.785. The quarter stop loss in US dollars is 2.044290310042133. The full stop loss in US dollars is 8.177161240168532. The maximum favorable excursion, or MFE, over 20 days, denoted as MFE20, has a value of 213. The median percentage of MFE20 is 5.8684160071170055. The 90th percentile of MFE20 is 18.769612027084108. The target coverage percentage of MFE20 is 8.450704225352112. The actual size is 0. This information was computed on September 12 from completed daily open, high, low, and close, or OHLC, data reconstructed for only 24-hour days. The hypothetical scenario assumes an ask price plus a 10 basis point slip, a stop loss at the entry price minus 2 times the average true range over 20 days, and 25 basis point fees on each side, credited to the coin, with a 10 basis point sell slip. The given formula is used to calculate the target price. The maximum favorable excursion of 213 over 20 days is an unconditional high excursion and not an executable target with given odds. The SOL quote is stale, and all actual sizes are 0. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://docs.alpaca.markets/us/docs/crypto-fees

A news scan found the headline 'IBIT C91208 flow statistics'. The flow statistics for IBIT C91208 include a 5-day flow in US dollars of 526,617,506.64 and a 20-day flow in US dollars of 2,911,974,094.80. The daily mean 5-day flow in US dollars is 105,323,501.33. A study was done on this data, which included a training set of 497 data points, with a correlation of -0.0262, an intercept of 0.0019, and a beta of -0.0016. The study also found that when the flow is positive, the next mean percentage change is 0.1923, and when the flow is not positive, the next mean percentage change is 0.1863. A separate set of 173 data points, known as the holdout set, had a correlation of 0.039, an intercept of -0.0005, and a beta of 0.329. For this set, when the flow is positive, the next mean percentage change is -0.0252, and when the flow is not positive, the next mean percentage change is -0.0656. This data was observed on September 12 and includes 675 rows in the primary issuer historical tab. The data from 2024 to 2025 is already known, and the data from 2026 is also known, but the final response in 2025 crosses into January 1 by one session, so sensitivity is needed before making any conclusions. The ordinary least squares, or OLS, method was used to analyze the next NAV return on the same-day fractional share change, but the actual publication time is unknown and there is no costed trading rule. https://www.blackrock.com/varnish-api/blk-one01-product-data/product-data/api/v1/get-fund-document?appSubType=ISHARES&appType=PRODUCT_PAGE&component=fundDownload&locale=en_US&portfolioId=333011&targetSite=us-ishares&userType=individual

A news scan found the following data about IBIT C91208 historical shares. The data shows the number of shares and their value on certain dates. On 2026-09-03, the net asset value, or NAV, was 46.134102, with 1,375,120,000 shares, a change, or delta, of 2,640,000 shares, and a flow of 121,794,029.28. On 2026-09-04, the NAV was 45.145313, with 1,384,960,000 shares, a delta of 9,840,000 shares, and a flow of 444,229,879.92. On 2026-09-08, the NAV was 44.39645, with 1,387,560,000 shares, a delta of 2,600,000 shares, and a flow of 115,430,770. On 2026-09-09, the NAV was 44.384992, with 1,387,800,000 shares, a delta of 240,000 shares, and a flow of 10,652,398.08. On 2026-09-10, the NAV was 43.674034, with 1,387,360,000 shares, a delta of -440,000 shares, and a flow of -19,216,574.96. On 2026-09-11, the NAV was 43.71244, with 1,386,800,000 shares, a delta of -560,000 shares, and a flow of -24,478,966.4. This data comes from a primary issuer workbook, which has 675 rows and two missing cells. There is a discrepancy between the number of shares on September 10 in the holdings sheet and the historical data, with 1,386,800,000 shares in the holdings sheet and 1,387,360,000 shares in the historical data. The flow is calculated as the delta in shares times the same-day NAV, but this has not been verified as actual cash or the time of spot execution. https://www.blackrock.com/varnish-api/blk-one01-product-data/product-data/api/v1/get-fund-document?appSubType=ISHARES&appType=PRODUCT_PAGE&component=fundDownload&locale=en_US&portfolioId=333011&targetSite=us-ishares&userType=individual

A news scan found the following data: this is a joint dependence and bootstrap panel report. The correlation between cryptocurrencies is as follows: Bitcoin (BTC) has a correlation of 1 with itself, 0.8880003912183593 with Ethereum (ETH), and 0.8379097255194173 with Solana (SOL). Ethereum has a correlation of 0.8880003912183593 with Bitcoin, 1 with itself, and 0.8490836574363663 with Solana. Solana has a correlation of 0.8379097255194173 with Bitcoin, 0.8490836574363663 with Ethereum, and 1 with itself. A regression analysis was performed, which is a statistical method to establish a relationship between variables. For Ethereum, the analysis had 6101 data points, an hourly percentage intercept of -0.00020398877846032964, and a beta of 1.1792211388777825. For Solana, the analysis had 6101 data points, an hourly percentage intercept of -0.0009136087504618525, and a beta of 1.2013851704885283. A simulation was run with 10000 paths, a horizon of 16 hours, and blocks of 6 hours. The results showed a mean return percentage of -0.037461886667985556, a 1% return percentage of -6.000626806386544, a 5% return percentage of -3.6372404794102553, and a 95% return percentage of 3.634198087898594. The probability of a loss above 3% of equity was 0, and the worst path percentage was -11.815273870307585. The simulation was run using Python with a numpy seed of 91208, and the data was retrieved from cached API pages on September 12 at 08:06 UTC. https://runway.adityaperswal.workers.dev/data/crypto-bars

A data scan found the following information about the SOL symbol. The data was observed on 2026-09-12 at 08:07:50.086322+00:00 from the source https://runway.adityaperswal.workers.dev/data/crypto-bars. The data includes 6101 observations with 1 missing hour. The last observed price was 101.645. The hourly standard deviation percentage, which measures how much prices typically vary from their average, was 0.6716982821684605. The standard deviation percentage was 0.5896224891614008 on weekends and 0.7017390316702797 on weekdays. The lowest 1% of hourly returns were -1.9401703232696188%, and the highest 1% were 2.080474790373521%. The daily standard deviation percentage was 3.3377745344400167. Funding data from https://www.okx.com/api/v5/public/funding-rate-history showed 100 observations from 2026-08-10 08:00:00+00:00 to 2026-09-12 08:00:00+00:00. The latest realized funding rate, which is the rate at which investors are paid to hold a position, was 0.8719517777829999 basis points. The prior median funding rate was 0.121511314192 basis points. The latest funding rate was in the 75.75757575757575 percentile, meaning it was higher than 75% of the prior rates. The sum of the funding rates over the period was 23.329550194861998 basis points. All data from Jan 1 to Sep 12 at 07 UTC was reindexed, with no forward filling and one missing hour per asset. The funding data includes 100 8-hour settlements from Aug 10 to Sep 12, with the latest rate ranked against the prior 99 rates. This is a small sample from a single venue and may not represent extreme universal values. https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://www.okx.com/api/v5/public/funding-rate-history