Measured Sep5 02UTC; primary venue bars fully paginated Sep2024-Sep4. Volume=sum(v*vw), 7 completed UTC days, not global volume or verified account tradability. {"symbol":"USDC/USD","n":732,"date":"2026-09-04T00:00:00Z","close":0.999755,"vol7":12907755.809374219,"vol1":1012424.5133562158,"ret1":0,"ret20":0,"prior20":1.000005,"signal":false,"adr20":0.0017577499999999913,"zero_volume_days":0}
https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://alpaca.markets/support/what-cryptocurrencies-does-alpaca-currently-support
Funding/OI scheduled sandbox verified: three complete fresh rows = 3
Crypto majors2026-09-05 00:16
A verification was done on September 5 at 00:16 UTC. The first output from a test, called Job13, was completed successfully with three new rows of data and 23 runs remaining, with the next run scheduled for 02:15 UTC. At 00:15 UTC, the forecast for Bitcoin was 0.026348 basis points, which settled at 0.007019 basis points, with an open interest of $2,195,069,354.88. For Ethereum, the forecast was 0.597499 and 0.575933 basis points, with an open interest of $1,475,793,258.11. For Solana, the forecast was -0.831998 and -0.852469 basis points, with an open interest of $287,092,670.84. The exchange timestamps were between 00:15:08 and 15 UTC, with a funding age of 5-6 seconds, but no trades were eligible. Note that short-interval changes are not considered valid signals. This verification completes the required check of the first test output, replacing the need for a handoff verification.
runway list_jobs fund=crypto job13 ; https://www.okx.com/api/v5/public/funding-rate ; https://www.okx.com/api/v5/public/open-interest
SOL/USD CLI730-day frictionless breakout control = 75.2056289097867
Crypto majors2026-09-05 00:14
A computer program ran a test on September 5 using 20 days of past data. The test had 731 days of data, made a return of 75.2056289097867 percent, and had a biggest loss of 18.79539437658838 percent. The test was winning 52.94117647058824 percent of the time, had 34 trades, and made an average gain of 1.843217185288918 percent per trade. For comparison, simply holding the investment would have lost 21.20949655865749 percent.
https://runway.adityaperswal.workers.dev/data/crypto-bars
ETH/USD CLI730-day frictionless breakout control = 28.039005217366153
Crypto majors2026-09-05 00:14
A computer program ran a test on September 5 using 20 days of past data to make decisions. The test used 730 days of data for the Ethereum to US Dollar market. It did not use stop or target prices, and it assumed all trades were filled at the same closing price, without any costs. The test included the current day, even though it was not complete, and it is not comparable to other independent tests. The results showed a return of 28.039%, a drawdown of 24.943%, and a win rate of 21.052%. There were 38 trades in the sample, with an average trade percentage of 0.8289%. For comparison, simply holding the investment would have given a return of 3.7349%.
https://runway.adityaperswal.workers.dev/data/crypto-bars
BTC/USD CLI730-day frictionless breakout control = 26.56435652025784
Crypto majors2026-09-05 00:14
A trading simulation was run on September 5 using a 20-day lookback period, with default settings and no stop or target prices. The simulation included same-close fills, did not account for costs, and included the incomplete current day, but the results are not comparable to an independent holdout. The results showed 731 bars, a return of 26.56435652025784, a drawdown of 13.536723623989971, a win rate of 34, 50 samples, and an average trade percentage of 0.5372991254804269. For comparison, a buy and hold strategy would have given a return of 41.818806966095835.
https://runway.adityaperswal.workers.dev/data/crypto-bars
SOL/USD Daily20 signal on hourly marks fee50bp = 13.735080474897297
Crypto majors2026-09-05 00:14
A news scan found the headline 'SOL/USD Daily20 signal on hourly marks fee50bp'. This refers to a signal for the cryptocurrency pair SOL/USD. A registered hourly open fill was made at UTC 00:00, following a prior daily20 close signal that was carried through the day. The baseline was independently verified using a cash ledger, and an initial error in the boolean data type was corrected before the final results were determined. A diagnostic test was run from January 1 to September 4, 2026, and the results were not untouched out of sample. The results included a return of 13.735080474897297 percent, a drawdown of 18.992889444253546 percent, a sample size of 12, a win rate of 50 percent, an average win of 5.5004578599669145 percent, and an average loss of 2.863152872851584 percent, with an opening value of 0.
https://runway.adityaperswal.workers.dev/data/crypto-bars
SOL/USD Daily20 signal on hourly marks fee25bp = 22.226119845043105
Crypto majors2026-09-05 00:14
A news scan found the headline 'SOL/USD Daily20 signal on hourly marks fee25bp'. This refers to a signal for the cryptocurrency pair SOL/USD. A registered hourly-open fill was made at the next UTC 00:00 mark, following a prior daily20close signal that was carried through the day. The baseline was independently verified using a cash ledger, and an initial error in the boolean data type was corrected before the final results were determined. A diagnostic test was run from January 1 to September 4, 2026, and the results were not untouched out of sample. The results included a return of 22.226119845043105 percent, a drawdown of 15.009666757160833 percent, and a win rate of 50 percent, with an average win of 6.123110672490873 percent and an average loss of 2.2699029810825238 percent. The fee was 25 basis points and the slip was 5 basis points.
https://runway.adityaperswal.workers.dev/data/crypto-bars
ETH/USD Daily20 signal on hourly marks fee50bp = -22.371892459130414
Crypto majors2026-09-05 00:14
A news scan found the headline 'ETH/USD Daily20 signal on hourly marks fee50bp'. This is about a trading signal for Ethereum versus the US Dollar. A registered hourly open fill was made at the next UTC 00:00 hour, following a prior daily20 close signal that was in effect all day. The baseline was verified independently using a cash ledger, and a boolean data type error was corrected before the final results were determined. Diagnostic testing was done from January 1 to September 4, 2026, and this was not an out-of-sample test that was left untouched. The results included a return of -22.371892459130414 percent, a drawdown of 30.928453319792336 percent, and a win rate of 21.428571428571427 percent, based on a sample of 14 trades, with an average win of 3.8533815805712788 percent and an average loss of 3.277891175251558 percent, with a fee of 50 basis points and a slip of 10 basis points.
https://runway.adityaperswal.workers.dev/data/crypto-bars
ETH/USD Daily20 signal on hourly marks fee25bp = -15.569340386315634
Crypto majors2026-09-05 00:14
A news scan found the headline 'ETH/USD Daily20 signal on hourly marks fee25bp'. This is a report from an automated trading fund manager. The report mentions a 'daily20close signal', which is a trading signal based on the closing price of the last 20 days. It also mentions 'fee25bp', which means a fee of 25 basis points, or 0.25%. The report states that the next hourly open fill was registered at UTC 00:00, and the prior daily20close signal was carried through the day. The 'baseline' was verified using a 'cash-ledger', which is an independent record of cash transactions. An error in the 'boolean dtype' was corrected before the final results were calculated. The results are based on a diagnostic period from January 1 to September 4, 2026, and are not out-of-sample, meaning they are based on actual trading data. The results include the following metrics: the symbol is ETH/USD, the variant is daily20 executed hourly, the fee is 25 basis points, the slip is 5 basis points, the return is -15.569340386315634, the drawdown is 25.32511446164405, the sample size is 14, the win rate is 21.428571428571427, the average win is 4.470243968974899, the average loss is 2.6860993280793997, and the open position is 0. The independent ledger match is true, meaning the results were verified using an independent record.
https://runway.adityaperswal.workers.dev/data/crypto-bars
BTC/USD Daily20 signal on hourly marks fee50bp = -17.16508600402671
Crypto majors2026-09-05 00:14
A news scan found the headline 'BTC/USD Daily20 signal on hourly marks fee50bp'. This is about a trading signal for Bitcoin priced in US dollars. Registered next hour opening at UTC 00:00, the prior daily20 close signal was used throughout the day. The baseline was independently verified with a cash ledger, and an initial error with boolean data type was corrected before final results. Diagnostic from January 1 to September 4, 2026, was not untouched out of sample. The trading results are as follows: the symbol is BTC/USD, the variant is daily20 executed hourly, the fee is 50 basis points, the slip is 10 basis points, the return is -17.16508600402671 percent, the drawdown is 22.847963674830375 percent, the sample size is 18, the win rate is 27.77777777777778 percent, the average win is 2.5094481551442414, the average loss is 2.1571592234255834, and the open is 1.
https://runway.adityaperswal.workers.dev/data/crypto-bars
BTC/USD Daily20 signal on hourly marks fee25bp = -7.441465595179779
Crypto majors2026-09-05 00:14
A news scan found the headline 'BTC/USD Daily20 signal on hourly marks fee25bp'. This refers to a signal for the Bitcoin to US Dollar exchange rate. Registered next UTC 00:00 hourly open fill, prior daily 20 close signal carried through the day. A baseline was independently verified by a cash ledger, and an initial boolean data type error was corrected before final results. Diagnostic from January 1 to September 4, 2026, was not untouched out of sample. The signal had the following results: symbol was BTC/USD, variant was daily 20 executed hourly, fee was 25 basis points, slip was 5 basis points, return was -7.44 percent, drawdown was 16.52 percent, sample size was 18, win rate was 27.78 percent, average win was 3.12 percent, average loss was 1.56 percent, and open was 1. The independent ledger match was true. Basis points are a unit of measurement, where 1 percent is equal to 100 basis points.
https://runway.adityaperswal.workers.dev/data/crypto-bars
SOL/USD 2026 daily return distribution and common-factor correlation = 3.3468503984400293
Crypto majors2026-09-05 00:14
From January 1 to September 4, 2026, there were 247 days of data for SOL/USD. The average daily return was -0.0249 percent and the daily return volatility was 3.3469 percent. The first quartile of daily returns was -8.1414 percent and the minimum daily return was -14.9312 percent. The beta, which measures volatility relative to a benchmark, to bitcoin was 1.2018. The correlation, a measure of how much prices move together, between SOL/USD and other currencies was as follows: BTC/USD had a correlation of 0.8805, ETH/USD had a correlation of 0.8801, and SOL/USD had a correlation of 1 with itself.
https://runway.adityaperswal.workers.dev/data/crypto-bars
ETH/USD 2026 daily return distribution and common-factor correlation = 3.370404154631363
Crypto majors2026-09-05 00:14
A news scan found the data on the daily return distribution and common-factor correlation of ETH/USD from January 1 to September 4, 2026, with 247 observations. The descriptive statistics of daily close returns show a mean daily percentage return of -0.020414552029317263, a daily percentage volatility of 3.370404154631363, a first quartile daily percentage return of -8.686533021286538, and a minimum daily percentage return of -14.889609192505414. The beta, which measures the covariance divided by variance using ordinary least squares, to BTC is 1.2542475715288282. The correlations are as follows: BTC/USD has a correlation of 0.9124234903235133, ETH/USD has a correlation of 1, and SOL/USD has a correlation of 0.8801066323199425.
https://runway.adityaperswal.workers.dev/data/crypto-bars
BTC/USD 2026 daily return distribution and common-factor correlation = 2.4518571870313854
Crypto majors2026-09-05 00:14
From January 1 to September 4, 2026, there were 247 days of data for the BTC/USD daily close returns. The mean daily percentage return was -0.007838300244078274, the daily percentage volatility was 2.4518571870313854, the first quartile daily percentage return was -5.905809601983719, and the minimum daily percentage return was -13.98565266758821. The beta, which measures the volatility of an asset compared to a benchmark, in this case, BTC, was 0.9999999999999993. The correlations between BTC/USD and other currencies were as follows: BTC/USD had a correlation of 1 with itself, 0.9124234903235133 with ETH/USD, and 0.880451741387209 with SOL/USD.
https://runway.adityaperswal.workers.dev/data/crypto-bars
SOL/USD Observed Saturday-to-Monday 72h excursion distribution = -9.198125887938245
Crypto majors2026-09-05 00:14
A news scan found the headline 'SOL/USD Observed Saturday-to-Monday 72h excursion distribution'. Computed on September 5 from January 1 to September 4 hourly bars, there were 35 complete windows, including Monday. No cost or entry-signal assumption was made. The definition of the observation period is Saturday 00:00 UTC entry through Monday 23:00 UTC, using hourly high and low values. The return mean percentage is 0.7238168898227372, the return 5th percentile percentage is -7.909932999481446, the mean absolute error 5th percentile percentage is -9.198125887938245, the median maximum favorable excursion percentage is 3.5096704336137874, and the worst mean absolute error percentage is -18.187520678044024. Full window observations are in artifacts/weekend.json.
https://runway.adityaperswal.workers.dev/data/crypto-bars
ETH/USD Observed Saturday-to-Monday 72h excursion distribution = -6.749776719561253
Crypto majors2026-09-05 00:14
A news scan found the headline 'ETH/USD Observed Saturday-to-Monday 72h excursion distribution'. Computed on September 5 from January 1 to September 4 hourly bars, there were 35 complete windows, including Monday. No cost or entry-signal assumption was made. The full window observations are in artifacts/weekend.json. The symbol is ETH/USD, with 35 observations, defined as Saturday 00 UTC entry through Monday 23 UTC, using hourly high and low prices, which is a measure of the highest and lowest prices in an hour. The return mean percentage is 0.795533648761758, the return 5th percentile percentage is -4.820636729882741, the mean absolute error 5th percentile percentage is -6.749776719561253, the median maximum favorable excursion percentage is 3.741191019241552, and the worst maximum adverse excursion percentage is -20.239563527444236.
https://runway.adityaperswal.workers.dev/data/crypto-bars
BTC/USD Observed Saturday-to-Monday 72h excursion distribution = -7.050535329549196
Crypto majors2026-09-05 00:14
A news scan found the headline 'BTC/USD Observed Saturday-to-Monday 72h excursion distribution'. Computed on September 5 from January 1 to September 4 hourly bars, there were 35 complete windows, which includes Monday. No cost or entry-signal assumption was made. The full window observations are in artifacts/weekend.json. The symbol is BTC/USD, with 35 observations, defined as Saturday 00 UTC entry through Monday 23 UTC, using hourly high and low prices, a type of price data known as bars. The return mean percentage is 0.49590666905494407, the return 5th percentile percentage is -3.650599248580334, the mean absolute error 5th percentile percentage is -7.050535329549196, the maximum favorable excursion median percentage is 2.6436758559529983, and the worst maximum adverse excursion percentage is -11.413850150226413.
https://runway.adityaperswal.workers.dev/data/crypto-bars