Notes

A news scan found the headline about a SOL/USD trading signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk and a net 3R target, and a 20-day timeout. The signal also includes a +2R breakeven and a 2 ATR absolute trail if enabled. The signal was trained on data from 2024 to 2025 and tested on data from January 1, 2026, to September 11, 2026. The test started at 02:18 UTC and used a daily holdout, with the last test on September 10. The signal uses a long-only vector, with a 25 basis point credit for asset fees, and a 10 to 20 basis point slip. The signal is checked every hour, and each entry event is consumed once. The results show a return percentage of -8.216356812719752, a drawdown percentage of 31.793786379107793, and a win rate of 16.666666666666664. The signal was tested on 6 samples, with a mean percentage of -0.638520804513883, an average win percentage of 27.06562348160628, and an average loss percentage of 6.179349661737916. The half Kelly criterion is 0, and the raw half Kelly criterion is -0.011795789684058464. The confidence interval is between -8.879696535176771 and 10.904320267414423. The signal was tested on 253 bars, with a maximum gap of 1440 minutes, and 1 armed trade. The buy and hold percentage is -20.81869466583686. The trade wealth was reconciled with a tolerance of 1e-10, and the fills and fees were also reconciled. The confidence intervals were calculated using 10,000 independent and identically distributed trade bootstraps, but the results are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a SOL/USD trading signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The net 3R target means the goal is to make three times the risk amount. The signal also has a 20-day timeout and a breakeven point of 2R, which is two times the risk amount, and an absolute trail of 2 ATR if enabled. The training data is from 2024 to 2025 and the testing data is from 2026 January 1 to September 11, with the last daily data being from September 10. The signal is for long positions only, with a vector backtest of 25 basis points credit for asset fees, and 10 to 20 basis points slip. The signal checks every hour and each entry event is consumed once. The results show a return percentage of 27.02404071439608 percent, a drawdown percentage of 27.306530042094945 percent, and a win rate of 44.44444444444444 percent. The sample size is 9, with a mean percentage of 4.051009994718789 percent, an average win percentage of 21.570596301785542 percent, and an average loss percentage of 9.964659050934614 percent. The half Kelly criterion, a measure of bet size, is 0.09390120555877862. The confidence interval is between -6.77535865556972 percent and 15.725042705227244 percent. The signal is based on 493 bars of data, with a maximum gap of 1440 minutes, and 2 armed trades. The buy and hold percentage is 586.352957997662 percent. The closed trade wealth is reconciled with a tolerance of 1e-10, and the confidence intervals are based on 10000 independent and identically distributed trade bootstraps. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a SOL/USD trading signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, a net 3R target, and a 20-day timeout. The signal also includes a +2R breakeven and a 2 ATR absolute trail if enabled. The training data is from 2024 to 2025 and the testing data is from 2026 January 1 to September 11, with a daily holdout on September 10. The signal is for long-only trades, with a 25 basis point credit for asset fees, a 10 to 20 basis point slip, and a next actual open for each hourly entry event consumed once. The results show a return percentage of 29.23%, a drawdown percentage of 27.02%, and a win rate of 44.44%. The signal was closed and the trade wealth was reconciled with raw fills and fees at a 1e-10 tolerance. The confidence intervals were calculated using 10,000 independent and identically distributed trade bootstraps. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes for ETH/USD. The notes mention a 1Hour signal with a 15min check. The strategy uses a 20day high and 10day low with a 2ATR risk, which means two times the average true range, a measure of price volatility. The target is a net 3R, which refers to three times the risk amount, with a 20day timeout. The notes also mention a +2R breakeven, which means doubling the initial risk amount to break even, and a 2ATR absolute trail if enabled, which means using the average true range to set a trailing stop loss. The data is from 2024-25 and 2026 Jan 1 to Sep 11 at 02UTC, with a holdout period, which is a portion of data set aside for testing, that was reused, and the last daily data point is from Sep 10. The strategy is long only, meaning it only buys, with a vector backtest, which is a type of simulation, and 25bp credit asset fees, which means 0.25% fees, and 10/20bps slip, which means a 0.1% to 0.2% slippage, or difference between expected and actual prices. The notes also mention an execution time of every hour, with each entry event consumed once. The results show a return of 18.74%, a drawdown of 20.05%, which means a 20.05% decline from the peak, and a win rate of 66.67%. The sample size is 6, with a mean return of 3.07%, an average win of 6.14%, and an average loss of 3.07%. The half Kelly criterion, which is a formula to determine the optimal bet size, is 0.25. The confidence interval, which is a range of values within which the true value is likely to lie, is between -1.18% and 7.97%. The notes also mention 24198 bars, which are units of time, and a max gap of 90 minutes. There are 0 armed trades, which means trades that are ready to be executed, and the buy and hold return is -17.30%. The closed trade wealth is reconciled with a tolerance of 1e-10, which means a very small margin of error. The notes also mention that the confidence intervals are not independent for forward evidence, and there is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes for ETH/USD. The notes are about a trading signal that uses a 1-hour time frame and checks every 15 minutes. The signal is trained on data from 2024 to 2025 and tested on data from January 1, 2026, to September 11, 2026. The trading strategy uses a risk management technique called 2 ATR risk, which means two times the average true range, a measure of price volatility. The strategy also uses a net 3R target, which means a profit target of three times the risk, and a 20-day timeout, which means the trade will be closed after 20 days if the target is not reached. The strategy has a return percentage of 91.07708075198823 percent and a drawdown percentage of 37.20039273450574 percent, which means the biggest loss during the testing period was 37.20 percent. The win rate is 55.55555555555556 percent, which means about 55 percent of trades were profitable. The average win percentage is 15.510070344664795 percent and the average loss percentage is 9.054578453776116 percent. The half Kelly criterion, a measure of optimal bet size, is 0.14804731875534533. The notes also mention that the trading strategy uses a long-only vector, which means it only buys assets, and a 25 basis point credit for asset fees, which is a type of fee for holding assets. The strategy also uses a 10 to 20 basis point slip, which means the price can move 10 to 20 basis points before the trade is executed. The notes include a data line with various statistics about the trading strategy, including the number of bars, which is a unit of time, and the maximum gap minutes, which is the longest time gap in the data. The notes conclude that the closed trade wealth has been reconciled with a tolerance of 1e-10, which means a very small margin of error, and that the confidence intervals, which are a measure of uncertainty, are not independent for forward evidence, which means the results may not be reliable for future predictions. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes are about the ETH/USD trading pair, with a 1-hour signal and a 15-minute check. The strategy uses a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target is a net 3R, which refers to three times the risk amount. There is a 20-day timeout and a breakeven point of 2R, with a 2 ATR absolute trail if enabled, meaning the stop-loss price will move as the price moves. The data is from January 1, 2024, to September 11, 2026, with a holdout period, which is a portion of data set aside for testing, from September 10. The strategy is long-only, meaning it only buys, with a vector backtest, which is a type of simulation, and 25 basis points credit asset fees, with 10 and 20 basis points slip, which refers to the difference between expected and actual prices. The results show a return of 21.81%, a drawdown of 20.21%, and a win rate of 66.67%. The sample size is 6, with a mean return of 3.53%, an average win of 6.93%, and an average loss of 3.27%. The half Kelly criterion, which is a formula to determine the optimal bet size, is 0.25. The confidence interval, which is a range of values within which a population parameter is likely to lie, is between -1.00% and 8.73%. The results are based on 24,198 bars of data, with a maximum gap of 90 minutes, and 2 armed trades, which are trades that are ready to be executed. The buy and hold return is -17.30%. The trades were reconciled with a tolerance of 1e-10, which is a very small number. The confidence intervals were calculated using 10,000 independent and identically distributed trade bootstraps, which is a method of resampling. There is no off-run trail or quote execution proof, which refers to the lack of evidence for the execution of trades. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes for ETH/USD. The notes mention a 1Hour signal with a 15-minute check. The strategy uses a 20-day high and 10-day low with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target is a 3R return, with a 20-day timeout. The breakeven point is set at 2R, which is two times the initial risk, with a 2 ATR absolute trail if enabled. The data is from January 1, 2026, to September 11, 2026, with a holdout period that ended on September 10. The strategy is long-only, meaning it only buys, with a vector backtest of 25 basis points credit for asset fees, and a 10 to 20 basis points slip. The next actual open is used for each hourly entry event, which is consumed once. The results show a return of 22.17%, a drawdown of 20.05%, and a win rate of 66.67%. The sample size is 6, with a mean return of 3.56%. The average win is 6.88% and the average loss is 3.07%. The half Kelly criterion, a measure of optimal bet size, is 0.26. The confidence interval is between -0.80% and 8.55%. The strategy has 24198 bars of data, with a maximum gap of 90 minutes, and 2 armed trades. The buy and hold return is -17.30%. The trade wealth has been reconciled with fills and fees at a tolerance of 1e-10. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes mentioned a trading strategy for ETH/USD, which is a cryptocurrency pair, with a 1Hour signal and 15min check. The strategy used a 20dayhigh and 10daylow, with a 2ATR risk, which means two times the average true range, a measure of price volatility. The target was a net 3R, which means three times the risk, and had a 20day timeout. The strategy also had a +2R breakeven, which means breaking even after two times the risk, and a 2ATR absolute trail if enabled, which means a stop-loss order that trails the price by two times the average true range. The training data was from 2024-25 and the testing data was from 2026 Jan 1 to Sep 11 02UTC. The holdout data was reused and the last daily data was from Sep 10. The strategy was long only, meaning it only bought the asset, and used a vectorbt backtester with 25bp credit asset fees, which means 25 basis points, or 0.25%, and 10/20bps slip, which means a 10 or 20 basis points difference between the expected and actual price. The strategy entered a trade every hour and each entry event was consumed once. The results showed a return of 57.98%, a drawdown of 36.56%, and a win rate of 47.37%. The sample size was 19 and the mean return was 3.28%. The average win was 16.08% and the average loss was 8.24%. The half Kelly criterion, a measure of optimal bet size, was 0.102. The confidence interval was between -2.47% and 9.50%. The strategy used 70128 bars of data and had a maximum gap of 195 minutes. There were 7 armed trades and the buy and hold return was 30.11%. The closed trade wealth was reconciled with a tolerance of 1e-10, which means a very small difference. The confidence intervals were calculated using 10000 independent and identically distributed trade bootstraps, which means resampling the data with replacement. There was no off-run trail or quote execution proof, which means no additional evidence to support the strategy's performance. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes are about the ETH/USD trading signal. The signal uses a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target is a net 3R, where R is a unit of profit, and there is a 20-day timeout. If the trail feature is enabled, the signal will adjust the stop loss to 2 ATR absolute or 2R breakeven. The training data is from 2024 to 2025 and the testing data is from 2026 January 1 to September 11, 02 UTC. The signal is long only, meaning it only buys, and uses a vector backtest with 25 basis points credit for asset fees, and 10 or 20 basis points slip, which is the difference between the expected and actual price. The signal checks every 15 minutes and executes every hour. The results show a return percentage of 76.33%, a drawdown percentage of 35.66%, and a win rate of 52.63%. The average win percentage is 15.38% and the average loss percentage is 8.92%. The half Kelly criterion, a measure of optimal bet size, is 0.1257. The confidence interval is between -1.84% and 9.96%. The signal has 19 samples and 70128 bars of data. The maximum gap between trades is 195 minutes and there are 8 armed trades. The buy and hold percentage is 30.11%. The closed trade wealth is reconciled with a tolerance of 1e-10. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes are about a trading signal for ETH/USD. The signal is based on a 1-hour time frame and has a 60-minute check. The strategy uses a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target is a 3R net, which means three times the risk amount, and has a 20-day timeout. If the trail feature is enabled, the strategy will adjust the stop loss to 2R breakeven or 2 ATR absolute trail. The training data is from 2024 to 2025 and the testing data is from January 1, 2026, to September 11, 2026. The holdout period is from September 10, 2026, and the last daily data is from September 10, 2026. The strategy is long only, meaning it only buys, and uses a vector backtest with 25 basis points credit asset fees, which is a type of fee, and 10/20 basis points slip, which is a type of price movement. The next actual open is used for each hour entry event, and each event is consumed once. The results show a return percentage of 15.84%, a drawdown percentage of 19.99%, and a win rate of 66.67%. The sample size is 6, and the mean percentage is 2.67%. The average win percentage is 5.64%, and the average loss percentage is 3.27%. The half Kelly criterion, a measure of bet size, is 0.24. The confidence interval, a measure of uncertainty, is between -1.78% and 8.18%. The number of bars, or price data points, is 6073, and the maximum gap minutes is 120. There is one armed trade, and the buy and hold percentage is -17.30%. The closed trade wealth is reconciled with fills and fees at a 1e-10 tolerance, which is a very small number. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, which is a type of statistical simulation. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91102 net3R 1Hour signal 60min check holdout slip10 trailtrue'. The trading fund manager's notes are about a trading strategy for Ethereum, denoted as ETH. The strategy uses a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The net 3R target is the profit goal, and there is a 20-day timeout. The strategy also includes a breakeven point at 2R, which is two times the initial risk, and an absolute trail of 2 ATR if enabled. The data used for training is from 2024 to 2025, and the testing period is from January 1, 2026, to September 11, 2026, at 02 UTC. The holdout period, which is a period of time used to evaluate the strategy, is from September 10, 2026, and the last daily data is used. The strategy is long-only, meaning it only buys, and uses a vector backtest with 25 basis points credit for asset fees, and 10 to 20 basis points for slippage, which is the difference between the expected and actual prices. The strategy enters a trade every hour, and each entry event is consumed once. The results show a return percentage of 17.24%, a drawdown percentage of 19.83%, and a win rate of 66.67%. The sample size is 6, and the mean percentage is 2.87%. The average win percentage is 5.85%, and the average loss percentage is 3.07%. The half Kelly criterion, which is a measure of the optimal bet size, is 0.25. The confidence interval, which is a range of values within which the true value is likely to lie, is between -1.58% and 8.39%. The strategy uses 6073 bars of data, and the maximum gap between trades is 120 minutes. There is one armed trade, which means one trade that is ready to be executed. The buy and hold percentage, which is the return of simply buying and holding the asset, is -17.30%. The closed trade wealth is reconciled with raw fills and fees at a tolerance of 1e-10. The confidence intervals are based on 10,000 independent and identically distributed trade bootstraps, which is a method of estimating the distribution of a statistic. There is no off-run trail or quote execution proof, which means there is no evidence that the strategy was executed correctly outside of the testing period. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes mention a trading strategy for ETH/USD with a 1Hour signal and 60min check. The strategy uses a 20day high and 10day low with 2ATR risk, a net 3R target, and a 20day timeout. It also includes a +2R breakeven and 2ATR absolute trail if enabled. The training data is from 2024-25 and the holdout period is from 2026 Jan 1 to Sep 11 02UTC. The strategy is long only, with a vectorbt 25bp credit and asset fees, 10/20bps slip, and next actual open. Each hour entry event is consumed once. The results show a return percentage of 30.279763080009133, a drawdown percentage of 40.558447324396404, and a win rate of 47.368421052631575. The strategy was trained on 19 samples with a mean percentage of 2.213130263049632. The average win percentage is 14.094122130843392 and the average loss percentage is 8.479762417964753. The half Kelly criterion is 0.07851252609080367. The notes also mention that the closed trade wealth was reconciled with raw fills and fees at a 1e-10 tolerance. Additionally, the notes state that 10000 IID trade bootstraps are not independent for forward evidence and there is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the signal data for ETH/USD. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, net 3R target, and a 20-day timeout. The breakeven is set at +2R and the absolute trail is 2 ATR if enabled. The training data is from 2024-25 and the testing data is from January 1, 2026, to September 11, 2026, at 02 UTC. The holdout data is reused and the last daily data is from September 10. The signal is for long-only trades, with a vector backtest of 25 basis points credit for asset fees, a 10-20 basis points slip, and the next actual open for each hourly entry event consumed once. The results show a return percentage of 33.25%, a drawdown percentage of 39.72%, and a win rate of 47.37%. The sample size is 19, with a mean percentage of 2.31%, an average win percentage of 14.10%, and an average loss percentage of 8.30%. The half Kelly criterion is 0.0820. The confidence interval is between -3.23% and 8.41%. The total number of bars is 17541, with a maximum gap of 180 minutes and 7 armed trades. The buy and hold percentage is 30.11%. The closed trade wealth is reconciled with draw and fills at a 1e-10 tolerance. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes for ETH/USD. The notes are about a trading signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target is a net 3R, which means three times the risk amount. There is a 20-day timeout. If the trail is enabled, the breakeven is 2R, which means two times the risk amount, and the absolute trail is 2 ATR. The data used to train the model is from 2024 to 2025, and the holdout period, which is a period of time used to test the model, is from January 1, 2026, to September 11, 2026, at 02 UTC. The last daily data used is from September 10. The trading strategy is long only, which means it only buys assets, and it uses a vector backtest with a 25 basis point credit for asset fees, and a 10 or 20 basis point slip, which means a small allowed difference between the expected and actual prices. The next actual open price is used for each hourly entry event, and each event is consumed once. The trading signal is for a 1-day time frame and is checked every 1440 minutes. The results show a return percentage of -18.45%, a drawdown percentage of 27.86%, and a win rate of 20%. The sample size is 5, and the mean percentage is -3.71%. The average win percentage is 9.28%, and the average loss percentage is 6.96%. The half Kelly criterion, which is a measure of the optimal bet size, is 0. The raw half Kelly criterion is -0.20. The confidence interval low is -9.74%, and the confidence interval high is 3.13%. There are 253 bars of data, and the maximum gap minutes is 1440. There is 1 armed trade, which means one trade that is ready to be executed. The buy and hold percentage is -17.85%. The closed trade wealth is reconciled with a tolerance of 1e-10, which means a very small allowed difference. The confidence intervals are based on 10,000 independent and identically distributed trade bootstraps, which is a method of estimating the distribution of a statistic. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91102 net3R 1Day signal 1440min check holdout slip10 trailtrue'. This refers to a trading signal for Ethereum against the US Dollar. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, net 3R target, and a 20-day timeout. The signal also includes a 2R breakeven and 2 ATR absolute trail if enabled. The data is from January 1, 2026, to September 11, 2026, and was last updated on September 10. The signal is for long-only trades, with a 25 basis point credit for asset fees, and a 10 to 20 basis point slip. The signal is checked every hour, and each entry event is consumed once. The results show a return percentage of -17.63%, a drawdown percentage of 27.35%, and a win rate of 20%. The sample size is 5, with a mean percentage of -3.52%, an average win percentage of 9.50%, and an average loss percentage of -6.78%. The half Kelly is 0, and the raw half Kelly is -0.19. The confidence interval is between -9.56% and 3.34%. The data includes 253 bars, with a maximum gap of 1440 minutes, and 1 armed trade. The buy and hold percentage is -17.85%. The trade wealth has been reconciled with fills and fees at a 1e-10 tolerance. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes mentioned a trading strategy for ETH/USD with a 1-day signal and 1440-minute check. The strategy used a 20-day high and 10-day low with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target was a net 3R, which refers to three times the risk amount. The strategy also had a 20-day timeout and a breakeven point of 2R, which means two times the risk amount, with a 2 ATR absolute trail if enabled. The training data was from 2024 to 2025 and the testing data was from 2026 January 1 to September 11. The results showed a return percentage of -10.65%, a drawdown percentage of 40.61%, and a win rate of 37.5%. The average win percentage was 16.73% and the average loss percentage was 9.65%. The half Kelly criterion, a measure of optimal bet size, was 0.0072%. The results also included confidence intervals, which are ranges of values within which a population parameter is likely to lie, and other metrics such as the number of bars, which refers to the number of time periods, and the maximum gap minutes. The trades were closed and the wealth was reconciled with a tolerance of 1e-10, which means a very small margin of error. The notes also mentioned that the confidence intervals were based on 10,000 independent and identically distributed trade bootstraps, but were not independent for forward evidence. Additionally, there was no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about an automated trading fund manager's notes. The notes were about a trading signal for ETH/USD, which is a cryptocurrency pair, with a 1-day time frame and a 1440-minute check. The signal used a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The target was a 3R net, which means three times the risk, and there was a 20-day timeout. If the trail was enabled, the breakeven was 2R, which means two times the risk, and the absolute trail was 2 ATR. The training data was from 2024 to 2025, and the testing data was from January 1, 2026, to September 11, 2026, at 02 UTC. The holdout was reused, and the last daily data was from September 10. The trading strategy was long only, which means it only bought assets, and used a vector backtest with a 25 basis point credit for asset fees, and a 10 to 20 basis point slip. The next actual open was used for each hourly entry event, and each event was consumed once. The results showed a return percentage of -7.74933366394788, a drawdown percentage of 40.193167208940594, and a win rate of 37.5. The sample size was 16, and the mean percentage was 0.44324260216396655. The average win percentage was 16.96666763567675, and the average loss percentage was 9.470812417943703. The half Kelly criterion, a measure of optimal bet size, was 0.013062158453317485. The confidence interval was between -6.274339117302517 and 7.444067243348748. The number of bars was 731, and the maximum gap minutes was 1440. There were 5 armed trades, and the buy and hold percentage was 30.106603470859604. The closed trade wealth was reconciled with a tolerance of 1e-10, and the confidence intervals were calculated using 10000 independent and identically distributed trade bootstraps. There was no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a BTC/USD signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The net 3R target and 20-day timeout are part of the strategy. There is also a +2R breakeven and 2 ATR absolute trail if enabled. The data is from January 1, 2026, to September 11, 2026, at 02 UTC, and the holdout period is from September 10. The signal is for long-only trades, with a 25 basis point credit for asset fees and a 10 to 20 basis point slip. The next actual open is used for each hourly entry event, which is consumed once. The results show a return percentage of -2.81737%, a drawdown percentage of 22.49195%, and a win rate of 42.85714%. The sample size is 7, with a mean percentage of -0.17460%, an average win percentage of 6.42709%, and an average loss percentage of 5.12586%. The half Kelly criterion, a measure of optimal bet size, is 0. The raw half Kelly is -0.01358, and the confidence interval is between -4.97275 and 5.10366. There are 24291 bars of data, with a maximum gap of 90 minutes. The armed trades are 0. The buy and hold percentage is -12.07234. The trade wealth is reconciled with fills and fees at a 1e-10 tolerance. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a BTC/USD trading signal. The signal is based on a 1-hour time frame and checked every 15 minutes. The training period is from 2024 to 2025 and the testing period is from January 1, 2026, to September 11, 2026. The signal uses a long-only strategy, which means it only buys the asset, and it has a 25 basis point credit for asset fees. A basis point is a unit of measurement, equal to 0.01%. The signal also has a 10 to 20 basis point slip, which is the difference between the expected and actual price of a trade. The results show a return of 19.68%, a drawdown of 28.14%, and a win rate of 38.10%. The average win is 12.05% and the average loss is 5.28%. The half Kelly criterion, a formula used to determine the optimal bet size, is 0.055. The confidence interval, a range of values within which a result is likely to lie, is between -2.76% and 5.88%. The signal was tested on 70,030 bars of data, which are units of time, and the maximum gap between trades was 195 minutes. The results were reconciled to ensure accuracy, with a tolerance of 1e-10, which is a very small number. The confidence intervals were estimated using 10,000 independent and identically distributed trade bootstraps, which is a statistical method. There is no evidence of off-run trail or quote execution proof, which refers to the ability to execute trades at the quoted price. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91102 net3R 1Hour signal 15min check holdout slip20 trailtrue'. This refers to a trading signal for Bitcoin in US dollars. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, net 3R target, and a 20-day timeout. The signal also includes a 2R breakeven and 2 ATR absolute trail if enabled. The data is from January 1, 2026, to September 11, 2026, at 02 UTC, and uses a holdout set that was last updated on September 10. The trading strategy is long only, with a vector backtest of 25 basis points credit for asset fees, and a 10 to 20 basis points slip. The signal checks every hour, with each entry event consumed once. The results show a return percentage of -3.8283, a drawdown percentage of 23.1095, and a win rate of 42.8571. The mean percentage is -0.3218, the average win percentage is 6.3363, and the average loss percentage is 5.3154. The half Kelly criterion is 0, and the raw half Kelly is -0.0254. The confidence interval is between -5.1378 and 5.0287. The data includes 24291 bars, with a maximum gap of 90 minutes, and 2 armed trades. The buy and hold percentage is -12.0723. The trade wealth has been reconciled with fills and fees at a tolerance of 1e-10. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a BTC/USD C91102 net3R 1Hour signal with a 15min check and holdout slip of 10, using a trailing stop. The signal uses a 20-day high and 10-day low with a 2 ATR risk, a net 3R target, and a 20-day timeout. It also uses a 2R breakeven and 2 ATR absolute trail if enabled. The signal was trained from 2024 to 2025 and tested from January 1, 2026, to September 11, 2026, at 02 UTC, with a reused holdout from September 10. The signal is long only, with a vector backtest of 25 basis points credit for asset fees, a 10 to 20 basis points slip, and the next actual open for each hourly entry event consumed once. The results show a return percentage of -2.47%, a drawdown percentage of 22.49%, and a win rate of 42.86%. The signal was closed with a trade wealth reconciliation of draw fills plus fees at a 1e-10 tolerance. The confidence intervals were calculated using 10,000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a BTC/USD trading signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, net 3R target, and a 20-day timeout. The signal also includes a +2R breakeven and 2 ATR absolute trail if enabled. The training data is from 2024-25 and the testing data is from 2026 January 1 to September 11, with a daily holdout on September 10. The trading strategy is long only, with a vector backtest of 25 basis points credit and 10/20 basis points slip, and the next actual open is used for each hourly entry event. The results show a return percentage of 4.8533%, a drawdown percentage of 30.0635%, and a win rate of 38.0952%. The sample size is 21, with a mean percentage of 0.6314%, an average win percentage of 10.5385%, and an average loss percentage of 5.4653%. The half Kelly criterion is 0.0299%, and the raw half Kelly criterion is also 0.0299%. The confidence interval is between -3.1285% and 4.7883%, with 70030 bars and a maximum gap of 195 minutes. There were 5 armed trades, and the buy and hold percentage is 106.9549%. The closed trade wealth is reconciled with draw and fills, plus fees, at a 1e-10 tolerance. The confidence intervals are based on 10000 IID trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a BTC/USD signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, net 3R target, and a 20-day timeout. The signal also includes a +2R breakeven and 2 ATR absolute trail if enabled. The training data is from 2024-25 and the testing data is from 2026 January 1 to September 11, 02 UTC. The signal is for long-only trades, with a vector backtest of 25 basis points credit asset fees, 10/20 basis points slip, and next actual open. Each hour entry event is consumed once. The results show a return percentage of 9.177016227811933, a drawdown percentage of 28.13602603996801, and a win rate of 38.095238095238095. The signal was trained on 21 samples, with a mean percentage of 0.8151292396630752, an average win percentage of 10.713335881173801, and an average loss percentage of 5.276074847420449. The half Kelly criterion is 0.03804273704773295. The confidence interval is between -2.9184050562272015 and 4.919916410981979. The signal was tested on 70030 bars, with a maximum gap of 195 minutes and 5 armed trades. The buy and hold percentage is 106.95486135300149. The closed trade wealth was reconciled with draw and fills at a 1e-10 tolerance. The confidence intervals were calculated using 10000 independent and identically distributed trade bootstraps. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91102 net3R 1Hour signal 60min check holdout slip20 trailtrue'. This refers to a trading signal for Bitcoin in US dollars. The signal was generated using a 20-day high and 10-day low, with a 2 ATR risk, which means two times the average true range, a measure of price volatility. The net 3R target means the target profit is three times the risk. The signal also has a 20-day timeout and a breakeven point of 2R, which is two times the initial risk, with an absolute trail if enabled, meaning the stop-loss price will trail the market price. The data used to train the model was from 2024 to 2025, and the holdout period, which is a period of time used to test the model, was from January 1, 2026, to September 11, 2026, at 02 UTC. The last daily data used was from September 10. The trading strategy is long only, meaning it only buys assets, and it uses a vector backtest, which is a method of testing a trading strategy, with a 25 basis point credit for asset fees, and a 10 to 20 basis point slip, which is the difference between the expected and actual price of a trade. The signal is checked every hour, and each hour's entry event is consumed once. The results of the trading strategy are as follows: the return percentage was -4.97%, the drawdown percentage, which is the maximum loss from a peak to a trough, was 22.61%, and the win rate was 42.86%. The sample size was 7, and the mean percentage return was -0.49%. The average win percentage was 5.88%, and the average loss percentage was 5.27%. The half Kelly criterion, which is a measure of the optimal bet size, was 0. The raw half Kelly criterion was -0.04. The confidence interval, which is a range of values within which a population parameter is likely to lie, was between -5.24 and 4.82. The number of bars, which are units of time, was 6073, and the maximum gap minutes, which is the maximum time between trades, was 120. There was one armed trade, which is a trade that is ready to be executed. The buy and hold percentage return, which is the return if the asset was bought and held for the entire period, was -12.07%. The closed trade wealth was reconciled, which means it was matched with the actual trade data, with a tolerance of 1e-10, which is a very small number. The confidence intervals were generated using 10,000 independent and identically distributed trade bootstraps, which is a method of generating random samples. There is no off-run trail or quote execution proof, which means there is no evidence that the trading strategy was executed in a live market. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline about a BTC/USD trading signal. The signal is based on a 20-day high and 10-day low, with a 2 ATR risk, a net 3R target, and a 20-day timeout. The signal also includes a +2R breakeven and a 2 ATR absolute trail if enabled. The data is from January 1, 2026, to September 11, 2026, and the holdout period is from September 10. The trading strategy is long only, with a vector backtest of 25 basis points credit asset fees, a 10/20 basis points slip, and a next actual open for each hour entry event consumed once. The results show a return percentage of -3.63%, a drawdown percentage of 21.99%, and a win rate of 42.86%. The sample size is 7, with a mean percentage of -0.29%, an average win percentage of 6.09%, and an average loss percentage of 5.08%. The half Kelly criterion is 0, and the raw half Kelly criterion is -0.024. The confidence interval is between -5.05% and 5.03%. The number of bars is 6073, with a maximum gap of 120 minutes and 1 armed trade. The buy and hold percentage is -12.07%. The trade wealth has been reconciled with fills and fees at a 1e-10 tolerance. The confidence intervals are based on 10000 independent and identically distributed trade bootstraps, but are not independent for forward evidence. There is no off-run trail or quote execution proof. https://runway.adityaperswal.workers.dev/data/crypto-bars