A diagnostic report for SOL/USD, a type of cryptocurrency pair, shows the following results: the return percentage is 72.15172735196003, the buy and hold percentage is -35.26931637519872, the maximum drawdown percentage is 18.795394376588376, the win rate is 52.77777777777778, and the average trade percentage is 1.6935012342155107, based on 36 samples and 749 bars of data. This report was generated using a node and a backtest script with the following parameters: symbols BTC/USD, ETH/USD, and SOL/USD, strategy breakout, lookback 20, and 981 days. The report includes same-close fills, no fee, and an independent holdout, with the current partial day included. It uses a different close-based rule and does not infer a net 3R manager edge.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'ETH/USD C91306 required CLI gross breakout diagnostic'. The results of a backtest are as follows: for the symbol ETH/USD, there were 982 bars, a return percentage of 41.3454091225117, and a buy and hold percentage of 12.5441419019507. The maximum drawdown percentage was 24.943256404302126, the win rate was 32.72727272727273, and the average trade percentage was 0.7666518724459392, based on a sample of 55. This was run using the command node /app/agent/dist/backtest.js with the flags --symbols BTC/USD,ETH/USD,SOL/USD, --strategy breakout, --lookback 20, and --days 981. The test used same-close fills, no fee, and included the current partial day, with a different close-based rule, and did not infer a net 3R manager edge.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'BTC/USD C91306 required CLI gross breakout diagnostic'. A diagnostic report for the BTC/USD symbol shows 982 bars of data, with a return percentage of 46.46%. The buy and hold percentage is 75.71%, and the maximum drawdown percentage is 14.69%. The win rate is 34.33%, and the average trade percentage is 0.64%. The sample size is 67. A backtest was run using the command node /app/agent/dist/backtest.js with the following flags: --symbols BTC/USD,ETH/USD,SOL/USD, --strategy breakout, --lookback 20, and --days 981. The test used same-close fills, no fee, and included the current partial day. It also used a different close-based rule, and did not infer a net 3R manager edge.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Hour manager holdout slip30bp'. The trading results for SOL/USD are as follows: the return percentage is -6.497253116745284, the drawdown is 29.310515249898717, and the win rate is 33.33333333333333, based on a sample of 6 trades. The half Kelly criterion is 0, and 1 trade had a skipped stale price. The symbol is SOL, the time frame is 1 hour, the mode is manager, the slippage is 30 basis points, and the window is holdout. There were 26 events missed. The trade returns percentages are -9.5056720269299, -10.090001881370936, -6.661054067556993, -0.7914710095135558, 19.306813998002337, and 4.02024470029112. The trading strategy includes 15-minute open checks, a 2 ATR stop, and a residual target for a net 3R. A stop, or a price level that triggers a trade exit, is set at 2 ATR, which means 2 times the average true range, a measure of price volatility. The horizon exits wait for a specific cadence, or timing. The first event is retained on collisions, and the expiry is 2 hours. Fees are 25 basis points each, plus 10 or 30 basis points for slippage. The training data is from 2024 to 2025, and the known holdout data is from 2026. There is a forced terminal close. The full sample missed count includes pre-2024 warm-up trades with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill, meaning it does not execute trades at unfavorable prices. All 48 cells are included, and timestamps are managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line: return percentage is -12.913589047584736, drawdown is 32.44086751982418, win rate is 40, sample size is 10, half Kelly is 0, skipped stale price is 3, symbol is SOL, time frame is 1 Hour, mode is manager, slippage in basis points is 30, window is train, and events missed greater than 2 hours full sample is 26. Trade returns percentage is equal to a list of numbers: -2.8260315000860903, -9.124881554541153, 33.12014171328561, -17.990140285500967, -12.154517318652125, 14.934289574423554, 3.2992046776715744, 10.672500401766548, -10.989133398208661, -12.078144518620093. The strategy uses 15-minute open checks, a 2ATR stop, which means a stop-loss order set at 2 times the average true range, a plus 2R quarter or break-even at selected cadence, and a residual target for net 3R; horizon exits wait for the selected cadence. The first event is retained on collisions, expiry is 2 hours; fees are 25 basis points each plus 10 or 30 basis points slippage. The training period is 2024-2025, with 2026 as a known holdout, and there is a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history; this should not be interpreted as within-window misses. The model does not fill, meaning it does not execute trades when there is no signal; all 48 cells are used, and timestamps are managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Hour manager holdout slip10bp'. The trading fund manager's notes are as follows: the return percentage is 2.23884532897776, the drawdown is 23.01142571081386, and the win rate is 33.33333333333333, based on a sample of 6. The half Kelly criterion, a measure of optimal bet size, is 0.7170145661278271. One trade was skipped due to a stale price. The symbol being traded is SOL, the time frame is 1 hour, and the mode is manager. The slippage, or loss of value due to buying or selling, is 10 basis points. The window of time being looked at is called the holdout window. There were 26 events that were missed because they occurred more than 2 hours after the last event. The trade returns percentages are -9.142968219310955, -1.3616762128767945, -7.708974249350264, -0.3938404395199213, 18.825864531712266, and 4.43716076196381. The trading strategy includes 15-minute open checks, a stop loss of 2 average true ranges, and a profit target of 2 times the risk, with a residual target for a net 3 times the risk. The horizon exits wait for a specific cadence. In the event of a collision, the first event is retained, and the expiry is 2 hours. There are fees of 25 basis points, plus 10 or 30 basis points for slippage. The training data is from 2024-2025, and the known holdout data is from 2026. The model does not fill all possible trades. All 48 cells of data are managed, and the timestamps are in the managed_results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following trade results for SOL/USD: the return percentage is -11.319965224128794, the drawdown is 32.1700886197668, and the win rate is 40. The sample size is 10, and the half Kelly is 0. The symbol is SOL, the time frame is 1 hour, and the mode is manager. The slippage is 10 basis points, and the window is train. There were 26 events missed. The trade returns percentage is [-2.436555505669602, -8.760651528016183, 32.41191169100419, -17.66144246490834, -11.802430157342636, 15.394949434420857, 3.713230790686617, 11.11607889108772, -10.67704936962015, -12.781484035842716]. The trade strategy includes 15-minute open checks, a 2ATR stop, and a quarter profit target at a selected cadence, with a residual target for a net 3R. The horizon exits wait for the cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each plus 10/30 basis points slippage. The training period is 2024-2025, with a known 2026 holdout and a forced terminal close. The full sample missed count includes a pre-2024 warm-up with no 15-minute history. The model does not fill, and all 48 cells are managed, with timestamps in managed_results.json.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Hour immediate15 holdout slip30bp'. The trading fund manager's notes are as follows: The fund's return percentage is -4.661790690341718, the drawdown is 24.073027272110682, and the win rate is 33.33333333333333, based on a sample of 6 trades. The symbol being traded is SOL, with a 1 Hour time frame, immediate15 mode, and 30 basis points of slippage, in the holdout window. The trade returns percentages are -9.677178454459323, -1.1721289362510157, -7.291363638226267, 0.2506280146966855, 19.401777604495575, and -3.7563761109367544. The strategy includes 15-minute open checks, a 2ATR stop, and a residual target for a net 3R return. The model does not fill orders. The training data is from 2024-2025, with a known 2026 holdout period, and a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history. All 48 cells of data are managed, with timestamps recorded in the managed_results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Hour immediate15 train slip30bp'. The body of the report is a data line that includes the following information: the return percentage is 56.22943310896961, the drawdown is 36.362924954319176, and the win rate is 63.63636363636363. The sample size is 11, the half Kelly is 2.3286894776729596, and no stale prices were skipped. The symbol is SOL, the time frame is 1 hour, the mode is immediate 15, and the slippage is 30 basis points. The window is train, and 4 events were missed. Trade returns percentage is a list of numbers: -1.364949246044389, 20.906462248186884, 17.828097086949256, 4.819190118587446, 16.846387691576446, 3.2992046776715522, 13.191529873229314, -10.990765815137504, -11.935135835610488, 8.964738874730639, -9.104477141549106. The report also mentions that 15-minute open checks were used, with a 2 ATR stop, which is a type of stop-loss order that uses the average true range, and a 2R quarter or break even at a selected cadence, where R refers to the risk amount. A residual target was used for a net 3R, and horizon exits waited for the cadence. The first event was retained on collisions, with an expiry of 2 hours, and fees were 25 basis points each plus 10 or 30 basis points slippage. The train data is from 2024-25, with known 2026 holdout, and a forced terminal close. The full sample missed count includes a pre-2024 warm-up with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill orders, and all 48 cells were used, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Hour immediate15 holdout slip10bp'. The trade details are as follows: the return percentage is 4.7448785135718765, the drawdown is 23.68254187049399, and the win rate is 50. The sample size is 6, and the half Kelly criterion, a measure of optimal bet size, is 1.2224725885354886. The symbol being traded is SOL, the time frame is 1 hour, and the mode is immediate 15. The slippage, or difference between expected and actual trade price, is 10 basis points, and the window is holdout. The trade returns percentages are -9.93606093990158, -0.7760240539167351, -6.919784820407083, 0.6524353468417399, 18.630228300696473, and 5.4601057186226765. The trading strategy includes 15-minute open checks, a 2 ATR, or average true range, stop, and a quarter profit target at the break-even point, with a residual target for a net 3R, a measure of profit. Horizon exits wait for the selected cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each plus 10 or 30 basis points slippage. The training data is from 2024 to 2025, and the known holdout data is from 2026, with a forced terminal close. The full sample missed count includes pre-2024 warm-up data with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill, or execute, all orders, and all 48 cells are managed, with timestamps in the managed_results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Hour immediate15 train slip10bp'. The trading fund manager's notes are as follows: The return percentage is 60.81861551447694, the drawdown is 36.1055745942662, and the win rate is 63.63636363636363, based on a sample of 11. The half Kelly criterion, a measure of optimal bet size, is 2.4622409060517656. The symbol being traded is SOL, with a 1 Hour time frame and immediate15 mode. The slippage, or difference between expected and actual trade price, is 10 basis points. The window is set to train. There were 4 events missed due to being greater than 2 hours old in the full sample. The trade returns percentage is listed as follows: -0.9696171929261643, 21.24005870904511, 18.300355408715575, 4.575898910530585, 17.314711294519448, 3.713230790686617, 13.645204703407043, -10.907987293943755, -11.971217188346895, 9.401472607897077, -8.740165333563898. The trading strategy includes 15-minute open checks, a 2ATR stop, and a quarter profit target at a selected cadence, with a residual target for a net 3R return. The horizon exits wait for the selected cadence. In the event of a collision, the first event is retained, with an expiry of 2 hours. Fees are 25 basis points each, plus 10 or 30 basis points slippage. The training period is from 2024 to 2025, with a known holdout period in 2026, and a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill orders. All 48 cells are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Day manager holdout slip30bp'. The details are as follows: the return percentage is -24.44%, the drawdown is 37.56%, and the win rate is 14.29%. This is based on a sample of 7 events. The half Kelly criterion, a measure of optimal bet size, is 0. The symbol being traded is SOL, with a 1Day time frame and manager mode. The slippage, or difference between expected and actual trade price, is 30 basis points. The window of time being looked at is called the holdout period. There were 2 events that were missed because they occurred more than 2 hours after the previous event. The trade returns percentages are -9.88%, -14.61%, -1.05%, -7.54%, -0.05%, 18.59%, and -9.45%. The trading strategy includes 15-minute open checks, a stop loss set at 2 times the average true range, and a take profit target at 2 times the risk. The residual target is set for a net 3 times the risk. The horizon exits wait for a specific cadence. In case of collisions, the first event is retained and the expiry is 2 hours. Fees are 25 basis points each, plus 10 or 30 basis points for slippage. The model was trained on data from 2024 to 2025 and tested on known data from 2026, with a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, but this should not be interpreted as within-window misses. The model does not fill all orders. All 48 cells of data are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data: the SOL/USD C91306 managed 1Day manager train slip30bp has a return percentage of 41.689738130383304, a drawdown of 28.186901760039472, and a win rate of 55.55555555555556. The sample size is 9 and the half Kelly is 1.7353182387132127. The symbol is SOL, the time frame is 1Day, the mode is manager, the slippage is 30 basis points, and the window is train. The trade returns percentage is [-3.743604098893616, 17.78938256705438, 13.943486152356478, -10.839553754042752, 22.028455509325173, -9.663162824577709, 10.672500401766527, -10.812062806925704, 13.049613681174232]. The strategy uses 15-minute open checks, a 2ATR stop, and a residual target for a net 3R, with horizon exits waiting for the selected cadence. The first event is retained on collisions, with an expiry of 2 hours and fees of 25 basis points each plus 10/30 basis points slippage. The train data is from 2024-25, with a known 2026 holdout and a forced terminal close. The full sample missed count includes a pre-2024 warmup with no 15-minute history. The model does not fill, and all 48 cells are managed, with timestamps in the managed_results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Day manager holdout slip10bp'. The trading fund manager's notes are as follows: the return percentage is -23.114313182756362, the drawdown is 36.66451078761456, and the win rate is 28.57142857142857, based on a sample of 7 trades. The half Kelly criterion, a measure of optimal bet size, is 0. The symbol being traded is SOL, with a 1Day time frame, in manager mode, and a slippage of 10 basis points. The window is holdout, and 2 events were missed that were greater than 2 hours, out of a full sample. Trade returns percentages are -9.682045501358882, -14.271986248231627, -0.6536125097975076, -7.173238628992884, 0.3496189576871478, 18.03250161457324, and -9.091178942868172. The strategy includes 15-minute open checks, a 2ATR stop, which is a stop-loss order set at 2 times the average true range, a +2R quarter or break-even point at a selected cadence, and a residual target for a net 3R. Horizon exits wait for the cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each, plus 10 or 30 basis points slippage. The training data is from 2024-25, with a known 2026 holdout, and a forced terminal close. The full sample missed count includes a pre-2024 warm-up with no 15-minute history, but this should not be interpreted as within-window misses. The model does not fill orders. All 48 cells are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line for the SOL/USD C91306 managed 1Day manager train slip10bp: return percentage is 46.51, drawdown is 27.83, and win rate is 55.56. The sample size is 9, and the half Kelly is 1.89. The symbol is SOL, the time frame is 1Day, the mode is manager, the slippage is 10 basis points, and the window is train. The number of events missed greater than 2 hours for the full sample is 2. Trade returns percentage is equal to a list of numbers: -3.36, 18.26, 14.40, -10.48, 22.52, -9.30, 11.12, -10.68, and 13.50. The strategy uses 15-minute open checks, a 2 ATR stop, which means a stop loss set at 2 times the average true range, and a quarter of the risk reward ratio at the break even point at a selected cadence. The residual target is for a net 3 risk reward ratio. Horizon exits wait for the cadence. The first event is retained on collisions, and the expiry is 2 hours. Fees are 25 basis points each plus 10 or 30 basis points slippage. The train data is from 2024 to 2025, and the known holdout data is from 2026, with a forced terminal close. The full sample missed count includes pre-2024 warmup data with no 15-minute history, and should not be interpreted as within window misses. The model does not fill orders. All 48 cells are included, and timestamps are managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line for the SOL/USD C91306 managed fund: the return percentage is -24.81196132288119, the drawdown is 37.555842224623824, the win rate is 14.285714285714285, and the sample size is 7. The fund uses a 1Day time frame, immediate15 mode, and a 30 basis point slippage. The trade returns percentages are -9.878101385506211, -14.614214872675657, -1.0502060624336895, -7.543805652131708, -0.05097952297773434, 18.014279062817028, and -9.45408949472466. The fund uses 15-minute open checks, a 2ATR stop, and a residual target for a net 3R return. The model was trained from 2024 to 2025 and tested on a holdout set from 2026, with a forced terminal close. The results are based on all 48 cells and are managed in the results.json file. A basis point, or bp, is a unit of measurement, and ATR refers to average true range, a measure of price volatility.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Day immediate15 train slip30bp'. The trading fund manager's notes are as follows: The SOL/USD trading strategy had a return percentage of 41.52147769142425, a drawdown of 28.203289819312783, and a win rate of 55.55555555555556, based on a sample of 9 trades. The half Kelly criterion, a measure of optimal bet size, was 1.7329408431566498. The strategy skipped 0 trades due to stale prices. The symbol traded was SOL, with a 1-day time frame and immediate 15-minute mode. The slippage, or difference between expected and actual trade prices, was 30 basis points. The strategy was trained on a window of data. The strategy missed 2 events that occurred more than 2 hours after the full sample. Trade returns were as follows: -3.743604098893616, 17.786132986779755, 14.053369818975824, -10.839553754042742, 22.028455509325173, -9.663162824577709, 10.437620762901133, -10.812062806925704, 13.049613681174232. The strategy used 15-minute open checks, a 2 ATR stop, which is a volatility-based stop loss, and a quarter of the potential profit as a benchmark to close the trade, with a residual target for a net 3R, which is a measure of profit. The strategy also used horizon exits, which wait for a specific cadence, or timing. In the event of a collision, the first event was retained, with an expiry of 2 hours. Fees were 25 basis points each, plus 10 or 30 basis points for slippage. The training data was from 2024 to 2025, with 2026 held out for testing, and the strategy forced a terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill, or complete, trades. All 48 cells of data were used, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line: return percentage is -23.58855665335652, drawdown is 36.66451078761456, and win rate is 28.57142857142857. The sample size is 7, half Kelly is 0, and skipped stale price is 0. The symbol is SOL, time frame is 1 Day, mode is immediate 15, slippage is 10 basis points, and window is holdout. There were 2 events missed greater than 2 hours in the full sample. Trade returns percentage is -9.682045501358882, -14.271986248231627, -0.6536125097975076, -7.173238628992884, 0.3496189576871478, 17.30445787162831, and -9.091178942868172. The strategy uses 15-minute open checks, a 2 ATR stop, which means a stop loss set at 2 times the average true range, and a quarter of the risk reward ratio at the selected cadence, with a residual target for a net 3 risk reward ratio. Horizon exits wait for the selected cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each plus 10 or 30 basis points slippage. The model was trained from 2024 to 2025, with a known 2026 holdout, and had a forced terminal close. The full sample missed count includes a pre-2024 warm-up with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill orders. All 48 cells are managed, with timestamps in the managed results json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'SOL/USD C91306 managed 1Day immediate15 train slip10bp'. The trading fund manager's notes for SOL/USD are as follows: the return percentage is 46.256440014381425, the drawdown is 27.844944199800736, and the win rate is 55.55555555555556, based on a sample of 9. The half Kelly criterion, a measure of optimal bet size, is 1.8893496247368675. The symbol is SOL, the time frame is 1Day, the mode is immediate15, and the slippage is 10 basis points. The window is train, and 2 events were missed due to being greater than 2 hours old for the full sample. Trade returns percentages are -3.357805761218824, 18.192229520517312, 14.51049892785683, -10.482196181841786, 22.517548985595436, -9.301090244498134, 10.880257848034258, -10.68103057481975, and 13.502719707269218. The strategy includes 15-minute open checks, a 2ATR stop, and a residual target for a net 3R, with horizon exits waiting for the selected cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each plus 10 or 30 basis points for slippage. The training data is from 2024-25, with 2026 held out, and includes a forced terminal close. The full sample missed count includes pre-2024 warmup data with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill orders. All 48 cells are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line for the ETH/USD C91306 managed 1Hour manager holdout slip30bp: the return percentage is -7.178192826427093, the drawdown is 21.36119085850403, the win rate is 28.57142857142857, the sample size is 7, the half Kelly is 0, the number of skipped stale prices is 1, the symbol is ETH, the time frame is 1 hour, the mode is manager, the slippage in basis points is 30, the window is holdout, and the number of events missed greater than 2 hours for the full sample is 54. The trade returns percentages are -7.436424235093863, -1.3955344967712335, 1.6772121208703306, -1.6655163022845754, 14.516936917923484, -6.476034979495848, and -5.028957018372859. The strategy includes 15-minute open checks, a 2ATR stop, and a residual target for a net 3R, with horizon exits waiting for the selected cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees of 25 basis points each plus 10 or 30 basis points slippage. The training period is 2024-2025, with a known 2026 holdout and a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill all orders. All 48 cells are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line: the ETH/USD C91306 managed 1Hour manager train slip30bp has a return percentage of 35.66754750309045, a drawdown of 34.58914251732863, and a win rate of 50. The sample size is 18, the half Kelly is 0.7886536080660962, and 2 stale prices were skipped. The symbol is ETH, the time frame is 1 hour, the mode is manager, the slippage is 30 basis points, and the window is train. The number of events missed greater than 2 hours for the full sample is 54. Trade returns percentage is equal to a list of 18 numbers: -11.218118222244078, 21.444805013636127, 23.3865030997525, -13.427055620814288, 4.655904386896559, -10.114305910757725, -8.102439044972753, 17.118288275195148, 9.465006845581136, 26.17499553594993, -9.240646238189443, -14.021461364918386, -11.302113717080164, 30.27820522432294, 0.2479365525007271, 14.925794239460854, -8.715254592606048, -11.175337358700165. The strategy uses 15-minute open checks, a 2 ATR stop, which means a stop loss set at 2 times the average true range, and a quarter of the risk reward ratio at the break even point, with a residual target for a net 3 times the risk reward ratio, at a selected cadence. Horizon exits wait for the cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each plus 10 or 30 basis points slippage. The training period is from 2024 to 2025, with a known holdout period in 2026, and a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill, and all 48 cells are used, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line for the ETH/USD C91306 managed 1Hour manager holdout slip10bp: return percentage is -4.761685347058432, drawdown is 21.22734247320084, win rate is 28.57142857142857, sample size is 7, half Kelly is 0, skipped stale price is 1, symbol is ETH, time frame is 1 Hour, mode is manager, slippage in basis points is 10, window is holdout, and events missed greater than 2 hours for full sample is 54. Trade returns percentage is a list of numbers: -7.27887498823624, -1.0003250301967692, 2.084737242171797, -1.2713889303677561, 14.97592401675416, -6.101188301517824, -4.648310411292389. The strategy uses 15-minute open checks, a 2ATR stop, which means a stop-loss order set at 2 times the average true range, and a quarter of the risk reward ratio at the break-even point, with a residual target for a net 3 times the risk reward ratio, at a selected cadence, which refers to the timing of trades. Horizon exits wait for the cadence. The first event is retained on collisions, with an expiry of 2 hours. Fees are 25 basis points each, plus 10 or 30 basis points for slippage. The training data is from 2024 to 2025, and the known holdout data is from 2026, with a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill orders. All 48 cells have timestamps, which are managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line: return percentage is 41.33153378262807, drawdown is 33.799480308373944, win rate is 50, sample size is 18, half Kelly is 0.8641467590204002, skipped stale price is 2, symbol is ETH, time frame is 1 Hour, mode is manager, slippage in basis points is 10, window is train, and events missed greater than 2 hours full sample is 54. Trade returns percentage equals [-10.86227794706186, 21.46168232678498, 22.82812035908528, -13.080068828753056, 5.075368190442653, -9.754041524835767, -7.734111040159009, 17.587701663312316, 9.903745666822417, 25.927236591719492, -8.876880199767234, -13.676856967770545, -10.94661009819986, 30.03340408624624, 0.6497330971897641, 15.386420049917016, -8.349382769715197, -11.428660324324358]. The strategy uses 15-minute open checks, a 2ATR stop, which means a stop loss set at 2 times the average true range, and a quarter of the risk reward ratio at break even at a selected cadence, with a residual target for a net 3 times the risk reward ratio. Horizon exits wait for the selected cadence. The first event is retained on collisions, with an expiry of 2 hours, and fees are 25 basis points each plus 10 or 30 basis points slippage. The training period is 2024-2025, with a known 2026 holdout, and a forced terminal close. The full sample missed count includes a pre-2024 warm-up period with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill orders. All 48 cells are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the headline 'ETH/USD C91306 managed 1Hour immediate15 holdout slip30bp'. The ETH/USD trading fund manager reported the following results: the return percentage was 11.930566976766073%, the drawdown was 22.88487479356266%, and the win rate was 57.14285714285714%. The sample size was 7, and the half Kelly criterion, a measure of optimal bet size, was 2.6922138540248413. The fund manager used a 1Hour time frame, immediate15 mode, and a 30 basis point slippage. The trade returns percentages were -8.077293163146349, -1.838083645999522, 1.6772121208703306, 4.971711322873751, 16.206512192564947, 5.3089517077953285, and -5.028957018372882. The strategy included 15-minute open checks, a 2ATR stop, and a residual target for a net 3R return. The model did not fill all orders and managed 48 cells. The training data was from 2024-2025, and the known holdout data was from 2026. The fund manager forced a terminal close and reported 27 missed events, including a pre-2024 warm-up period with no 15-minute history.
https://runway.adityaperswal.workers.dev/data/crypto-bars
A news scan found the following data line: the ETH/USD C91306 managed 1Hour immediate15 train slip30bp has a return percentage of 39.68391678335925, a drawdown of 31.471634415770154, and a win rate of 44.44444444444444. The sample size is 18, the half Kelly is 0.7694123471500396, and 1 stale price was skipped. The symbol is ETH, the time frame is 1 hour, the mode is immediate 15, the slippage is 30 basis points, and the window is train. There were 27 events missed greater than 2 hours in the full sample. Trade returns percentage is a list of numbers including -10.982894179811852, 21.77969716047312, and 17.118288275195106. The strategy includes 15-minute open checks, a 2 ATR stop, which means a stop loss set at 2 times the average true range, and a residual target for a net 3R, which means three times the risk. Horizon exits wait for a specific cadence, or timing. The first event is retained on collisions, and expiry is 2 hours, with fees of 25 basis points each plus 10 or 30 basis points slippage. The train data is from 2024 to 2025, with 2026 held out, and there is a forced terminal close. The full sample missed count includes pre-2024 warm-up with no 15-minute history, and should not be interpreted as within-window misses. The model does not fill, and all 48 cells are included, with timestamps managed in the results.json file.
https://runway.adityaperswal.workers.dev/data/crypto-bars