Notes

A news scan found the headline 'SOL/USD C91306 delayed HIGH 1Day train delay1 slip10bp'. The return was 25.420885921607205, the drawdown was 30.301096474759305, the win rate was 57.14285714285714, and the sample size was 7. The mean trade percentage was 4.488646425436358 and the half Kelly fraction was 1.259588472030996. There were 3 skipped entries. The trades were as follows: the first trade entered on 2024-09-21 00:00:00+00:00 and exited on 2024-10-11 00:00:00+00:00 with a return percentage of -5.940878090250434 due to time, the second trade entered on 2024-10-22 00:00:00+00:00 and exited on 2024-11-11 00:00:00+00:00 with a return percentage of 25.827002614712647 due to target, the third trade entered on 2024-11-12 00:00:00+00:00 and exited on 2024-12-02 00:00:00+00:00 with a return percentage of 5.817015718997176 due to time, the fourth trade entered on 2025-01-20 00:00:00+00:00 and exited on 2025-01-29 00:00:00+00:00 with a return percentage of -10.819310381242374 due to stop, the fifth trade entered on 2025-04-21 00:00:00+00:00 and exited on 2025-05-11 00:00:00+00:00 with a return percentage of 28.014891148442487 due to time, the sixth trade entered on 2025-07-12 00:00:00+00:00 and exited on 2025-08-01 00:00:00+00:00 with a return percentage of 4.887419249289904 due to time, and the seventh trade entered on 2025-09-12 00:00:00+00:00 and exited on 2025-09-26 00:00:00+00:00 with a return percentage of -16.365615281894897 due to stop. The symbol was SOL, the time frame was 1 day, the delay was 1, and the slippage was 10 basis points. The window was train and the days were 731. This data was retrieved on September 13 at 06:03-04 UTC and was computed through September 12. The model was derived and not based on fills. The train data was from 2024 to 2025 and the holdout data was from 2026. The full gross was 3R, 2ATR, and 20 days, with exits at sampled bar opens and including terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'SOL/USD C91306 delayed HIGH 1Day holdout delay0 slip30bp'. The return was -26.398607832625544, the drawdown was 35.29850243033073, and the win rate was 14.285714285714285. The mean trade percentage was -3.8927833838910852 and the half Kelly fraction was 0. There were 7 samples and 0 skipped entries. The trades were as follows: from 2026-01-06 to 2026-01-21 with a return percentage of -9.878101385506211 due to a stop, from 2026-03-17 to 2026-03-28 with a return percentage of -14.596053157714273 due to a stop, from 2026-04-17 to 2026-05-07 with a return percentage of -1.0502060624336895 due to time, from 2026-05-09 to 2026-05-17 with a return percentage of -6.778889164655933 due to a stop, from 2026-07-02 to 2026-07-22 with a return percentage of -0.050979522977723235 due to time, from 2026-08-20 to 2026-08-25 with a return percentage of 14.58604099099845 due to a target, and from 2026-08-28 to 2026-09-02 with a return percentage of -9.481295384948218 due to a stop. The symbol was SOL, the time frame was 1 day, the delay was 0, and the slippage was 30 basis points. The window was a holdout and the days were 255. This was retrieved on September 13 at 06:03 UTC and computed through September 12. It was model-derived and not based on actual fills. The model was trained from 2024 to 2025 and tested on 2026 data. The strategy used a 3R, 2ATR, and 20-day period, with exits at sampled bar opens and terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'SOL/USD C91306 delayed HIGH 1Day holdout delay0 slip10bp'. The report shows the results of a trading model for the SOL/USD symbol. The model had a return of -29.80% and a drawdown of 39.03%. The win rate was 28.57%. The model made 7 trades. The trades were as follows: from 2026-01-06 to 2026-01-26 with a return of -14.60%, from 2026-03-17 to 2026-03-28 with a return of -14.25%, from 2026-04-17 to 2026-05-07 with a return of -0.65%, from 2026-05-09 to 2026-05-18 with a return of -8.04%, from 2026-07-02 to 2026-07-22 with a return of 0.35%, from 2026-08-20 to 2026-08-25 with a return of 15.05%, and from 2026-08-28 to 2026-09-02 with a return of -9.12%. The model used a 1-day time frame, a delay of 0, and a slippage of 10 basis points. The results were computed through September 12 and retrieved on September 13 at 06:03 UTC. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Hour holdout delay4 slip30bp'. The return was -9.16236274477933, the drawdown was 22.91093510051304, and the win rate was 42.857142857142854. This means the return is the profit or loss, the drawdown is the largest decrease in value, and the win rate is the percentage of successful trades. The sample size was 7, the mean trade percentage was -1.1056681534807686, and the half Kelly fraction was 0. The half Kelly fraction is a measure of the optimal bet size. There were 4 skipped entries. The trades were as follows: a trade from 2026-01-14 03:00:00+00:00 to 2026-01-20 07:00:00+00:00 with a return percentage of -7.36266711393786, a trade from 2026-03-04 20:00:00+00:00 to 2026-03-08 19:00:00+00:00 with a return percentage of -12.480106344745124, a trade from 2026-04-08 04:00:00+00:00 to 2026-04-28 04:00:00+00:00 with a return percentage of 1.2589188467300838, a trade from 2026-07-14 17:00:00+00:00 to 2026-08-03 17:00:00+00:00 with a return percentage of -1.5547742877098925, a trade from 2026-08-19 19:00:00+00:00 to 2026-08-21 02:00:00+00:00 with a return percentage of 11.461501860743617, a trade from 2026-08-21 06:00:00+00:00 to 2026-09-10 06:00:00+00:00 with a return percentage of 3.6716645934749437, and a trade from 2026-09-11 18:00:00+00:00 to 2026-09-13 00:00:00+00:00 with a return percentage of -2.73421462892115. The symbol was ETH, the time frame was 1 hour, the delay was 4, and the slippage was 30 basis points. The data was retrieved on September 13 at 06:03-04 UTC and was computed through September 12. The model was trained on data from 2024-25 and tested on data from 2026. The exits were at sampled bar opens and included terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Hour holdout delay4 slip10bp'. The return was 2.4861778723921333, the drawdown was 19.11969765402035, and the win rate was 42.857142857142854. The sample size was 7, the mean trade percentage was 0.5093714812180413, and the half Kelly fraction was 1.2278289411408494. There were 4 skipped entries. The trades were as follows: a trade entered on 2026-01-14 03:00:00+00:00 and exited on 2026-01-20 07:00:00+00:00 with a return percentage of -6.991374082451529 due to a stop, a trade entered on 2026-03-04 20:00:00+00:00 and exited on 2026-03-24 20:00:00+00:00 with a return percentage of -3.5986455236284076 due to time, a trade entered on 2026-04-08 04:00:00+00:00 and exited on 2026-04-28 04:00:00+00:00 with a return percentage of 1.6647674368430065 due to time, a trade entered on 2026-07-14 17:00:00+00:00 and exited on 2026-08-03 17:00:00+00:00 with a return percentage of -1.1602030586890022 due to time, a trade entered on 2026-08-19 19:00:00+00:00 and exited on 2026-08-21 02:00:00+00:00 with a return percentage of 11.90824269006776 due to a target, a trade entered on 2026-08-21 06:00:00+00:00 and exited on 2026-09-10 06:00:00+00:00 with a return percentage of 4.087183536291272 due to time, and a trade entered on 2026-09-11 18:00:00+00:00 and exited on 2026-09-13 00:00:00+00:00 with a return percentage of -2.344370629906811 due to a boundary. The symbol was ETH, the time frame was 1 hour, the delay was 4, and the slippage was 10 basis points. The window was a holdout. The data was retrieved on September 13 at 06:03-04 UTC and computed through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 and applied to a known 2026 holdout. The gross return was calculated using a 3R, 2ATR, and 20-day formula, with exits at sampled bar opens, including terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Hour holdout delay2 slip30bp'. The return was -7.3096712370086445, the drawdown was 22.58129595322985, the win rate was 33.33333333333333, the sample size was 6, and the mean trade percentage was -1.047269795883162. The half Kelly fraction, which is a measure of the optimal bet size, was 0. There were 4 skipped entries. The trades were as follows: on January 5, 2026, at 03:00:00, a trade was entered and exited on January 20, 2026, at 23:00:00, with a return percentage of -8.65762038476734, due to a stop, which is an order to sell when a certain price is reached; on March 4, 2026, at 18:00:00, a trade was entered and exited on March 24, 2026, at 18:00:00, with a return percentage of -2.8945602063975784, due to time, meaning the trade was held for a certain amount of time; on April 8, 2026, at 02:00:00, a trade was entered and exited on April 28, 2026, at 02:00:00, with a return percentage of 1.7934412280117717, also due to time; on July 14, 2026, at 15:00:00, a trade was entered and exited on August 3, 2026, at 15:00:00, with a return percentage of -1.587183198304909, again due to time; on August 19, 2026, at 17:00:00, a trade was entered and exited on August 21, 2026, at 02:00:00, with a return percentage of 11.538338765654931, due to a target, which is a predetermined price level; and on August 22, 2026, at 00:00:00, a trade was entered and exited on August 23, 2026, at 05:00:00, with a return percentage of -6.476034979495848, due to a stop. The symbol was ETH, the time frame was 1 hour, the delay was 2, the slippage was 30 basis points, and the window was holdout. The data was retrieved on September 13 at 06:03-04 UTC and computed complete calendar bars through September 12. The model was derived, not from actual fills, and was trained on data from 2024-2025 and tested on data from 2026. The gross return was calculated using a 3R, 2ATR, 20-day model, and exits were at sampled bar opens, including terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Hour holdout delay2 slip10bp'. This is a report on an automated trading fund manager's performance. The return was -5.058183811965861, the drawdown was 22.270999613474316, and the win rate was 33.33333333333333. A drawdown refers to the difference between the highest and lowest values of an investment during a certain period. The sample size was 6, and the mean trade percentage was -0.650664474608174. The half Kelly fraction, which is a measure of the optimal bet size, was 0. There were 4 skipped entries. The trades are as follows: a trade entered on 2026-01-05 03:00:00+00:00 and exited on 2026-01-20 23:00:00+00:00 with a return percentage of -8.291517562353112, another trade entered on 2026-03-04 18:00:00+00:00 and exited on 2026-03-24 18:00:00+00:00 with a return percentage of -2.5053588769603063, a trade entered on 2026-04-08 02:00:00+00:00 and exited on 2026-04-28 02:00:00+00:00 with a return percentage of 2.2014321988384156, a trade entered on 2026-07-14 15:00:00+00:00 and exited on 2026-08-03 15:00:00+00:00 with a return percentage of -1.192741865107716, a trade entered on 2026-08-19 17:00:00+00:00 and exited on 2026-08-21 02:00:00+00:00 with a return percentage of 11.98538755945151, and a trade entered on 2026-08-22 00:00:00+00:00 and exited on 2026-08-23 05:00:00+00:00 with a return percentage of -6.101188301517835. The symbol was ETH, the time frame was 1 hour, the delay was 2, and the slippage in basis points was 10. The window was holdout, and the days were 255. This report was retrieved on September 13 at 06:03-04 UTC and computed complete calendar bars through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 with a known 2026 holdout. The report includes a full gross 3R, 2 ATR, and 20-day period, with exits at sampled bar opens, and includes terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Hour holdout delay0 slip30bp'. The return was 10.59334598278349, the drawdown was 22.646299516825607, and the win rate was 57.14285714285714. The sample size was 7, the mean trade percentage was 1.6891732222168443, and the half Kelly fraction was 2.5415440053830203. There were 0 skipped entries. The trades were as follows: a trade entered on 2026-01-05 01:00:00+00:00 and exited on 2026-01-20 23:00:00+00:00 with a return percentage of -8.04698787657111 due to a stop, a trade entered on 2026-03-04 16:00:00+00:00 and exited on 2026-03-24 16:00:00+00:00 with a return percentage of -1.838083645999511 due to time, a trade entered on 2026-04-08 00:00:00+00:00 and exited on 2026-04-28 00:00:00+00:00 with a return percentage of 1.6772121208703306 due to time, a trade entered on 2026-07-06 22:00:00+00:00 and exited on 2026-07-26 22:00:00+00:00 with a return percentage of 4.971711322873751 due to time, a trade entered on 2026-08-19 15:00:00+00:00 and exited on 2026-08-19 22:00:00+00:00 with a return percentage of 14.78036594492198 due to a target, a trade entered on 2026-08-20 17:00:00+00:00 and exited on 2026-09-09 17:00:00+00:00 with a return percentage of 5.308951707795351 due to time, and a trade entered on 2026-09-11 14:00:00+00:00 and exited on 2026-09-13 00:00:00+00:00 with a return percentage of -5.028957018372882 due to a boundary. The symbol was ETH, the time frame was 1 hour, the delay was 0, and the slippage was 30 basis points. The window was a holdout. The data was retrieved on September 13 at 06:03-04 UTC and computed through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 with a known 2026 holdout. The gross return was calculated using a 3R, 2ATR, and 20-day period, with exits at sampled bar opens, including terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Hour holdout delay0 slip10bp'. The return was 13.733733534189312, the drawdown was 22.33626371317833, and the win rate was 57.14285714285714. The sample size was 7, the mean trade percentage was 2.0967462839483537, and the half Kelly fraction was 3.2111305573046494. There were 0 skipped entries. The trades were as follows: a trade that entered on 2026-01-05 01:00:00+00:00 and exited on 2026-01-20 23:00:00+00:00 with a return percentage of -7.678437621917322 due to a stop, a trade that entered on 2026-03-04 16:00:00+00:00 and exited on 2026-03-24 16:00:00+00:00 with a return percentage of -1.4446479288420577 due to time, a trade that entered on 2026-04-08 00:00:00+00:00 and exited on 2026-04-28 00:00:00+00:00 with a return percentage of 2.084737242171797 due to time, a trade that entered on 2026-07-06 22:00:00+00:00 and exited on 2026-07-26 22:00:00+00:00 with a return percentage of 5.392440889487182 due to time, a trade that entered on 2026-08-19 15:00:00+00:00 and exited on 2026-08-19 22:00:00+00:00 with a return percentage of 15.240408874689981 due to a target, a trade that entered on 2026-08-20 17:00:00+00:00 and exited on 2026-09-09 17:00:00+00:00 with a return percentage of 5.7310329433412965 due to time, and a trade that entered on 2026-09-11 14:00:00+00:00 and exited on 2026-09-13 00:00:00+00:00 with a return percentage of -4.6483104112924 due to a boundary. The symbol was ETH, the time frame was 1 hour, the delay was 0, and the slippage was 10 basis points. The window was a holdout. The data was retrieved on September 13 at 06:03-04 UTC and was computed through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 with a holdout in 2026. The gross return was 3R, the average true range was 2ATR, and the period was 20 days. Exits occurred at the opening of sampled bars and included terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Day holdout delay1 slip30bp'. The return was -28.89, the drawdown was 32.89, the win rate was 20, the sample size was 5, and the mean trade percentage was -6.36. The half Kelly fraction was 0, and 3 entries were skipped. There were 5 trades: from 2026-01-07 to 2026-01-10 with a return percentage of -7.46, from 2026-01-15 to 2026-01-21 with a return percentage of -13.42, from 2026-03-18 to 2026-03-23 with a return percentage of -12.34, from 2026-07-16 to 2026-08-05 with a return percentage of -3.53, and from 2026-08-21 to 2026-09-10 with a return percentage of 4.94. The symbol was ETH, the time frame was 1 day, the delay was 1 day, and the slippage was 30 basis points. The window was a holdout period. This data was retrieved on September 13 at 06:03 UTC and computed through September 12. The model was trained on data from 2024 to 2025 and applied to known data from 2026. The model used a gross 3R, 2 ATR, and 20-day period, exited at sampled bar opens, and included terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Day holdout delay1 slip10bp'. The return was -27.46%, the drawdown was 31.94%, and the win rate was 20%. The sample size was 5, and the mean trade percentage was -5.99%. The half Kelly fraction was 0, and 3 entries were skipped. There were 5 trades: from 2026-01-07 to 2026-01-10 with a return of -7.09% due to a stop, from 2026-01-15 to 2026-01-21 with a return of -13.07% due to a stop, from 2026-03-18 to 2026-03-23 with a return of -11.98% due to a stop, from 2026-07-16 to 2026-08-05 with a return of -3.15% due to time, and from 2026-08-21 to 2026-09-10 with a return of 5.36% due to time. The symbol was ETH, the time frame was 1 day, the delay was 1 day, and the slippage was 10 basis points. The window was a holdout period. This data was retrieved on September 13 at 06:03 UTC and computed through September 12. It is a model-derived report, not based on actual fills, and was trained on data from 2024 to 2025 and tested on data from 2026. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Day holdout delay0 slip30bp'. The return was -15.965104127502231, the drawdown was 27.24834401499575, and the win rate was 20. The sample size was 5, and the mean trade percentage was -3.0941328333627305. The half Kelly fraction was 0, and there were 0 skipped entries. There were 5 trades: one that entered on 2026-01-06 00:00:00+00:00 and exited on 2026-01-21 00:00:00+00:00 with a return percentage of -10.06109385592019 due to a stop, one that entered on 2026-03-17 00:00:00+00:00 and exited on 2026-03-22 00:00:00+00:00 with a return percentage of -12.346100593109377 due to a stop, one that entered on 2026-04-12 00:00:00+00:00 and exited on 2026-05-02 00:00:00+00:00 with a return percentage of -0.5971873745373446 due to time, one that entered on 2026-07-15 00:00:00+00:00 and exited on 2026-08-04 00:00:00+00:00 with a return percentage of -2.8605724458390602 due to time, and one that entered on 2026-08-20 00:00:00+00:00 and exited on 2026-08-22 00:00:00+00:00 with a return percentage of 10.394290102592318 due to a target. The symbol was ETH, the time frame was 1 day, the delay was 0, and the slippage was 30 basis points. The window was a holdout, and the days were 255. This was retrieved on September 13 at 06:03-04UTC and computed through September 12. It was model-derived and not based on actual fills. The model was trained from 2024 to 2025 and tested on 2026 holdout data. The full gross was 3R, 2ATR, and 20 days, with exits at sampled bar opens, including terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'ETH/USD C91306 delayed HIGH 1Day holdout delay0 slip10bp'. The return was -14.267479191675875, the drawdown was 26.66399395121769, and the win rate was 20. The sample size was 5, and the mean trade percentage was -2.705731396618394. The half Kelly fraction was 0, and there were 0 skipped entries. There were 5 trades: from 2026-01-06 to 2026-01-21 with a return percentage of -9.70061619458824, stopped with a 2.5202710029793924 percent overshoot; from 2026-03-17 to 2026-03-22 with a return percentage of -11.994781302943625, stopped with a 0.895813778525584 percent overshoot; from 2026-04-12 to 2026-05-02 with a return percentage of -0.19877811027920744, exited on time; from 2026-07-15 to 2026-08-04 with a return percentage of -2.4712348925030936, exited on time; and from 2026-08-20 to 2026-08-22 with a return percentage of 10.836753517222197, exited on target. The data covers 255 days, and the symbol is ETH. The time frame is 1 day, the delay is 0, and the slippage is 10 basis points. The window is a holdout. This data was retrieved on September 13 at 06:03 UTC and computed through September 12. It is model-derived and does not include fills. The model was trained from 2024 to 2025 and tested on 2026 holdout data. The gross return is 3R, the average true range is 2ATR, and the average duration is 20 days. Exits occur at the open of sampled bars and include terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Hour holdout delay4 slip30bp'. The report is about the performance of a trading model for the Bitcoin symbol, which is a digital currency. The model was trained on data from 2024 to 2025 and tested on data from 2026. The results show a return of 0.665531823142814, a drawdown of 14.89383800138655, and a win rate of 50. The model made 6 trades, with a mean trade percentage of 0.3055213597247953. The half Kelly fraction, which is a measure of the optimal bet size, is 0.5123497529498771. The model skipped 4 entries. The trades are listed below: a trade that entered on 2026-01-04 05:00:00+00:00 and exited on 2026-01-24 05:00:00+00:00 with a return percentage of -2.894548306054401, a trade that entered on 2026-04-14 03:00:00+00:00 and exited on 2026-05-04 03:00:00+00:00 with a return percentage of 6.644411774958381, a trade that entered on 2026-05-06 13:00:00+00:00 and exited on 2026-05-16 11:00:00+00:00 with a return percentage of -6.231867478426367, a trade that entered on 2026-07-21 11:00:00+00:00 and exited on 2026-07-31 15:00:00+00:00 with a return percentage of -6.363769214355508, a trade that entered on 2026-08-19 19:00:00+00:00 and exited on 2026-08-21 08:00:00+00:00 with a return percentage of 10.346202681259719, and a trade that entered on 2026-08-21 12:00:00+00:00 and exited on 2026-09-10 12:00:00+00:00 with a return percentage of 0.3326987009669491. The model used a 1Hour time frame, a delay of 4, and a slippage of 30 basis points. The report was retrieved on September 13 at 06:03-04UTC and computed complete calendar bars through September 12. The model-derived results are not based on actual fills and include terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Hour holdout delay4 slip10bp'. This is a report on a trading model for the Bitcoin symbol, which is a digital currency. The model was trained on data from 2024 to 2025 and tested on data from 2026. The report shows the results of six trades. The first trade started on January 4, 2026, and ended on January 24, 2026, with a return of -2.5053 percent. The second trade started on April 14, 2026, and ended on May 4, 2026, with a return of 7.0718 percent. The third trade started on May 6, 2026, and ended on May 17, 2026, with a return of -6.1207 percent. The fourth trade started on July 21, 2026, and ended on August 1, 2026, with a return of -6.2071 percent. The fifth trade started on August 19, 2026, and ended on August 21, 2026, with a return of 10.7885 percent. The sixth trade started on August 21, 2026, and ended on September 10, 2026, with a return of 0.7348 percent. The model had a return of 2.5818 percent and a drawdown of 13.9648 percent. The win rate was 50 percent. The model used a one-hour time frame and a delay of four hours. The slippage was 10 basis points, which is a measure of the difference between the expected price of a trade and the actual price. The report was retrieved on September 13 at 06:03 UTC and was computed using complete calendar bars through September 12. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Hour holdout delay2 slip30bp'. This is a report on a trading model for the Bitcoin symbol, which is a digital currency. The model was trained on data from 2024 to 2025 and tested on data from 2026. The report shows the results of six trades. The return on investment was -2.4671337768238715 percent, and the biggest loss was 17.01318308981469 percent. The model won 33.33333333333333 percent of the time. The average trade lost 0.23990301556031843 percent. The model used a delay of 2 and a slippage of 30 basis points, which is a measure of the difference between the expected price and the actual price of a trade. The model was tested on a holdout set, which is a set of data that is not used to train the model. The results are based on a one-hour time frame and include six trades. The trades were made on the following dates: January 4 to 24, 2026, April 14 to May 4, 2026, May 6 to 16, 2026, July 21 to 31, 2026, August 19 to 21, 2026, and August 21 to September 10, 2026. The reasons for exiting the trades were time, stop, or target, which refers to a predetermined price level. The model also used a trend indicator, which is a measure of the direction of the market. The report was retrieved on September 13 at 06:03 UTC and includes data up to September 12. The model is based on a gross return of three times the risk, two times the average true range, and a 20-day time frame. The exits were made at the open of the sampled bars, and the model includes terminal liquidation, which means that all positions are closed at the end of the trading period. The model has 60 predetermined cells, which are used to make trading decisions. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Hour holdout delay2 slip10bp'. This is a report on a trading model for the Bitcoin symbol, which is a digital currency, and the US dollar. The report includes the following information: the return, or profit, was -0.3715, the drawdown, or largest loss, was 15.9056, and the win rate, or percentage of successful trades, was 33.3333. The sample size was 6 trades. The mean trade percentage, or average profit per trade, was 0.1169. The half Kelly fraction, a measure of risk, was 0.1962. There were 4 skipped entries, which are trades that were not made. The trades are listed below: a trade that started on January 4, 2026, and ended on January 24, 2026, had a return of -2.3535, a trade that started on April 14, 2026, and ended on May 4, 2026, had a return of 5.1161, a trade that started on May 6, 2026, and ended on May 17, 2026, had a return of -5.9859, a trade that started on July 21, 2026, and ended on August 1, 2026, had a return of -6.1613, a trade that started on August 19, 2026, and ended on August 21, 2026, had a return of 10.6834, and a trade that started on August 21, 2026, and ended on September 10, 2026, had a return of -0.5970. The report also includes the following information: the symbol is BTC, the time frame is 1 hour, the delay is 2, the slippage in basis points is 10, and the window is holdout. The report was retrieved on September 13 at 06:03-04 UTC and was computed through September 12. The model is derived, not from actual trades, and includes terminal liquidation, which means that all positions are closed at the end of the trading period. The report is based on a training period of 2024-2025 and a holdout period of 2026. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Hour holdout delay0 slip30bp'. The report shows the results of a trading model for the symbol BTC, which is a type of cryptocurrency, with a time frame of 1 hour. The model had a return of 1.8313690835902605, a drawdown of 15.627950319962947, and a win rate of 50. The mean trade percentage was 0.4475923423890635 and the half Kelly fraction was 1.1118445833222594. The model made 6 trades, with the following details: it entered the market on 2026-01-04 01:00:00+00:00 and exited on 2026-01-24 01:00:00+00:00 with a return percentage of -3.239170441482997, entered on 2026-04-10 21:00:00+00:00 and exited on 2026-04-30 21:00:00+00:00 with a return percentage of 3.043756815599874, entered on 2026-05-06 09:00:00+00:00 and exited on 2026-05-17 02:00:00+00:00 with a return percentage of -6.209396589424299, entered on 2026-07-14 16:00:00+00:00 and exited on 2026-08-03 16:00:00+00:00 with a return percentage of -2.7348661895421067, entered on 2026-08-19 15:00:00+00:00 and exited on 2026-08-21 01:00:00+00:00 with a return percentage of 10.600286039543171, and entered on 2026-08-21 08:00:00+00:00 and exited on 2026-09-10 08:00:00+00:00 with a return percentage of 1.2249444196407389. The model used a delay of 0 and a slippage of 30 basis points, which is a measure of the difference between the expected price of a trade and the actual price. The report was retrieved on September 13 at 06:03-04 UTC and was computed using complete calendar bars through September 12. The model was trained on data from 2024 to 2025 and tested on data from 2026. The model uses a type of technical analysis called gross 3R, 2ATR, and 20d, which are measures of volatility and risk. The model exits trades at the opening of sampled bars and includes terminal liquidation, which means that all positions are closed at the end of the trading period. All parameters were predetermined and used 60 cells. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Hour holdout delay0 slip10bp'. This is a report on a trading model for Bitcoin, symbol BTC. The model was trained on data from 2024 to 2025 and tested on data from 2026. The report covers 255 days, from January 4, 2026, to September 10, 2026. The model made six trades during this time. The trades were as follows: a trade that started on January 4, 2026, and ended on January 24, 2026, with a return of -2.8513503195446877 percent, a trade that started on April 10, 2026, and ended on April 30, 2026, with a return of 3.456759086410699 percent, a trade that started on May 6, 2026, and ended on May 18, 2026, with a return of -6.230542604572598 percent, a trade that started on July 14, 2026, and ended on August 3, 2026, with a return of -2.34502480200105 percent, a trade that started on August 19, 2026, and ended on August 21, 2026, with a return of 11.04357509167273 percent, and a trade that started on August 21, 2026, and ended on September 10, 2026, with a return of 1.6306568392958631 percent. The model had a return of 3.865085512335696 percent and a drawdown of 14.969439154901975 percent. The win rate was 50 percent. The model used a time frame of one hour and a delay of zero. Slippage, which is the difference between the expected price and the actual price, was 10 basis points, which is a measure of the difference in price, with one basis point being equal to 0.01 percent. The model also used a holdout window, which is a period of time used to test the model. The report was retrieved on September 13, 2026, at 06:03 UTC. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Day holdout delay1 slip30bp'. The return was -0.21227381922949462, the drawdown was 13.034926770540778, and the win rate was 40. The sample size was 5, the mean trade percentage was 0.19279753982137687, and the half Kelly fraction was 0.22387335315499568. There was 1 skipped entry. The trades were as follows: a trade that entered on 2026-01-06 00:00:00+00:00 and exited on 2026-01-21 00:00:00+00:00 with a return percentage of -6.91492366781723 due to a stop, a trade that entered on 2026-04-15 00:00:00+00:00 and exited on 2026-05-05 00:00:00+00:00 with a return percentage of 6.470688171922179 due to time, a trade that entered on 2026-05-07 00:00:00+00:00 and exited on 2026-05-19 00:00:00+00:00 with a return percentage of -6.5541896764442225 due to a stop, a trade that entered on 2026-07-16 00:00:00+00:00 and exited on 2026-08-05 00:00:00+00:00 with a return percentage of -2.1296073677002147 due to time, and a trade that entered on 2026-08-21 00:00:00+00:00 and exited on 2026-09-04 00:00:00+00:00 with a return percentage of 10.092020239146372 due to a target. The symbol was BTC, the time frame was 1 day, the delay was 1, and the slippage was 30 basis points. The window was a holdout. This data was retrieved on September 13 at 06:03-04UTC and computed complete calendar bars through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 with a known 2026 holdout. The gross return was 3R, the average true range was 2ATR, and the exit was at the sampled bar opens, including terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Day holdout delay1 slip10bp'. The return was -0.2686556274030494, the drawdown was 13.776620128535388, and the win rate was 40. The sample size was 5, the mean trade percentage was 0.21390039637057848, and the half Kelly fraction was 0.2250106253509645. There was 1 skipped entry. The trades were as follows: a trade entered on 2026-01-06 00:00:00+00:00 and exited on 2026-01-26 00:00:00+00:00 with a return percentage of -8.444199515169505 due to a stop, a trade entered on 2026-04-15 00:00:00+00:00 and exited on 2026-05-05 00:00:00+00:00 with a return percentage of 6.8974256798382205 due to time, a trade entered on 2026-05-07 00:00:00+00:00 and exited on 2026-05-19 00:00:00+00:00 with a return percentage of -6.179656244686182 due to a stop, a trade entered on 2026-07-16 00:00:00+00:00 and exited on 2026-08-05 00:00:00+00:00 with a return percentage of -1.7373400858053833 due to time, and a trade entered on 2026-08-21 00:00:00+00:00 and exited on 2026-09-04 00:00:00+00:00 with a return percentage of 10.533272147675742 due to a target. The symbol was BTC, the time frame was 1 day, the delay was 1, and the slippage was 10 basis points. The window was a holdout. The data was retrieved on September 13 at 06:03-04UTC and computed complete calendar bars through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 with a known 2026 holdout. The gross return was 3R, the average true range was 2ATR, and the period was 20 days. Exits occurred at sampled bar opens and included terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Day holdout delay0 slip30bp'. The return was -12.248870136238835, the drawdown was 25.93474703271117, and the win rate was 33.33333333333333. This was based on a sample of 6 trades. The mean trade percentage was -1.8422946456240648 and the half Kelly fraction was 0. There were 0 skipped entries. The trades were as follows: from 2026-01-05 00:00:00+00:00 to 2026-01-25 00:00:00+00:00 with a return percentage of -3.6543182801291763 due to time, from 2026-03-17 00:00:00+00:00 to 2026-03-28 00:00:00+00:00 with a return percentage of -12.3469990171594 due to a stop, from 2026-04-14 00:00:00+00:00 to 2026-05-04 00:00:00+00:00 with a return percentage of 4.363714514214911 due to time, from 2026-05-06 00:00:00+00:00 to 2026-05-23 00:00:00+00:00 with a return percentage of -7.737911679736986 due to a stop, from 2026-07-15 00:00:00+00:00 to 2026-08-04 00:00:00+00:00 with a return percentage of -3.4556311216628277 due to time, and from 2026-08-20 00:00:00+00:00 to 2026-08-22 00:00:00+00:00 with a return percentage of 11.777377710729088 due to a target. The symbol was BTC, the time frame was 1 day, the delay was 0, and the slippage was 30 basis points. The window was a holdout. This data was retrieved on September 13 at 06:03-04 UTC and was computed through September 12. The model was derived, not from actual fills, and was trained from 2024 to 2025 with a holdout in 2026. The gross return was 3R, the average true range was 2ATR, and the period was 20 days. Exits occurred at the opening of sampled bars and included terminal liquidation. All 60 cells were predetermined. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the headline 'BTC/USD C91306 delayed HIGH 1Day holdout delay0 slip10bp'. This is a report on a trading model for the Bitcoin to US Dollar currency pair, symbol BTC. The model had a return of -10.117357948740391 and a drawdown of 24.74016665003479. The win rate was 33.33333333333333 percent. The model made 6 trades. The mean trade percentage was -1.4488758062713305. The half Kelly fraction, which is a measure of risk, was 0. The model skipped 0 entries. The trades were as follows: the model entered the market on 2026-01-05 00:00:00+00:00 and exited on 2026-01-25 00:00:00+00:00 with a return percentage of -3.2681620823856283 due to a time limit, it entered on 2026-03-17 00:00:00+00:00 and exited on 2026-03-28 00:00:00+00:00 with a return percentage of -11.995683327902462 due to a stop, which is a price level that triggers a sale, it entered on 2026-04-14 00:00:00+00:00 and exited on 2026-05-04 00:00:00+00:00 with a return percentage of 4.782007212546557 due to a time limit, it entered on 2026-05-06 00:00:00+00:00 and exited on 2026-05-23 00:00:00+00:00 with a return percentage of -7.368122639007346 due to a stop, it entered on 2026-07-15 00:00:00+00:00 and exited on 2026-08-04 00:00:00+00:00 with a return percentage of -3.0686785802089056 due to a time limit, and it entered on 2026-08-20 00:00:00+00:00 and exited on 2026-08-22 00:00:00+00:00 with a return percentage of 12.225384579329802 due to reaching a target price. The model used a 1-day time frame and had a delay of 0 and a slippage of 10 basis points, which is a measure of the difference between the expected price and the actual price. The model was trained on data from 2024 to 2025 and tested on data from 2026. The report was retrieved on September 13 at 06:03-04 UTC and was computed using complete calendar bars through September 12. https://runway.adityaperswal.workers.dev/data/crypto-bars

A news scan found the data line for SFIX R91306, which includes various metrics. The train bootstrap mean 95 percent is between -19.51 and 32.15 percent. The train bootstrap positive percent is 63.92 percent. The half Kelly scenario dollars amount to 196.92 dollars. The train missing 2 mean bounds percent is between -21.60 and 78.40 percent. The holdout missing 1 mean bounds percent is between -33.01 and 166.99 percent. The RSI return correlation train is -0.70, and the RV return correlation train is 0.31. There is one entry negative intrinsic gap row, which occurred on March 11, 2026, with a lag of 2 and an entry intrinsic gap of -0.04. The MC cost sensitivity analysis includes two scenarios: one with stress set to false, which has a cash p05 p50 p95 of 66.39, 123.23, and 184.98, a loss probability percent of 27.24, a median drawdown percent of 19.58, and a below 25 percent of 0.33; and one with stress set to true, which has a cash p05 p50 p95 of 11.76, 24.83, and 77.42, a loss probability percent of 99.27, a median drawdown percent of 75.83, and a below 25 percent of 51.66. Note that missing bounds assume unobserved returns are between -100 percent and 300 percent, and are not hard bounds. The holdout N1 bootstrap is degenerate. The scenario Kelly is from censored pre-event marks and ignores the unobserved full premium tail, and does not include live sizing, which refers to the process of adjusting the size of a trade based on market conditions. https://runway.adityaperswal.workers.dev/data/option-bars ; research/extra.json NumPyseed913061

A descriptive trade replay for SFIX shows the following statistics: the phase is all, block is 3, sample is 7, paths are 20000, and opportunities are 20. The cash percentiles are 95.9078891393956, 144.1308856550751, and 199.54638518822176. The loss probability is 7.704999999999999 percent, the median drawdown is 12.852735324118386 percent, and the cash below ticket percentile is 0. The mean tickets are 20 and the 95th percentile loss streak is 4. This replay is not executable and is for validation purposes only, using a vectorbt return and drawdown calculation with one contract per event and a $1000 initial investment. The returns are normalized and permit fractional contracts, but this is not available in live trading. The MC25USD constant all-in exposure scales the empirical costed returns, not the quoted 20 premium plus 5 cost. A block 3 circular bootstrap was used, and the known-calendar holdout is not an independent discovery. Note that near-zero floating returns may affect the loss streak figure, so it should not be used. https://runway.adityaperswal.workers.dev/data/option-bars ; research/study.py R91306