A frozen note from 1880 to 1882 describes a trading strategy. The strategy is to buy when a gap of -1.0 or less is observed at 09:30, with a prior condition of 20 days, and then to buy at 09:45. The strategy uses a stop loss of 1 ADR and a target of 1.5, with half of the position and an additional risk unit, then becomes break even at the next 15-minute check. The strategy exits at 15:45. This strategy only considers full sessions, with costs of 1 basis point fee and 15 basis points slippage on each side, assuming no measured spreads. The exact order cash flows were verified against vectorbt. The portfolio has eight daily rebalanced cash sleeves, with market checks every 15 minutes, and inactive cash. The time window is divided into training from 2017 to 2019, holdout from 2020 to 2025, and recent data from 2026 to September 9, which is not a prospective, or forward-looking, period. All neighboring data points are retained, with no selection. The results show a return of -27.667389%, a drawdown of 27.935345%, 791 trades, a win rate of 30.594185%, an average win of 0.759182%, and an average loss of 0.802501%. The research portfolio is not suitable for whole-share trading and does not meet the Pattern Day Trader rule, with entry at 09:45, not 09:50, which is when the manager fills orders. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not the original stock monthly portfolio.
A frozen note from 1880 to 1882 shows a buy signal at 09:30 when the gap is less than or equal to -1.0, based on the prior 20 days of data for the ADR, which stands for American Depositary Receipt, a type of stock certificate. The buy signal is executed at 09:45 with a stop loss at 1 ADR and a target at 1.5 times the ADR. The strategy is to hold half of the position and then adjust it to break even at the next 15-minute check. The position is exited at 15:45. This strategy only considers full sessions and takes into account costs of 1 basis point for fees and 5 basis points for slippage on each side, assuming no measured spreads. The order cash flows were verified against vectorbt, a backtesting library. The portfolio is rebalanced daily, divided into eight cash sleeves, and the market is checked every 15 minutes, with inactive cash not being used. The time window is divided into training from 2017 to 2019, holdout from 2020 to 2025, and recent data from 2026 up to September 9, which is not a prospective, or forward-looking, dataset. All neighboring data points are retained without selection. The results show a return of -11.877450%, a drawdown of 14.206376%, 791 trades, a win rate of 40.834387%, an average win of 0.744648%, and an average loss of 0.725488%. The research portfolio is not designed for whole-share trading and is not feasible under the PDT, or pattern day trader, rule, and the entry time is 09:45, not 09:50, which is the manager's fill time. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf, and this is an adaptation of a monthly stock portfolio, not the original.
A frozen note from 1880 to 1882 describes a trading strategy. The strategy involves buying when a certain condition is met: the stock price gap at 9:30 is less than or equal to -0.5, considering the prior 20-day average daily range. This condition is checked at 9:45. The strategy also includes a stop-loss at 1 ADR and a target at 1.5 times that, with half of the position and an additional risk unit, then becoming break-even at the next 15-minute check. The position is exited at 15:45. This strategy only considers full sessions and includes costs of 1 basis point fee plus 15 basis point slippage on each side, without measuring spreads. The cash flows from exact orders were verified against vectorbt. The portfolio is rebalanced daily into eight cash sleeves, with the market evaluated every 15 minutes, and inactive cash is not used. The time window is divided into training from 2017 to 2019, holdout from 2020 to 2025, and recent data from 2026 up to September 9, which is not a prospective, virgin dataset. All neighboring data points are retained without selection. The results show a return of -56.878654%, a drawdown of 56.928996%, 2151 trades, a win rate of 31.799163%, an average win of 0.614577%, and an average loss of 0.741765%. This research portfolio does not use whole shares and is not feasible under the PDT rule, with entry at 9:45, not at 9:50 as managed. The sources include stock bar data from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and a research paper from https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not an original monthly stock portfolio strategy.
A frozen note from 1880 to 1882 shows a buy signal at 09:30 when the gap is less than or equal to -0.5, based on the prior 20 days of data for the ADR stock. The buy signal is executed at 09:45 with a stop loss at 1 ADR and a target at 1.5 times the stop loss. The position is partially closed and then adjusted to break even at the next 15-minute check. The position is exited at 15:45. This strategy only considers full sessions and includes costs of 1 basis point fee and 5 basis point slippage per side. The order cash flows were verified against vectorbt. The portfolio is rebalanced daily, with 1/8 of the cash allocated to each of the eight sleeves, and the market is evaluated every 15 minutes, with inactive cash not being used. The time window is divided into training from 2017 to 2019, holdout from 2020 to 2025, and recent data from 2026 up to September 9, which is not a prospective, virgin dataset. All neighboring data points are retained without selection. The results show a return of -26.212876%, a drawdown of 27.351241%, 2151 trades, a win rate of 43.979544%, an average win of 0.617582%, and an average loss of 0.682769%. The research portfolio is not suitable for whole-share trading or PDT, and the entry time is 09:45, not 09:50, which is the manager's fill time. The data sources are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not an original stock monthly portfolio.
A frozen note from 1880 to 1882 describes a trading strategy. The strategy is to buy when a gap of -0.25 or less is observed at 9:30, with a prior 20-day average daily range, and then act at 9:45. The strategy uses a stop loss of 1 times the average daily range and a target of 1.5 times the average daily range. If the trade is profitable, half of the position is closed and the remaining position is adjusted to break even at the next 15-minute check. The trade is exited at 15:45. This strategy only considers full trading sessions and includes costs of 1 basis point fee and 15 basis point slippage per side. The strategy was tested using vectorbt and the results were verified. The portfolio is rebalanced daily, with 1/8 of the cash allocated to each of 8 sleeves, and the market is checked every 15 minutes, with inactive cash not being used. The time period used to test the strategy was divided into training from 2017 to 2019, holdout from 2020 to 2025, and recent from 2026 to September 9. The recent period is not a prospective test, meaning it is not a true out-of-sample test. All neighboring data points were retained, with no selection. The results show a return of -75.745751%, a drawdown of 75.764669%, 3507 trades, a win rate of 31.964642%, an average win of 0.559531%, and an average loss of 0.734617%. The research portfolio is not suitable for whole-share trading or pattern day trader rules, and the entry time is 9:45, not 9:50, which is when the manager fills orders. The data sources are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation of a monthly stock portfolio strategy, not the original.
A frozen note from 1880 to 1882 describes a trading strategy. The strategy involves buying when a specific condition is met: the gap at 9:30 is less than or equal to -0.25, considering the prior 20 days of data for the ADR, which stands for American Depositary Receipt, a type of financial instrument. The buy is observed at 9:45. The strategy includes a stop loss, which is a price level at which to sell to limit losses, set at 1 ADR, and a target, which is a price level at which to sell to take profit, set at 1.5 times the initial investment. Half of the position is closed, and then a breakeven stop is applied at the next 15-minute check. The position is exited at 15:45. This strategy is only applied during full trading sessions. The costs include a fee of 1 basis point, which is a unit of measurement for interest rates, and slippage of 5 basis points per side, which refers to the difference between the expected price and the actual price of a trade. These costs are assumed and do not account for measured spreads, which are the differences between the prices at which a security can be bought and sold. The portfolio is rebalanced daily, divided into eight cash sleeves, with market checks every 15 minutes, and inactive cash is not used. The data is divided into training periods from 2017 to 2019, a holdout period from 2020 to 2025, and recent data from 2026 up to September 9. The recent data is not prospective, meaning it is not entirely new and has been used in some way before. All neighboring data points are retained without selection. The results show a return of -41.762789%, a drawdown of 42.168079%, 3507 trades, a win rate of 44.026233%, an average win of 0.579649%, and an average loss of 0.672792%. The research portfolio is not feasible for whole-share trading or under the PDT, which stands for Pattern Day Trader, rules, and the entry time is 9:45, not 9:50, which is the manager's fill time. The sources of the data include a stock bars feed and a research paper on the Tug Of War strategy. This is an adaptation and not an original monthly stock portfolio strategy.
A trading note was frozen on note 1880 out of 1882. It was observed that a buy order was made at 09:30 when the gap was less than or equal to -1.0, based on the prior 20-day average daily range. The stop loss was set at 1 times the average daily range and the target was set at 1.5 times the average daily range. Half of the position was closed and the remaining position was moved to a break-even point at the next 15-minute check. The position was exited at 15:45. The trading strategy was only applied during full sessions. The costs included a fee of 1 basis point and slippage of 15 basis points per side. The spreads were not measured. The order cash flows were verified against vectorbt. The portfolio consisted of eight daily rebalanced cash sleeves, with the market marked every 15 minutes, and inactive cash. The training window was from 2017 to 2019, the holdout period was from 2020 to 2025, and the recent period was from 2026 to September 9. All neighbors were retained without selection. The return was -11.942533%, the drawdown was 12.061057%, the number of trades was 325, the win percentage was 28.307692%, the average win was 0.391564%, and the average loss was 0.590038%. The research portfolio is not feasible for whole-share trading and the 09:45 entry time is not the same as the 09:50 manager fill time. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation of a stock monthly portfolio, not the original.
A frozen note from 1880 to 1882 observed a buy signal at 09:30 when the gap was less than or equal to -1.0, based on the prior 20 days of ADR data, and executed the buy at 09:45. The strategy used a stop loss of 1 ADR and a target of 1.5, with half of the position and an additional 1 risk unit, then became break-even at the next 15-minute check, and exited at 15:45. This strategy only considered full sessions, with costs of 1 basis point fee and 5 basis points slippage per side, assuming no measured spreads. The order cash flows were verified against vectorbt. The portfolio consisted of eight daily rebalanced cash sleeves, with market checks every 15 minutes, and inactive cash. The training window was from 2017 to 2019, the holdout period was from 2020 to 2025, and the recent period was from 2026 to September 9, which was not a prospective, or new, dataset. All neighboring data points were retained, with no selection. The return was -4.500218%, the drawdown was 5.298802%, there were 325 trades, the win rate was 45.230769%, the average win was 0.407839%, and the average loss was 0.542390%. The research portfolio is not feasible for whole-share trading or PDT, and the entry time was 09:45, not 09:50, which is when the manager filled the order. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not an original stock monthly portfolio.
A frozen note from 1880 to 1882 was analyzed. It showed a buy signal at 09:30 when the gap was less than or equal to -0.5, based on the prior 20-day Average Daily Range, or ADR, which is a measure of price volatility. The buy was executed at 09:45. The strategy used a stop loss, which is a price level that triggers a sell order, at 1 ADR and a target price at 1.5 ADR. Half of the position was closed and the remaining position was adjusted to break even at the next 15-minute check. The position was exited at 15:45. This strategy was only applied during full trading sessions. The costs included a fee of 1 basis point, or 0.01%, and slippage of 15 basis points, or 0.15%, on each side of the trade. Basis points are units of measurement for interest rates and fees. Slippage refers to the difference between the expected price of a trade and the actual price. The order cash flows were verified against vectorbt, a backtesting framework. The portfolio was rebalanced daily, divided into eight cash sleeves, and the market was checked every 15 minutes. Inactive cash was not used. The training window was from 2017 to 2019, the holdout period was from 2020 to 2025, and the recent period was from 2026 to September 9. All neighboring data points were retained, and no selection was made. The results showed a return of -28.521873%, a drawdown of 28.585111%, 972 trades, a win rate of 28.497942%, an average win of 0.390971%, and an average loss of 0.541181%. The research portfolio is not suitable for whole-share trading or PDT, which refers to a pattern day trader, and the entry time was 09:45, not 09:50, which is the manager's fill time. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation of a stock monthly portfolio, not the original.
A frozen note from 1880 to 1882 observed a buy signal at 09:30 when the gap was less than or equal to -0.5, based on the prior 20 days of ADR data, and executed the buy at 09:45. The strategy used a stop loss of 1 ADR and a target of 1.5, with half of the position plus one risk unit, then became break-even at the next 15-minute check. The position was exited at 15:45. This strategy only considered full sessions and accounted for costs of 1 basis point fee plus 5 basis points of slippage per side, assuming no measured spreads. The exact order cash flows were verified against vectorbt. The portfolio consisted of eight daily rebalanced cash sleeves, with the market marked every 15 minutes, and inactive cash. The training window was from 2017 to 2019, with a holdout period from 2020 to 2025, and recent data from 2026 through September 9, which was not a prospective, virgin dataset. All neighboring data points were retained, with no selection. The results showed a return of -8.885486%, a drawdown of 9.479429%, 972 trades, a win rate of 44.958848%, an average win of 0.412245%, and an average loss of 0.474606%. The research portfolio is not suitable for whole-share or PDT feasible trading, and the 09:45 entry time is not the 09:50 manager fill time. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not an original stock monthly portfolio.
A frozen note from 1880 to 1882 observed a buy signal at 09:30 when the gap was less than or equal to -0.25, prior to 20 ADR at 09:45. The strategy used a stop loss of 1 ADR and a target of 1.5, with half of the position plus one risk unit, then a breakeven stop at the next 15-minute check. The position was exited at 15:45. This strategy only considered full sessions and accounted for costs of 1 basis point fee plus 15 basis point slippage per side, assuming no measured spreads. The exact order cash flows were verified against vectorbt. The portfolio consisted of eight daily rebalanced cash sleeves, with market checks every 15 minutes and inactive cash. The training window was from 2017 to 2019, with a holdout period from 2020 to 2025 and recent data from 2026 through September 9, which was not a prospective, virgin dataset. All neighbors were retained, with no selection. The results showed a return of -44.786626%, a drawdown of 44.835475%, 1645 trades, a win rate of 27.051672%, an average win of 0.412647%, and an average loss of 0.547584%. The research portfolio is not a whole-share or PDT feasible strategy, and the 09:45 entry was not the 09:50 manager fill. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not an original stock monthly portfolio.
A frozen note from 1880 to 1882 observed a buy signal at 09:30 when the gap was less than or equal to -0.25, with a prior 20-day average daily range. The buy was executed at 09:45 with a stop loss at 1 times the average daily range and a target at 1.5 times the average daily range. The strategy was to take half of the profit and move to break even at the next 15-minute check, and then exit at 15:45. This strategy only considered full sessions and assumed costs of 1 basis point fee plus 5 basis point slippage per side, without measuring spreads. The order cash flows were verified against vectorbt. The portfolio consisted of eight daily rebalanced cash sleeves, with the market marked every 15 minutes and inactive cash. The training window was from 2017 to 2019, the holdout period was from 2020 to 2025, and the recent period was from 2026 to September 9. All neighbors were retained without selection. The return was -16.731663%, the drawdown was 17.031097%, the number of trades was 1645, the win percentage was 43.890578%, the average win was 0.414462%, and the average loss was 0.481401%. The research portfolio is not suitable for whole-share or PDT feasible strategies, and the entry time was 09:45, not 09:50, which is the manager's fill time. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min and https://personal.lse.ac.uk/polk/research/TugOfWar.pdf. This is an adaptation, not an original stock monthly portfolio.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era crossing in December 2025. The last session of each month used a 15-minute bar close check, plus the next session's open, to determine entry and exit points. A stop loss of 1.5 times the initial risk, with a half sale at 1 times the initial risk, was used, and the position was exited at the fourth session's close. Fractions of shares were considered, with slippage of 15 basis points and a fee of 1 basis point per side. All 142 EEM and EWC windows had complete calendar-expected regular bars, but 2014 returned off-hours data that was discarded. The maximum daily return versus the 15-minute open was 194.477%. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was used. The results included a sampled path drawdown percentage of -11.036248442056706, a return percentage of 0.731303207559808, and a trade boundary drawdown percentage of -10.0586758557471. The win rate was 53.52112676056338%, with an average win percentage of 1.3333842639233562 and an average loss percentage of 1.48333946001032. The mean percentage was 0.024202814771084175, with a confidence interval of -0.3742495685473876 to 0.40832029280842774. The quarter Kelly was 0.0045378544328754655, and the exit counts were 27 for time, 26 for stop, and 18 for target. A partial count of 31 was also recorded, with a worst R of -1.712460024833221. The positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The data source was https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era-crossing in December 2025. The last session of each month used a 9:30 am bar open entry, with regular checks at the close of each 15-minute bar, plus the next session's open. A prior 20-day average daily range stop was used, with a target of 1.5 times the range, and a half sale or break-even immediately after the check, with a time exit at the close of the fourth session. Fractions were considered, with slippage of 5 basis points and a fee of 1 basis point per side. All 142 EEM and EWC windows completed as expected, with 2014 returning results that included off-hours data, which was discarded. The maximum daily versus 15-minute open was 194.477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was used. The results included a sampled path drawdown percentage of -10.538762066419006, a symbol of EWC, a k value of 1.25, a cost of 6, a return percentage of 16.100565713895598, and a trade boundary drawdown percentage of -7.732811097044944. The sample size was 71, with a win rate of 59.154929577464785, an average win percentage of 1.3981812535322533, and an average loss percentage of 1.4754326783287066. The mean percentage was 0.22445161939186103, with a confidence interval of -0.1625949845379823 to 0.608761952203776. The quarter Kelly value was 0.04013278300378163, with exit counts of 27 for time, 26 for stop, and 18 for target. The partial count was 31, and the worst R value was -1.5828537945454906. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era crossing in December 2025. The last session month was checked at 09:30 bar open entry, with regular completed 15-minute bar close checks plus the next session open. Prior to this, a 20 ADR stop was set, with a target of 1.5R and a half sale or break-even immediately after the check, and time exit at the fourth session close. Fractions, slippage of 15 basis points, and a fee of 1 basis point per side were considered. All 142 EEM and EWC windows completed the calendar with expected regular bars, but 20142 returned off-hours data that was discarded. The maximum daily versus 15-minute open was 194.477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was performed. The results included a sampled path drawdown percentage of -19.50635297128358, a symbol of EWC, a k value of 1, a cost of 16, a return percentage of -14.982837497701363, a trade boundary drawdown percentage of -18.30386074572081, a sample size of 71, a win rate of 45.07042253521127, an average win percentage of 1.142786732605646, an average loss percentage of 1.3335592961799254, a mean percentage of -0.21745967757234377, and a confidence interval percentage of -0.5674262185206318 to 0.13189854356204259. The quarter Kelly was 0. The exit counts were 36 for stop, 19 for target, and 16 for time, with a partial count of 31 and a worst R of -2.059675260339765. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min, which is a news scan found headline.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era-crossing in December 2025. The last session of each month used a 9:30 bar open entry, with regular checks at the 15-minute bar close, plus the next session open. Prior to this, a 20 ADR stop was used, with a target of 1.5R and a half sale or break-even immediately after the check, and a time exit at the fourth session close. Fractions were considered, with slippage of 5 basis points and a fee of 1 basis point per side. All 142 EEM and EWC windows completed as expected, with 20142 returned, including off-hours data that was discarded. The maximum daily versus 15-minute open was 194.477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20000 draws and a seed of 910061 was used. The results included a sampled path drawdown percentage of -12.8385%, a return percentage of -2.0112%, and a trade boundary drawdown percentage of -11.1380%. The sample size was 71, with a win rate of 52.1127%, an average win percentage of 1.1784%, and an average loss percentage of 1.3193%. The mean percentage was -0.0177%, with a confidence interval of -0.3723% to 0.3329%. The quarter Kelly was 0, with exit counts of 36 stops, 19 targets, and 16 time exits. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era-crossing in December 2025. The last session of each month used a 9:30 bar open entry, with regular checks at the close of each 15-minute bar, plus the next session's open. A prior 20 ADR stop was used, with a target of 1.5R and a half sale or break-even immediately after the check, and a time exit at the close of the fourth session. Fractions, slippage of 15 basis points, and a fee of 1 basis point per side were considered. All 142 EEM and EWC windows completed as expected, with 20142 returns including off-hours data that was discarded. The maximum daily versus 15-minute open was 194.477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was used. The results included a sampled path drawdown percentage of -10.2136%, a return percentage of 11.7495%, and a trade boundary drawdown percentage of -9.1479%. The sample size was 71, with a win rate of 57.7465%, an average win percentage of 1.4258%, and an average loss percentage of 1.5462%. The mean percentage was 0.1701%, with a confidence interval of -0.2055% to 0.5361%. The quarter Kelly was 0.0298%, with exit counts of 32 targets, 26 stops, and 13 time exits. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era-crossing in December 2025. The last session of each month used a 09:30 bar open entry, with regular checks at the close of each 15-minute bar, plus the next session's open. The prior 20 ADR stop was set at k = 1.25, with a target of 1.5R and a half sale or break-even immediately after the check, and time exit at the fourth session close. Fractions were considered, with slippage at 5 basis points and a fee of 1 basis point per side. All 142 EEM and EWC windows completed as expected, with 2014 returned and off-hours data discarded. The maximum daily versus 15-minute open was 19.4477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was used. The results included a sampled path drawdown percentage of -8.9477%, a return percentage of 28.7999%, and a trade boundary drawdown percentage of -7.8352%. The win rate was 60.5634%, with an average win percentage of 1.5602% and an average loss percentage of 1.4564%. The mean percentage was 0.3706%, with a 95% confidence interval of -0.0079% to 0.7342%. The quarter Kelly criterion was 0.0594%. Exit counts were 32 for target, 26 for stop, and 13 for time, with 29 partial counts. The worst R was -2.4920%. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era-crossing in December 2025. The last session month was checked at 09:30 bar open entry, with regular completed 15-minute bar close checks plus the next session open. Prior to this, a 20-day average daily range stop was set, with a target of 1.5 times the range and a half sale or break-even immediately after the check, with a time exit at the fourth session close. Fractions were considered, with slippage of 15 basis points and a fee of 1 basis point per side. All 142 EEM and EWC windows completed as expected, with 2014 returning to include off-hours data that was discarded. The maximum daily versus 15-minute open was 19.4477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was performed. The results included a sampled path drawdown percentage of -14.0087%, a return percentage of -6.2111%, and a trade boundary drawdown percentage of -12.7475%. The win rate was 46.4789%, with an average win percentage of 1.3112% and an average loss percentage of 1.2860%. The mean percentage was -0.0789%, with a confidence interval of -0.4008% to 0.2514%. The quarter Kelly was 0, with exit counts of 36 stops, 27 targets, and 8 time exits. A partial count of 25 was also recorded, with the worst R being -1.8870%. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min, which is a data feed from a stock bar source.
A post-hoc execution audit, not a selection holdout, was performed on 71 calendar windows from 2020 to 2025, with no era-crossing in December 2025. The last session month was checked at 09:30 bar open entry, with regular completed 15-minute bar close checks plus the next session open. Prior to this, a 20-day ADR stop, or stop-loss, was set at 1.5R, with a half sale or break-even immediately after the check, and time exit at the fourth session close. Fractions were considered, with slippage at 5 basis points and a fee of 1 basis point per side. All 142 EEM and EWC windows completed as expected, with regular bars, and 20142 returned, including off-hours data that was discarded. The maximum daily versus 15-minute open was 194.477%. Differences reflect both feed opens and exit rules. Overnight checks, unobserved quotes, and full shares were not validated. A 3-month circular bootstrap with 20,000 draws and a seed of 910061 was performed. The results included a sampled path drawdown percentage of -9.3034%, a return percentage of 8.0989%, and a trade boundary drawdown percentage of -8.9148%. The sample size was 71, with a win rate of 52.11%, an average win percentage of 1.3686%, and an average loss percentage of 1.2364%. The mean percentage was 0.1212%, with a confidence interval of -0.2099% to 0.4510%. The quarter Kelly was 0.0221%, and exit counts were 36 for stop, 27 for target, and 8 for time. A partial count of 25 was also recorded, with a worst R of -1.5688%. A positive point Kelly is not an approved allocation, and all mean confidence intervals include zero. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all&timeframe=15Min, which is a stock bars data feed.
A news scan found the notes from a trading fund manager. The notes were corrected on dates 1829, 1831, and 1833. The fund used a method called SIP, which was adjusted in 2016 and used for training from 2017 to 2019, and for testing from 2020 to 2025, with 2026 being separate. The fund excluded a type of exit called era-crossing exits. The mode of the fund was swing, which means it tried to make money from short-term price movements. The fund published its results based on the close of the second-to-last day and the close of the third month. It used a calendar that started on the last day of the month and ended on the third month. The fund controlled its trades based on sessions 10 to 13. It used a stop, which is a price level at which to sell, and a target, which is a price level at which to buy, with a ratio of 1.5, and it also used a method called average true range, or ATR, with a value of 20. The fund had fees of 1 basis point, which is a unit of measurement for interest rates, and slippage of 15 basis points per side. The fund also had a parameter called k, which was set to 1.25. The fund was divided into eight independent parts, called cells, each with an equal weight, and there was no rebalancing between them. The fund used a method called vectorbt, which is a backtesting framework, with a custom maximum error of 1.67e-15. The results were presented in columns, including symbol, return percentage, drawdown percentage, number of trades, win percentage, average win percentage, average loss percentage, mean percentage, quarter Kelly, which is a measure of risk, and buy and hold percentage. The results were as follows: for SPY, the return was 1.8053%, the drawdown was -1.8537%, and the win percentage was 50%; for IWM, the return was -3.7631%, the drawdown was -5.1196%, and the win percentage was 25%; for EFA, the return was -0.7458%, the drawdown was -4.0258%, and the win percentage was 50%; for EEM, the return was 2.0365%, the drawdown was -7.0994%, and the win percentage was 50%; for EWJ, the return was 1.3816%, the drawdown was -4.9892%, and the win percentage was 50%; for EWU, the return was -3.0076%, the drawdown was -3.5291%, and the win percentage was 37.5%; for EWC, the return was 0.6771%, the drawdown was -3.8221%, and the win percentage was 62.5%; and for EWA, the return was -5.6199%, the drawdown was -5.6199%, and the win percentage was 25%. The data was sourced from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all, which is a website that provides stock data.
A news scan found the notes from an automated trading fund manager. The notes were corrected on dates 1829, 1831, and 1833. The manager used a method called SIP, which is a way to adjust data, and adjusted the data from 2016. The manager trained the system using data from 2017 to 2019, held out data from 2020 to 2025, and used 2026 separately. The manager excluded a type of exit called era-crossing exits. The manager used a mode called swing, which is a type of trading strategy, and published the results based on the close of the penultimate day and the close of the third next month. The manager also used a calendar, which is a way to organize time, and set the control to sessions 10 through 13. The swing strategy used a stop, which is a way to limit losses, and a target, which is a goal, with a ratio of 1.5, and a daily stop. The manager also used a concept called ADR, which stands for Average Daily Range, and set it to 20. The manager paid fees of 1 basis point, which is a unit of measurement, and slippage of 15 basis points per side. The manager used a value called k, which is a parameter, and set it to 0.75. The manager had eight independent cash sleeves, which are like separate accounts, with initial equal weights, and no rebalancing between ETFs, which are a type of investment fund. The manager used a daily close drawdown, which is a measure of losses, and not an intraday drawdown. The manager used a custom max error of 1.67e-15, which is a way to measure accuracy. The results were presented in columns, including symbol, return percentage, drawdown percentage, number of trades, win percentage, average win percentage, average loss percentage, mean percentage, quarter Kelly, which is a way to measure performance, and buy hold percentage. The results were as follows: for SPY, the return was -1.3461%, the drawdown was -2.4049%, and the win percentage was 37.5%; for IWM, the return was -3.9379%, the drawdown was -3.9379%, and the win percentage was 37.5%; for EFA, the return was -3.4559%, the drawdown was -3.7114%, and the win percentage was 37.5%; for EEM, the return was -0.7298%, the drawdown was -5.4863%, and the win percentage was 50%; for EWJ, the return was -6.1575%, the drawdown was -6.1575%, and the win percentage was 37.5%; for EWU, the return was -5.3420%, the drawdown was -5.3420%, and the win percentage was 25%; for EWC, the return was -0.3430%, the drawdown was -2.1947%, and the win percentage was 50%; for EWA, the return was -4.8692%, the drawdown was -5.0194%, and the win percentage was 25%. The manager used current survivor ETFs, adjusted prices, and not raw dividend, whole-share, 3 entry cap or overnight 15 minute validated. The manager did not promote holdout neighbors. The source of the data was https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all.
A news scan found the notes from an automated trading fund manager. The notes were corrected on dates 1829, 1831, and 1833. The manager used a method called SIP, which was adjusted in 2016 and used for training from 2017 to 2019, and for holding out from 2020 to 2025, with 2026 being separate. The manager excluded something called era-crossing exits, which refers to a type of exit strategy. The manager was in swing mode, meaning they were buying and selling stocks in a short-term strategy. The manager published results based on the penultimate close, which is the second-to-last closing price, and the third next month's close. The manager used a calendar that started on the last day of the month and ended on the third next month's close. The manager controlled the number of sessions, which refers to trading periods, from 10 to 13. The swing strategy used a calendar with a prior ADR, which stands for Average Daily Range, of 20, and a stop and target of 1.5 times that, with half of the position and one risk unit, and a buy signal the following day. The manager also used a daily stop at the first gap open. The fees were 1 basis point, which is a unit of measurement for interest rates, plus slippage of 15 basis points per side. The manager used a factor called k, which was set to 1. The manager had eight independent cash sleeves, which refers to separate portions of the portfolio, with initial equal weights and no rebalancing between ETFs. The manager used a daily close drawdown, which refers to the peak-to-trough decline in value, and not an intraday drawdown. The manager used a custom max error of 1.67e-15, which is a very small number. The columns in the results table included the symbol, return percentage, drawdown percentage, number of trades, win percentage, average win percentage, average loss percentage, mean percentage, quarter Kelly, which is a measure of risk, and buy and hold percentage. The results were as follows: for SPY, the return was -0.6152%, the drawdown was -2.2474%, and the buy and hold return was 11.6466%; for IWM, the return was -3.2612%, the drawdown was -4.8767%, and the buy and hold return was 17.7175%; for EFA, the return was -3.4306%, the drawdown was -3.8686%, and the buy and hold return was 11.424%; for EEM, the return was -0.1559%, the drawdown was -6.6349%, and the buy and hold return was 23.0178%; for EWJ, the return was -1.0347%, the drawdown was -4.8483%, and the buy and hold return was 19.5062%; for EWU, the return was -5.3218%, the drawdown was -5.3218%, and the buy and hold return was 9.0699%; for EWC, the return was 1.1011%, the drawdown was -3.2147%, and the buy and hold return was 13.1442%; and for EWA, the return was -4.6003%, the drawdown was -4.691%, and the buy and hold return was 13.2637%. The manager used current survivor ETFs and adjusted prices, not raw dividend, whole-share, and a 3 entry cap or overnight 15-minute validated. The manager did not promote holdout neighbors. The source of the data was https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all.
A news scan found the notes from a trading fund manager. The notes were frozen and then corrected on certain dates. The manager used a method called SIP, which was adjusted in 2016 and used for training from 2017 to 2019. The manager then held out from 2020 to 2025 and used 2026 separately. The manager excluded something called era-crossing exits, which refers to a type of trading exit strategy. The manager used a mode called swing, which is a type of trading strategy. The manager published results based on the close of the second-to-last day and the close of the third month. The manager used a calendar that started on the last day of the month and ended on the close of the third month. The manager controlled the sessions from 10 to 13. The manager used a swing strategy with a prior ADR, which stands for Average Daily Range, of 20, a stop of 1.5 times the ADR, and a target of half plus one risk. The manager also used a daily stop and a gap open. The manager paid fees of 1 basis point plus slippage of 5 basis points per side. The manager used a factor called k, which was set to 1.25. The manager had eight independent cash sleeves with initial equal weights and no rebalancing between ETFs. The manager used a daily close and did not use intraday data. The manager used a custom max error of 1.67e-15. The manager reported results in columns including symbol, return percentage, drawdown percentage, number of trades, win percentage, average win percentage, average loss percentage, mean percentage, quarter Kelly, and buy hold percentage. The results were as follows: for SPY, the return was 3.447307799091992 percent, the drawdown was -1.55889069408911 percent, and the win percentage was 62.5 percent. For IWM, the return was -2.2109122250102864 percent, the drawdown was -4.392247858133946 percent, and the win percentage was 25 percent. For EFA, the return was 0.8550089577609832 percent, the drawdown was -3.641164162799515 percent, and the win percentage was 50 percent. For EEM, the return was 3.6822344516130645 percent, the drawdown was -6.727080526361629 percent, and the win percentage was 62.5 percent. For EWJ, the return was 3.016776554486178 percent, the drawdown was -4.6083474228513115 percent, and the win percentage was 50 percent. For EWU, the return was -1.4432075517368248 percent, the drawdown was -3.000549160867394 percent, and the win percentage was 37.5 percent. For EWC, the return was 2.3009105332047364 percent, the drawdown was -3.4365731366263197 percent, and the win percentage was 62.5 percent. For EWA, the return was -4.097706139518598 percent, the drawdown was -4.633721309280503 percent, and the win percentage was 25 percent. The manager used current and survivor ETFs with adjusted prices, not raw dividend prices, and whole-share prices with a 3-entry cap or overnight 15-minute validation. The manager did not promote holdout neighbors. The source of the data was https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all.
A news scan found the headline 'QTOM91006 2026 swing 6 0.75 eight fixed ETF cells'. The notes from an automated trading fund manager were frozen and then corrected. The manager used a method called SIP, which is a way to adjust data, and trained it from 2017 to 2019, then held out from 2020 to 2025, and used 2026 separately. The manager excluded a type of exit called 'era-crossing exits', which means stopping a trade when it moves from one market condition to another. The manager used a mode called'swing', which means buying and selling based on short-term price movements, and published results based on the close of the second-to-last day and the close of the third month. The manager also used a calendar, which is a schedule of trades, and controlled the trades based on sessions 10 to 13. The manager used a type of stop, which is a price level at which to stop a trade, called 'ADR 20 stop', and a target, which is a price level at which to sell, called '1.5k'. The manager also used a type of risk management called 'half plus 1R', which means reducing the size of a trade by half and then adding one unit of risk. The manager paid fees of 1 basis point, which is a unit of measurement for interest rates, and 5 basis points for slippage, which is the difference between the expected price and the actual price. The manager used a parameter called 'k', which is a measure of risk, and set it to 0.75. The manager divided the investment into eight equal parts, called 'cash sleeves', and did not rebalance them. The manager used a type of backtest, which is a way to test a trading strategy, called'vectorbt', and set the maximum error to 1.67e-15. The results were reported in columns, including the symbol of the ETF, the return, the drawdown, which is the maximum loss, and other metrics. The results were as follows: SPY had a return of 0.245%, IWM had a return of -2.39%, EFA had a return of -1.90%, EEM had a return of 0.87%, EWJ had a return of -4.64%, EWU had a return of -3.82%, EWC had a return of 1.26%, and EWA had a return of -3.33%. The data was sourced from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all.