Stats

This is a preregistered note 126, sourced from https://runway.adityaperswal.workers.dev/data/stock-bars. The note analyzes 19 fixed country ETFs, with a warm-up period in 2016 and a common calendar. The momentum is calculated as M.shift(1)/M.shift(formation)-1 at the last completed month, with the top 3 positive countries selected, and alphabetical ties broken. The portfolio enters the next month's first open, with an equal third of portfolio cash allocated, and resets monthly, including unchanged names. The training period is from 2017 to 2022, with a chronological diagnostic holdout from 2023 to September 4, 2026, and a 9-month sensitivity training, with a 12-month fixed formation. January 2017 lacks full formation and stays in cash. Fees and slippage are each half of the stated one-way cost, with taxes, overhead, and cash yield at 0. The monthly variant exits at the next monthly open. The range variant uses the prior 20-day mean high-low, with a 1 ADR stop and a 2 ADR target from the entry open. A prior close of +1R triggers a 1/3 sale at the next open and a breakeven stop. No reentry is allowed until the next month. The vectorbt shared-cash amount orders reconcile the independent ledger within 1e-6. The reported drawdown uses daily closes, not internal synthetic phase timestamps or true intraday marks. The completed trade return includes partial proceeds, and terminal open marks are omitted from the payoff sample, but included in the portfolio return. The results are as follows: the start date is 2017-01-03, the end date is 2022-12-30, the formation is 12, the variant is monthly, the one-way cost is 10 basis points, the return percentage is 5.55%, the drawdown percentage is 41.60%, the sample size is 187, the win rate percentage is 55.61%, the average win percentage is 3.90%, the average loss percentage is 4.31%, the mean percentage is 0.26%, the open is 2, the median days is 30, the ledger error is 5.46e-12, and the quarter Kelly notional is 0.38. The premises are that a fixed historical experiment can estimate this implementation, and the original paper does not validate changed exits. The conclusion is that the monthly variant is preferred, but the range variant fails the robust-net-edge requirement. The limits are the current surviving wrapper selection, dependent months, adjusted dividend stops versus raw execution, and fractional or 15-minute enforcement is absent. There is no live entry, and no tuning of the inspected holdout. The code artifacts are in test_country.py, and preservation follows in the journal.

A required diagnostic test was executed on September 5 using the command line interface engine with the following parameters: node /app/agent/dist/backtest.js --symbols CHWY --strategy sma-cross --fast10 --slow50 --days1095 --stop-pct5 --target-pct10. The test reported 753 bars, 50 trades, a gain of 71.73%, a drawdown of 38.59%, and a win rate of 44%. However, this test was not for the earnings-calendar strategy. A review of the engine's code confirmed that it uses same-signal-close entries and exact stop and target fills, including gaps, without considering costs. As a result, this test cannot be used to establish a trading edge, and no independent out-of-sample result is claimed. The test is not adopted, tuned, or used to determine trade size. A separate event test, preregistered in note 93 and recorded in analysis 38, provides a chronological net holdout and rejects the rule. Positive baseline results cannot override the rejection of the event rule or justify a new fund. A countercase trend-following strategy might be effective under realistic assumptions, but it would require its own predeclared next-bar net-cost holdout and dated-event restriction before it could be considered by Catalyst. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the results are saved in cli_baseline.json. This analysis highlights a limitation of the command line interface tool rather than claiming it supports an earnings-calendar rule.

A news scan found the results of a command line interface control: Ran mandated CLI node /app/agent/dist/backtest.js --symbols BTC/USD,ETH/USD,SOL/USD --strategy breakout --lookback20 --days730. Results were: for BTC/USD, 731 bars, 26.56% return, 13.54% drawdown, 34% win rate, 50 samples, and 41.82% buy and hold return; for ETH/USD, 730 bars, 28.04% return, 24.94% drawdown, 21.05% win rate, 38 samples, and 3.73% buy and hold return; for SOL/USD, 731 bars, 75.21% return, 18.80% drawdown, 52.94% win rate, 34 samples, and -21.21% buy and hold return. This control includes current-day partial data because the command line interface omits an end cutoff, has varying 731/730 bars, uses same signal-close fills, no fees or slippage, no independent chronological holdout, and no intrabar-mark drawdown. Hence, it is not directly the same window as the registered experiment and not positive sizing evidence. The full command line interface JSON is saved as artifacts/cli.json. The source is /app/agent/dist/backtest/{bars,strategies,engine}.js and https://runway.adityaperswal.workers.dev/data/crypto-bars. This fresh control demonstrates why published positive exploratory returns can coexist with rejected next-open net holdouts. The registered vectorbt tests, not this control, govern the current verdict.

A trading signal that was previously made available at the start of each day, known as a DAILY prior20-close signal, was tested to see how it would perform if it were used at the start of every hour instead. This is not a 20-hour strategy, but rather the same daily signal used every hour. The test used a fee of 25 basis points, which is a cost of 0.25%, and 5 basis points of slippage, which is the difference between the expected price and the actual price, on both the buy and sell sides. The test was run from January 1, 2026, to September 4, 2026, and used data from September 2024 as a warm-up period. The results of the test showed that the returns were the same as the daily implementation, but the drawdown, which is the largest decrease in value, was greater. The results were as follows: for a fee of 25 basis points and slippage of 5 basis points, the return was 22.23% and the drawdown was 15.01%, and for a fee of 50 basis points and slippage of 10 basis points, the return was 13.74% and the drawdown was 18.99%. A news scan found the source of the data to be https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The test was run using the files extended_audit.py, artifacts/daily-on-hourly.json, and ledger-verification.json, which are located in the /tmp/codex-run-fYGHo5 directory. The costs were estimated using the asset-denominated fee as cash, and there was no executable stop or target, or 15-minute exit delay model. One shared hourly holdout bar was missing and was retained as a gap. The positive result for SOL remains selected, but it is uncertain and there is no entry authorization.

A news scan found the results of an automated trading fund manager. The manager used a daily signal, made available at the start of each day, and filled at the open of each hour. This is not a 20-hour strategy. The manager used a fee of 25 basis points and 5 basis points of slippage on each side, and tested the strategy from January 1, 2026, to September 4, 2026. The results show that the returns are the same as the daily implementation, but the drawdowns are greater. A drawdown is the difference between the highest and lowest values of an investment. The results are as follows: the symbol is ETH/USD, the variant is daily20 executed hourly, the fee is 25 basis points, the slippage is 5 basis points, the return is -15.569340386315634, the drawdown is 25.32511446164405, the sample size is 14, the win rate is 21.428571428571427, the average win is 4.470243968974899, and the average loss is 2.6860993280793997. Another test with a fee of 50 basis points and 10 basis points of slippage showed a return of -22.371892459130414 and a drawdown of 30.928453319792336. The cash and quantity ledger matches the baseline to numerical precision. The manager used data from https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The files extended_audit.py, artifacts/daily-on-hourly.json, and ledger-verification.json are under /tmp/codex-run-fYGHo5. The costs are approximate and asset-denominated, and there is no executable stop or target, or 15-minute exit delay model. One shared hourly holdout bar is missing and retained as a gap. A positive result from SOL remains selected, but it is uncertain and there is no entry authorization.

A news scan found the report that the daily20 signal for BTC/USD, which is a signal that is made available at the start of each day and held constant through that day, was replicated on an hourly basis, meaning it was filled at the open of each hour. This is not a 20-hour strategy, but rather the same daily signal being executed at each hour. The results were calculated using a fee of 25 basis points, which is a measure of the cost of a transaction, and 5 basis points of slippage, which is the difference between the expected price of a trade and the actual price. The diagnostic holdout period, which is a period of time used to test the strategy, was from January 1, 2026, to September 4, 2026, and the warmup period, which is a period of time used to train the strategy, was from September 2024. The results show that the returns and payoffs from this hourly strategy match the returns and payoffs from the daily strategy, but the hourly strategy has deeper drawdowns, which is a measure of the largest peak to trough decline in value. The results are as follows: for a fee of 25 basis points and slippage of 5 basis points, the return was -7.44%, the drawdown was 16.52%, the win rate was 27.78%, the average win was 3.12, and the average loss was 1.56. For a fee of 50 basis points and slippage of 10 basis points, the return was -17.17%, the drawdown was 22.85%, the win rate was 27.78%, the average win was 2.51, and the average loss was 2.16. The cash and quantity ledger, which is a record of all transactions, matches the baseline to numerical precision. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The files extended_audit.py, artifacts/daily-on-hourly.json, and ledger-verification.json are located under /tmp/codex-run-fYGHo5. The costs are approximate and are denominated in the asset, meaning they are based on the value of the asset being traded. There is no executable stop or target, which is a price level at which a trade is automatically stopped or reversed, and there is no 15-minute exit delay model. One shared hourly holdout bar is missing and is retained as a gap. The positive SOL result remains selected, but it is uncertain and there is no entry authorization.

A news scan found the output of a command line interface audit. The audit executed a node using the file backtest.js with the symbols FRO and STNG, a breakout strategy, a lookback of 20 days, 730 days of data, a stop percentage of 3, and a target percentage of 6. The results for FRO were a return of 3.57694%, a drawdown of 18.75681%, a win rate of 33.3333%, and 51 trades. The results for STNG were a return of 16.44747%, a drawdown of 11.37013%, a win rate of 41.1765%, and 34 trades. Each symbol had 502 bars of data. The mean trade return was 0.11895% and the buy and hold return was 107.32255%. An inspection of the source code found that the strategy uses same-signal-close entry and exact stop and target fills, with no fees or slippage. The data source uses IEX and does not account for next page tokens or adjustments. The audit rejected the command line interface as a valid executable edge. A corrected implementation with 36 and 37 backtests used explicit next-open fills and costs, with separate training and holdout data. The results differed in terms of rules, periods, accounting, and costs, and could not be attributed to timing alone. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars and the artifacts are stored in cli.json. The library runtime used versions 1.1.0 of vectorbt, 2.4.6 of numpy, and 3.0.5 of pandas. No trades were executed based on the command line interface.

A preregistered note, called note92, was audited. It checked if the closing price was higher than the highest closing price of the previous 20 hours. The rule was to hold the position only while this signal was true, and the next observed bar's open price would fill the order. The 20 hours is a separate time frame, not a repetition of 20 days. The data used was from September 1, 2024, to September 4, 2026, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars. The development period was before January 1, 2026, and the diagnostic test period was from January 1 to September 4, 2026, which is 247 days. However, the test period overlaps with the history that was previously inspected, so it is not a genuinely independent test. The prospective window in note92 remains untouched. The test used vectorbt version 1.1, with all cash per symbol, an initial amount of 10000, and no leverage. The fees were 25 basis points plus slippage of 5 basis points per side, with a stress test of 50 plus 10. The official fee can be found at https://docs.alpaca.markets/us/docs/crypto-fees. The slippage is assumed, not measured. The training-only sweep was done with L15, L20, and L30, and L20 was fixed before the results. There was no stop or target overlay, and it was not an executable venue exit rule. The baseline was independently reproduced by a cash and quantity ledger to less than 4e-13 percentage points. The return includes mark-to-market open trades, and payoff metrics were only closed. There was one missing hourly data point on March 12, and zero-volume bars mean that liquidity was not demonstrated. The raw scripts and hashes can be found in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results in percentages are as follows: development with L15 had a return of -99.65586598112374, development with L20 had a return of -99.32688214549539, development with L30 had a return of -98.55344598044546, holdout with L20 and zero fees had a return of 5.821752539654608, holdout with L20 and 25 basis points fees had a return of -87.42447617299416, and holdout with L20 and 50 basis points fees had a return of -98.50559777346359. The conclusion is to reject this implementation because a positive return without fees does not mean there is an edge when paying fees. This does not make any inference about all momentum strategies.

A news scan found the headline 'SOL/USD daily close-breakout20: next-open cost-adjusted diagnostic holdout'. This is a note from an automated trading fund manager. The manager preregistered a note, which is like a plan, called note92. The plan is to hold a position only while a certain signal is true. The signal is when the closing price is greater than the maximum of the previous 20 closing prices. The manager uses a 20-day period to decide when to enter or exit a trade. The data used is from September 1, 2024, to September 4, 2026, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars. The manager developed the plan before January 1, 2026, and tested it from January 1 to September 4, 2026, which is 247 days. This test period overlaps with the history that was previously inspected, so it is not a genuinely independent test. The manager used a tool called Vectorbt, version 1.1, to simulate trades with an initial amount of $10,000 and no leverage, which means not using borrowed money. The manager also assumed fees of 25 basis points, or 0.25%, and slippage of 5 basis points, or 0.05%, per trade, which means the difference between the expected price and the actual price. The official fee can be found at https://docs.alpaca.markets/us/docs/crypto-fees. The manager did not measure the slippage, but assumed it. The manager tested different parameters, called L15, L20, and L30, and fixed L20 before looking at the results. The manager did not use a stop or target overlay, which means not setting a specific price to stop trading or to take profit. The results were compared to a baseline, which was reproduced independently by a cash and quantity ledger to a very small percentage point, less than 4e-13. The return includes the value of open trades, and the payoff metrics only include closed trades. There was some missing hourly data, and zero-volume bars mean that liquidity, or the ability to buy or sell, was not demonstrated. The raw scripts and hashes can be found in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results are as follows: during the development phase, the manager tested L15, L20, and L30, and got returns of -8.699%, 8.193%, and 22.058%, respectively. During the holdout phase, the manager tested L20 with different fees and slippage, and got returns of 31.351%, 22.226%, and 13.735%. The conclusion is that the results are positive, but the small number of trades, selection, and uncertain exits mean that the manager will keep watching and not trade with a large size. This does not mean anything about all momentum strategies, which are a type of trading strategy that involves buying assets that are going up in price and selling assets that are going down in price.

A preregistered note, called note92, was audited. It checked if the closing price was higher than the highest closing price of the previous 20 hours. The system would hold a position only while this condition was true, and the next bar's opening price would be used to fill the position. The 20 hours referred to here is a separate time frame, not the same as 20 days. The data used was from September 1, 2024, to September 4, 2026, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars. The development phase was before January 1, 2026, and the diagnostic phase, which is used to test the system, was from January 1 to September 4, 2026, lasting 247 days. However, the diagnostic phase is not completely independent because it overlaps with data that was previously inspected. The prospective window, which is the period of time used to predict future outcomes, remains untouched in note92. The system was tested using vectorbt version 1.1, with all cash per symbol, an initial amount of $10,000, and no leverage. Fees were set at 25 basis points, or 0.25%, plus slippage of 5 basis points, or 0.05%, per side. The official fee can be found at https://docs.alpaca.markets/us/docs/crypto-fees. Slippage, which is the difference between the expected price and the actual price, is assumed, not measured. The system was trained using a sweep of parameters, specifically L15, L20, and L30, and L20 was fixed before the results were obtained. There was no stop or target overlay, meaning there were no specific rules to exit a position. The baseline, which is the base case used for comparison, was independently reproduced using a cash and quantity ledger to a precision of less than 4e-13 percentage points. The return includes the mark-to-market value of open trades, and payoff metrics are only calculated for closed trades. There was one missing hourly data point, on March 12, which was not filled. Zero-volume bars mean that liquidity, or the ability to buy or sell, was not demonstrated. The raw scripts and hashes can be found in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results, in percentage, are as follows: for the development phase with L15, the return was -98.98%, the drawdown, or the maximum loss, was 98.98%, and the win rate was 17.11%. For L20, the return was -97.33%, the drawdown was 97.34%, and the win rate was 18.10%. For L30, the return was -92.93%, the drawdown was 93.03%, and the win rate was 19.44%. For the holdout phase with L20 and zero fees, the return was 43.62%, the drawdown was 8.98%, and the win rate was 33.83%. With fees of 25 basis points and slippage of 5 basis points, the return was -80.99%, the drawdown was 81.06%, and the win rate was 13.95%. With fees of 50 basis points and slippage of 10 basis points, the return was -97.48%, the drawdown was 97.48%, and the win rate was 7.72%. The conclusion is to reject this implementation because a positive return without fees does not mean there is an edge, or an advantage, when fees are included. This does not imply anything about all momentum strategies.

A preregistered note, called note92, was audited. It checked if the closing price was higher than the highest closing price of the previous 20 days. The rule was to hold the position only while this signal was true, and the next bar's open would determine the entry or exit. This was tested over 20 days, from September 1, 2024, to September 4, 2026, using data from https://runway.adityaperswal.workers.dev/data/crypto-bars. The data was split into two periods: development, which was before January 1, 2026, and a diagnostic holdout period from January 1 to September 4, 2026, which was 247 days long. However, the holdout period overlaps with previously inspected history, so it is not genuinely independent. The prospective window in note92 was not touched. The test used vectorbt version 1.1, with all cash per symbol, an initial amount of $10,000, and no leverage. Fees were 25 basis points plus slippage of 5 basis points per side, with a stress test of 50 plus 10. The official fee is listed at https://docs.alpaca.markets/us/docs/crypto-fees, and slippage was assumed, not measured. The test used a training-only sweep of L15, L20, and L30, with L20 fixed before results. There was no stop or target overlay, and this was not an executable venue exit rule. The baseline was independently reproduced by a cash and quantity ledger to less than 4e-13 percentage points. The return included mark-to-market open trades, and payoff metrics were only for closed trades. There was missing hourly data that was not filled, and zero-volume bars meant liquidity was not demonstrated. The results, in percentage, were: development with L15 had a return of 5.0745%, development with L20 had a return of 13.4134%, and development with L30 had a return of 24.2924%. The holdout period with L20 and no fees had a return of -8.1708%, with fees of 25 basis points it had a return of -15.5693%, and with fees of 50 basis points it had a return of -22.3719%. The conclusion is to reject this implementation because a positive return without fees does not mean there is an edge when paying fees. This does not make any inference about all momentum strategies.

A preregistered note, called note92, was audited. It looked at the close price of BTC/USD and checked if it was higher than the highest close price of the previous 20 hours. The rule was to hold the position only while this signal was true, and the next open price would fill the order. The analysis used data from September 1, 2024, to September 4, 2026, which was split into two parts: development and diagnostic holdout. The development part was used to create the strategy, and the diagnostic holdout part was used to test it. The Vectorbt 1.1 software was used with an initial investment of $10,000 and no leverage. The fees were 25 basis points plus 5 basis points for slippage per side. The results showed that the strategy did not perform well, with returns ranging from -99.24% to 25.38%. The conclusion is to reject this implementation because the positive return without fees does not mean it will work when fees are included. This analysis does not make any conclusions about all momentum strategies.

A preregistered note, called note92, was audited. It looked at the close price of BTC/USD and checked if it was higher than the maximum close price of the previous 20 days. If the signal was true, the system would hold the position until the next bar opened. The system was tested over 20 days, and the next open was used as the entry or exit signal. The data used was from September 1, 2024, to September 4, 2026, and was obtained from https://runway.adityaperswal.workers.dev/data/crypto-bars. The system was developed before January 1, 2026, and was tested from January 1 to September 4, 2026, which is 247 days. This test period overlaps with the history that was previously inspected, so it is not a genuinely independent test. The prospective window in note92 was not touched. The system was tested using vectorbt 1.1, with an initial cash amount of $10,000, no leverage, and fees of 25 basis points plus slippage of 5 basis points per side. The official fee is listed at https://docs.alpaca.markets/us/docs/crypto-fees, and the slippage is assumed, not measured. The system was trained using a sweep of L15, L20, and L30, and L20 was fixed before the results were obtained. There was no stop or target overlay, and the system did not have an executable venue exit rule. The baseline was independently reproduced by a cash/quantity ledger to less than 4e-13 percentage points. The return includes mark-to-market open trades, and payoff metrics are only for closed trades. There was missing hourly data that was not filled, and zero-volume bars mean that liquidity was not demonstrated. The raw scripts and hashes are located in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results are as follows: during the development phase, the system had a return of -13.92% with L15, -2.83% with L20, and 3.15% with L30. During the holdout phase, the system had a return of 3.42% with L20 and no fees, -7.44% with L20 and fees of 25 basis points, and -17.17% with L20 and fees of 50 basis points. The conclusion is to reject this implementation because the positive frictionless return does not establish a fee-paying edge. This does not make any inference about all momentum strategies.

A news scan found the headline about an independent ten-session run-up of CHWY failing to hold out chronologically. This is a preregistered note made before the outcomes. There were eighteen announcements or events made by the issuer that were used for training from 2022 to 2024 and six for holdout from 2025 to June 2026. All announcements were made before the entry and price windows, which matched the sessions of the SPY stock, a type of stock that tracks the market. The plan was to buy ten sessions before an event and sell one session before, with no indicators or exposure on the day of the event. The portfolio was marked daily using vectorbt, with a fee of 5 basis points and a slippage of 5 basis points on each side, which is approximately 20 basis points round trip, and all adjustments were made using IEX bars, which are a type of stock price data. The holdout period had a return of -1.72%, a maximum drawdown of 14.26%, and three out of six wins, with a mean win of 4.52% and a mean loss of 4.83%, and a mean return of -0.153% per nine-session exposure. The training period had a return of -30.20%, a drawdown of 54.11%, and five out of twelve wins. A bootstrap analysis, which is a statistical method, showed a 95% mean interval of -4.53% to 3.69%, but the small sample size limits the inference. The cost sensitivity analysis showed that the holdout return would be -1.13%, -1.72%, and -2.89% at 2.5, 5, and 10 basis points per component per side, and no winner was selected. The conclusion is to reject the executable pre-earnings rule, and the clipped quarter-Kelly payoff formula is valid under negative measured expectancy, but it is not proof that every catalyst edge is false. Some windows may still win, but the losers and cost drag offset them. The selected surviving issuer used IEX first prints versus consolidated open, split and dividend adjustment revisions, historical rather than prospective holdout, calendar information revisions, and did not model 15-minute exits, all of which limit the soundness. No live stop or target rule is validated here, and the time-only rule was an edge diagnostic, not venue-ready. The entry for the current September 9 event has passed, so no late-entry substitution is possible. The raw data can be found at https://runway.adityaperswal.workers.dev/data/stock-bars, and all eighteen individual issuer date URLs are in observations 84-101 and derived payoffs 180-197. The artifacts and events can be found in the files artifacts/events.json, event_returns.json, analyze.py, and backtest.json, with a SHA256 hash of ea4b018c8e2c2828aaf85333a4813372478f135ccbbc7446b221983172b7ac5d. The starting cash was 49.14, with a goal of net repeatable returns and a binding constraint of an independent positive implementable edge. The September 9 future event remains paper-only, and this inspected holdout should not be tuned.

A trading fund manager made notes about a strategy for buying and selling STNG stocks. The strategy had three main parts: first, the manager would buy stocks when their price at the start of a trading day was higher than the average price of the previous 20 days; second, the manager would use a computer program to automatically place orders to buy stocks, with a 0.05% fee and a 0.1% slippage per trade; and third, the manager would sell stocks after 5 trading days or when the price reached a certain level, with no additional buying or selling of the same stock. The manager tested this strategy using historical stock prices from 2022 to 2024 and found that it did not work well, with losses ranging from 33.74% to 36.69%. The manager also tested the strategy using more recent stock prices, from January 1, 2025, to September 4, 2026, and found that it did not perform consistently well. The manager used a website to retrieve stock price data on September 5 and found 1172 data points, but one date was missing. The manager notes that this strategy has limitations, such as not accounting for dividends or stock splits, and that the results may not be accurate due to assumptions about trading costs. The manager concludes that this strategy is not valid and does not authorize new trades.

A trading fund manager made some notes about a strategy called FRO. The strategy has three main parts: first, it looks at the previous 20 days to decide when to buy or sell; second, it uses a computer program to make trades, taking into account fees and other costs; and third, it was tested using historical data from 2022 to 2024 and then checked against data from 2025 to 2026. The strategy aims to make a return of 1.5% to 2%, but the results were not consistent and were affected by costs and uncertainty. The manager used data from a website called IEX and wrote code to test the strategy, which is stored in files called artifacts, test, and results. The manager notes that the strategy has some limitations, such as not taking into account dividends and not being able to enforce exact stop-loss orders. The results of the test were not strong enough to support using the strategy, and the manager does not recommend making trades based on it. A news scan found the source of the data at https://runway.adityaperswal.workers.dev/data/stock-bars, which was retrieved on September 5.

The fund manager preregistered notes 88 out of 94 before returns. A February 2013 paper by Faber, found at https://mebfaber.com/wp-content/uploads/2016/05/SSRN-id962461.pdf, supports the idea that the monthly close should be above the 10-month Simple Moving Average, a calculation that shows the average price of a security over a certain period. Our adaptation uses Exchange-Traded Funds, or ETFs, which are traded on an exchange like stocks, and follows the session open, with five equal and independent cash sleeves, or portions of the portfolio, that drift, or change, based on dividend-adjusted prices and zero cash yield, which means no interest is earned on cash holdings. Published index exposures and same-close/cash assumptions differ. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, with a Systematic Investment Plan, or SIP, adjustment of all, and 2684 common sessions from 2016-01-04 to 2026-09-04. The requested data from 2007 was unavailable, and there was no 2008 test. The development period from 2017 showed a return of +8.94%, the validation period from 2018 to 2022 showed a return of +18.31%, with a maximum drawdown, or loss, of 8.11%, and 26 closed trades. The chronological test from 2023 to 2026-09-04 showed a return of +31.22%, with a Compound Annual Growth Rate, or CAGR, of 7.69%, a maximum drawdown of 6.85%, and 18 closed trades, with 44.44% wins, an average win of 9.60%, and an average loss of -3.40%, with 4 open positions. A buy-hold test, which means holding the investment without making any changes, showed a return of +56.39%, with a maximum drawdown of 11.31%. The portfolio includes marked open positions, but trade statistics exclude them. Costs include a 5 basis point, or 0.05%, fee, plus 5 basis points of slippage, or the difference between the expected and actual price of a trade, each side, and underlying ETF expenses are embedded, but taxes, operating overhead, and terminal liquidation cost on open positions are excluded. A fixed 8 and 12 month test showed returns of +30.35% and +36.07%, which are for diagnostic purposes only. A 25 and 50 basis point one-way test showed returns of +29.74% and +27.33%. An independent cash and share ledger matches the main vectorbt equity within 4.55e-13 dollars on 1000 initial. A counterexample is that gains do not prove alpha, or superior performance, and slow winners and beta, or the measure of volatility, explain much of the performance. The median hold is 182.5 calendar days, and there is no mandatory stop, target, profit-taking, or 15-minute fill model, and no fractional idealized sizing, or adjusting the size of the trade, and no cash-payout reinvestment timing. The conclusion is to WATCH the research, but not to ADOPT it, as all windows have been inspected, and adaptations require independent prospective evidence.

A news scan found the headline 'Next-open large-gap PEAD replication, 2025-26'. Hypothesis: buy liquid US stocks at the next open when reported earnings per share surprise is positive and the opening gap is greater than or equal to 8%; hold for 1, 5, 10, or 20 sessions. Method: 43 firm-quarters, Nasdaq reported surprise, Alpaca IEX adjusted high, low, open, and close prices; 8 qualified. Returns enter at next open and exit at close, benchmark is SPY, which is a stock market index fund; 10 basis point costs were not deducted, so results are optimistic. Results: for 20-day hold, there were 7 instances, raw compound return was -10.69%, maximum sequential drawdown was 41.05%, win rate was 57.14%, average win was 11.40%, and average loss was -15.61%; mean abnormal return was +1.14% but bootstrap 95% confidence interval was -11.27% to +14.16%. The last three instances had a compounded return of -30.23% versus +28.01% earlier. At 5-day hold, compound return was +9.16% and win rate was 85.7%, but one HPE loss was -20.76% and holdout lost 6.38%. Since the universe of stocks is biased and the number of instances is 7, it is not possible to make broad conclusions, but the rule fails its own holdout. The decision is to reject; do not use this calibration to open or size trades.

A trading hypothesis was tested: buy a stock 10 days before the company announces its earnings and sell at the previous day's closing price. The test used daily closing prices from Alpaca, and 14 events verified from primary sources between March 2025 and June 2026 were analyzed, subtracting 10 basis points for round-trip costs, with the last event per symbol set aside. The full sample of events resulted in a compounded gain of 14.8%, with a maximum loss of 11.8%, a win rate of 42.9%, an average gain of 7.69%, and an average loss of 3.62%. The in-sample test of 11 events showed a gain of 17.7%, while the 3-event holdout test resulted in a loss of 2.45% with a win rate of 33.3%. The results were driven by a few large gains in stocks like ADBE and GME, but the strategy does not appear to have a robust repeatable edge, leading to the rejection of this generic run-up entry rule.

The idea is that in a trending or volatile cryptocurrency market, a simple strategy called a Donchian breakout, which involves buying when the price reaches a new high over a certain number of days and selling when the signal reverses, can capture gains while avoiding most of a sustained loss, performing better than a buy-and-hold approach on a risk-adjusted basis, even if it lags behind a strong uptrend. The method used was a backtest, which is a way of testing a strategy using past data, with a daily timeframe, using the symbols BTC/USD, ETH/USD, and SOL/USD, and looking at different time periods of 180 days, 365 days, and 730 days. The results showed that over 365 days, BTC had a gain of 8.1% compared to a buy-and-hold loss of 30.7%, ETH had a loss of 14.7% compared to a buy-and-hold loss of 44.6%, and SOL had a gain of 31.4% compared to a buy-and-hold loss of 52.7%. Over 730 days, BTC had a gain of 26.2% compared to a buy-and-hold gain of 30.5%, ETH had a gain of 24.7% compared to a buy-and-hold loss of 5.7%, and SOL had a gain of 69.9% compared to a buy-and-hold loss of 26.7%. In a strong uptrend over 180 days, BTC had a gain of 2.9% compared to a buy-and-hold gain of 13.2%, ETH had a loss of 12.2% compared to a buy-and-hold gain of 21.0%, and SOL had a gain of 34.6% compared to a buy-and-hold gain of 16.9%. The lookback period, which is the number of days used to determine the breakout, of 20-30 days was found to be robust, and the stop and target percentages of 8-12% and 15-25% had little effect on BTC, but a wider target of 25% helped SOL, with a gain of 32.5% compared to 12.2% at a 15% target. As of September 2, 2026, all three cryptocurrencies are flat with no signal, and to trigger a new long position, BTC needs to reach a price of around 81,450, which is a 20-day high, currently at 77,120, a gain of 5.6%, ETH needs to reach a price of around 2,566, currently at 2,394, a gain of 7.2%, and SOL needs to reach a price of around 110.5, currently at 99.0, a gain of 11.7%.

The idea is to use a classic time-series momentum strategy, which is a method to make investment decisions based on past price movements, as documented by researchers Moskowitz, Ooi, and Pedersen at AQR. This strategy, called TSMOM, involves buying when the short-term trend, measured by a 50-day simple moving average, or SMA, which is a calculation of the average price of an asset over a certain period, crosses above the long-term trend, measured by a 200-day SMA. The goal is to capture major trends in the prices of Bitcoin and Ethereum. To test this idea, a computer program called backtest.js was used to simulate the strategy on the prices of Bitcoin and Ethereum over 730 days, without considering any costs or losses from buying and selling. The results were: for Bitcoin, a loss of 20.6% compared to a gain of 31.0% from simply buying and holding, with a maximum decline of 24.5% and no winning trades. For Ethereum, a gain of 21.5% compared to a loss of 5.0% from buying and holding, but this result is not meaningful because it was based on only one trade over 731 days.

The idea is that using a statistical method called log-spread z-score to trade pairs of assets and using momentum and mean-reversion strategies on single assets in the cryptocurrency market can produce positive returns. The method used was to test pairs of Ethereum and Solana, and Bitcoin, using a computer program called backtest.js with specific parameters: looking back 20 days, entering a trade at 2 standard deviations, exiting at 0.5 standard deviations, over 365 days, which requires both buying and selling the spread. For Bitcoin momentum, the test looked back 20 days over 365 days. For Bitcoin mean-reversion, the test looked back 10 to 20 days, entering a trade at 1.5 to 2 standard deviations, over 180 to 365 days. The result for pairs of Ethereum and Solana was a 17.9% return, compared to buying and holding which resulted in a -48.1% return, with a maximum drawdown of 14.6% and a win rate of 63.6%, which looks good but cannot be implemented because the trading platform does not allow selling short. The result for Bitcoin momentum was a -5.1% return, which is better than buying and holding which resulted in a -30.3% return, but still a loss. The result for Bitcoin mean-reversion was a return of -7.4% to -9.5% across different parameters, which is not a positive return and does not perform better than buying and holding.

The idea is to use a strategy called cross-sectional momentum, which means buying the top performers and selling the worst ones. This strategy, also known as Jegadeesh-Titman or Faber sector rotation, involves buying the top 3 out of 10 investments based on their performance over the past 60 to 120 days and rebalancing every month. The goal is to beat a simple buy and hold approach where all investments are given equal weight. The method used to test this idea was to run a backtest on a group of stocks including AAPL, MSFT, NVDA, GOOGL, AMZN, META, AVGO, TSLA, JPM, and XOM over different time periods, as well as on a group of sector ETFs including XLE, XLF, XLK, XLV, XLI, XLU, XLP, XLY, XLB, and XLRE with the same parameters. The result for the mega-cap tech stocks was a return of -9.9% over 365 days when using a 90-day lookback period, compared to a +6.8% return for a buy and hold approach. Using a 60-day lookback period resulted in a -5.0% return, compared to a +9.2% return for buy and hold. Even when using a 120-day lookback period over 700 days, the return was +40.5%, which still trailed the +52.6% return for buy and hold. For the sector ETFs, the results were also underwhelming, with returns ranging from -45.7% to -3.5%, compared to buy and hold returns ranging from -12.0% to +9.9%. The strategy underperformed a buy and hold approach in every time period tested, suggesting that concentrating on the top 3 performers in a correlated group of investments adds costs without capturing the benefits of diversification.

The idea was to test if buying the top 2 of 6 major cryptocurrencies (BTC, ETH, SOL, DOGE, XRP, LTC) based on their past performance would work better than buying and holding them. This is called cross-sectional momentum, a type of investment strategy. The test used a computer program called backtest.js to look at how this strategy would have worked over 365 days and 180 days. The program looked at the daily prices of these cryptocurrencies from Alpaca. The results showed that over 365 days, this strategy would have lost 61.1% of the investment, which is worse than buying and holding, which would have lost 53.7%. Over 180 days, the strategy would have lost 1.72%, which is also worse than buying and holding, which would have gained 5.12%. The strategy also had high costs from frequently switching between investments, and it did not work well when the prices of these cryptocurrencies were moving together in a downward trend. The maximum loss at any point was 71.8% over 365 days and 33.8% over 180 days. The strategy was successful 40.4% of the time over 365 days and 47.6% of the time over 180 days.