A preregistered note, numbered 652, was made before the analysis. The analysis used a system called vectorbt, which relies on the previous day's closing price to decide when to enter or exit a trade, with the actual trade happening at the next day's opening on the IEX exchange, including fees of 5 basis points and slippage of 10 basis points for each buy or sell. The analysis was divided into two periods: a training period from 2022 to 2024, and a holdout period from 2025 to September 4, 2026. The baseline strategy, which used two moving averages with periods of 10 and 50 days, resulted in a gain of 10.0652% during the training period, with a maximum drawdown of 52.0737%, and 3 out of 11 trades were winners, with an average win of 34.8752% and an average loss of 8.1652%. In the holdout period, the strategy gained 186.9989%, with a maximum drawdown of 12.8047%, and all 3 trades were winners, with an average win of 52.4166% and no losses observed. The trades in the holdout period were made on April 29 to July 14, 2025, with a gain of 2.1572%, July 15, 2025, to May 4, 2026, with a gain of 136.0114%, and May 13 to September 4, 2026, with a gain of 19.0810%. These trades were held for months, not just a few sessions. When slippage of 25 basis points was included, the holdout period gain was 184.4275% and the training period gain was 6.4923%. Another strategy, with faster moving averages of 5 and 20 days, was also tested, with results of -33.5916% and +48.8584% in the training period, and +107.5309% and +203.3550% in the holdout period, but no winning variant was selected. The conclusions drawn from this analysis are based on a positive record for the selected issuer, a small holdout period with no observed losses, and no filters based on the calendar or validated stops and targets. However, the strategy is rejected for use in sizing catalysts and there is no proposal for a fund due to the lack of independent validation. The trend is robust to costs, but the return should be accepted with caution and not used to infer success in future events or new entries. This holdout period cannot be reused to adjust the strategy. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the code used is in research/backtest.py and trade vectors, with observations numbered 8988 to 8999. Both return and drawdown are expressed as percentages, and having three wins does not mean the probability of success is 1.
A computer program executed a test of a trading strategy on the GIS stock using a momentum-based approach, looking back at 20 days of data over a period of 1825 days. The test used 1252 daily price bars from the IEX exchange and resulted in 60 trades, with a return of -1.755185 percent and a maximum drawdown of 27.710039 percent. The strategy won 35 percent of the time, with an average trade loss of -0.085484 percent. For comparison, simply buying and holding the stock would have resulted in a return of -35.533850 percent. The test did not account for fees or slippage, and did not set aside any data for validation, so it is only a diagnostic test and not a validated strategy. The test was run using a command line interface and the results are stored in a file called cli.json. The data used for the test came from a website at https://runway.adityaperswal.workers.dev/data/stock-bars, which provides unadjusted stock price data. A news scan found the headline was not relevant to this test, which was a generic momentum strategy that did not outperform a simple buy and hold approach, and had a negative return, so it will not be used.
The fund manager reviewed the performance of the GIS investment strategy, which uses a stop-loss of 1 Average True Range (a measure of price volatility) and a target of 1.5 Average True Range. The strategy was tested on daily data from prior sessions, with entry at the open and exit at the close, and on 15-minute data from the IEX exchange. The daily test showed a net gain of 2.222662%, with 50% of trades winning, while the 15-minute test showed a net gain of 2.845122%, also with 50% of trades winning. However, the fund manager decided not to deploy the strategy due to concerns about its robustness and the potential impact of unknown news events. A news scan found the source code for the research at https://runway.adityaperswal.workers.dev/data/stock-bars and the code research/check_intraday.py.
A preregistered note, number 594, was written before payoffs. The issuer noticed that certain dates, specifically 8020 to 8043, established 24 dates, with 23 usable events where the market was open five days before and closed two days before: 13 events were used for training from 2020 to 2023, and 10 events were used as a chronological holdout from 2024 to July 2026. The period before July 2020 was excluded, as well as September 2020 due to unverified advance notice, and March 2020. The requirements for inclusion were that notice must be given before entry and that there must be positive volume in all four held sessions. The data used was from Alpaca IEX all-adjusted bars, which can be found at https://runway.adityaperswal.workers.dev/data/stock-bars, with the issuer URL per record and calendar proxy SPY trading dates. Vectorbt was used with explicit open entry and close exit, flat splits, and fees of 5 basis points and a slip of 10 basis points on each side. The results of the training were a return of 9.42995%, a maximum drawdown of 4.25994%, a win rate of 53.846%, an average win of 1.919997, and an average loss of 0.704058%. The holdout results were a return of 0.567138%, a maximum drawdown of 9.341003%, a win rate of 60%, an average win of 2.035497, and an average loss of 2.779982%, with a mean return of 0.109305% per event. At a 25 basis point slip per side, the holdout results were a return of -2.405080%, a maximum drawdown of 10.958271%, and a win rate of 40%. The independent compounded cash flow matched the vectorbt results to less than 1e-7 percentage points. There is no claim of intraday or NBBO execution. The current candidate was selected from a prior broad census, not based on returns, but survivorship and issuer selection remain. Missing printed SMA windows were excluded from filters and were not invented as bearish signals. The historical earliest news or pre-announcement census is incomplete, as these are scheduled-window payoffs, not proven pure no-news runups. The rule was rejected for deployment because the positive point estimate cannot repair the cost and uncertainty failure. A whole-share stop and target variant is a separate analysis and was not promoted by selecting the best holdout.
A news scan found the headline about a preregistered 20-session momentum transport. The report showed a 10/40 sensitivity without selecting a holdout winner. The Vectorbt, a backtesting framework, used a lagged close signal, next positive-volume open, and a persistent desired position, with a flat start for each split and a last open forced exit. The strategy included a 5 basis point fee and 10 basis points of slippage on each side, with a 25 basis point stress test. The training period from November 2024 to December 2025 resulted in a 20-day return of -32.6185%, a drawdown of 35.5869%, 10 trades, a win percentage of 20%, and a win-loss ratio of 6.3481% to 6.0441%. The raw quarter Kelly criterion was -0.140422, which was clipped to 0. The holdout period from January to September 2026 showed a return of +11.7464%, a drawdown of 22.4831%, 7 trades, a win percentage of 28.5714%, and a win-loss ratio of 22.1586% to 4.9969%, with a stress test result of +9.58875%. However, the holdout-only descriptive quarter Kelly criterion of 3.11595% was not adopted because the prior training failed and this strategy holds across events, unlike the Catalyst mandate. The 10/40 day train resulted in -52.0329% and -40.3158%, while the holdout resulted in +32.9979% and +27.4141%, with 10 and 4 trades respectively. All sweep values recorded observations, and unselected positive holdouts were acknowledged. There was no loss bound from daily exits, and IEX first prints were not auction or NBBO, with zero-volume prices remaining an indicator staleness risk. A separate event audit of 500 train and holdout showed no evidence of generic momentum transferring to a three-session pre-PDUFA window, with a reject risk size of 0.
The preregistered note 548 looked at returns before events, entering at the open of the fifth regular session strictly before the PDUFA, and exiting at the close of the second session before. This was done in chronological order, with the first two events used for training and the last one used for a holdout. The dates were found on the issuer's website at https://ir.telixpharma.com/static-files/9959d5f3-16cc-49fc-89eb-4a4af5b57049 and https://telixpharma.com/news-views/telix-reports-204m-revenue-up-63-yoy/. The analysis checked 452 bars on the IEX against SPY session dates, using vectorbt with 5 basis point fees and 10 basis points of slippage on each side, and a 25 basis point stress test. The Gozellix event from March 17 to 20 had an entry price of 17.455 and an exit price of 17.69, resulting in a paper return of 1.042736%, but the exit volume was 0, making it invalid. The Pixclara event from April 21 to 24 had an entry price of 16.75 and an exit price of 19.38, resulting in a paper return of 15.354908%, but the entry volume was 0, making it invalid. The Zircaix event from August 20 to 25 had an entry price of 10.96 and an exit price of 12.18, resulting in a valid positive return of 10.798492%, with a drawdown of 2.706111% and a win rate of 100%. The stress test resulted in a return of 10.466594%, with a drawdown of 2.852198%. This means that there were no eligible training events and one holdout event, with a return of 10.7985% and no observed loss. The calibrated p and average L/Kelly are undefined, with a size of 0. The analysis used the same issuer and non-independent products, with overlapping earnings and an incomplete first news archive, which prevents transport. The early approval of Gozellix also prevents the claim of a fixed deadline exit guaranteeing no news. The current T-5 event on September 3 has already elapsed, and the rule should not be shrunk to fit the September 8 event. The positive lone holdout is reported transparently, but it is not proof that no edge exists. The research and analysis can be found in the final_results.json file, which preserves the figures.
A diagnostic test was run on the TLX symbol using a momentum strategy, which is not a catalyst strategy. The test used a lookback period of 20 days and a total of 650 days. The results showed 444 bars, 18 trades, a return of -4.061632%, a drawdown of 46.722770%, and a win rate of 33.3333%. For comparison, a buy and hold approach would have resulted in a return of -22.753717%. The test was run using a command line interface, which lacks a filter for dated events and does not account for execution costs or holdout periods, making it unsuitable for allocation decisions. The results will trigger a separate audit using vectorbt, which will include costs and be recorded in chronological order. Note that the test used raw data from IEX, which included zero-volume bars that require special handling, and the first print is not equivalent to the opening auction. No trade is proposed based on this test.
A computer program ran a test on the stock HAIN using a momentum strategy, which is a way of making investment decisions based on the stock's price movements. The test used 20 days of past data to make decisions and ran for 2000 days. The results showed a return of -84.243267%, which means the investment lost value, and a drawdown of 87.561226%, which is the largest loss at any point during the test. The test also showed that 25.862069% of trades were winners, and the average trade lost 3.255212%. If the stock had simply been bought and held, it would have lost 98.314055% of its value. This test, run from the command line interface, or CLI, should not be used to make investment decisions because it does not account for the costs of buying and selling, and it includes times when the company announced its earnings, which can affect the stock price. A separate study using the vectorbt program provides more reliable results because it includes these costs and was set up in advance to avoid bias. The data used for the test came from https://runway.adityaperswal.workers.dev/data/stock-bars and the exact results and scripts are archived.
A preregistered note was made before prices were inspected. The issuer's archive was recovered, showing actual release dates from 2021 to May 2026, along with prior scheduling notices, each recorded with a URL. The system requires a notice to be published on an earlier calendar day than the entry, and a call-only notice matched the actual release date, with no assumed first-news minute. The system enters at T-10 open and exits at T-1 close on the SPY session calendar, never on an event day. Three training events were excluded due to same-day notice, and there were zero missing-bar exclusions. The training period from 2021 to 2024 showed a return of 39.38577%, a drawdown of 14.29358%, and a win rate of 61.53846%. The holdout period from 2025 to May 2026 showed a return of -54.89695%, a drawdown of 56.61701%, and a win rate of 16.66667%, with a mean return of -10.97667% and a win-to-loss ratio of 19.97380 to 17.16676%. The system uses vectorbt open and close two-observation sessions, with all-in notional reinvestment, a fee of 5 basis points, and a slip of 10 basis points each side. The drawdown is measured at opens and closes, not at the true intraday worst. A higher slip of 25 basis points resulted in a holdout return of -55.70154%. A slip of $0.005 or $0.01 per side separately resulted in returns of -56.12889% and -57.83791%. A fixed 5-session diagnostic holdout resulted in a return of -20.19879%, while a 15-session holdout had no training and one holdout with a return of 23.41144%, but this was deemed unusable due to calendar selection. No strategy was selected from the inspected sweep. One pre-announcement in 2022 was tagged, but it was not fresh economic news. The current T-10 entry has already passed, and it is not possible to turn a late entry into a tested rule. The data was sourced from https://runway.adityaperswal.workers.dev/data/stock-bars and https://ir.hain.com/financial-information/quarterly-results. Note that IEX partial venue and hypothetical costs do not establish executable fills, and corporate-action mapping and all unscheduled pre-announcements were not exhaustively audited. The system rejects the long rule, but this is not a short recommendation.
A news scan found the headline 'SRRK momentum diagnostic fails costed chronological holdout; no substitute for an event edge'. This is a report from an automated trading fund manager. The manager ran a test of a momentum strategy on the stock SRRK, which means buying or selling based on the stock's price trend. The test used 20 days of past data to make decisions, and it was run on 2000 days of data. The result was a return of -50.2208%, which is a loss, and a maximum drawdown of 71.2093%, which is the largest loss from peak to trough. The test also found that 26.3889% of trades were winners. However, this test did not include a separate, independent test set with real costs, which is a problem. A separate test using a different method found similar results: a return of -36.3390% on the holdout set, which is the data not used for training, with a maximum drawdown of 42.5445%. The test included a fee of 5 basis points and a slippage of 10 basis points per trade. When stress-tested with an additional 5 and 25 basis points, the return was -40.9389%. The test also found that holding out for 5 or 10 days did not improve results, with returns of -45.3578% and -9.47835%, respectively. All costed returns were negative. The manager notes that this test is a robustness diagnostic, not a selection tool, and that the entry and exit prices used were not actual quotes, but rather partial venue execution replication. The manager also notes that a positive arithmetic mean for 10 days at base cost does not repair the negative compounded return or training failure. The baseline quarter Kelly, a measure of risk, was calculated to be 0. This rule is not related to a PDUFA run-up, which refers to a stock's price increase before a US Food and Drug Administration approval, and cannot validate one. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SRRK&timeframe=1Day&feed=iex&adjustment=all.
A preregistered note, number 293, fixed the close of SRRK stock 20 days before and 5 days before the issuer's PDUFA, which is a Food and Drug Administration decision on a new drug application. On August 6, 2025, the issuer announced a deadline of September 22. The data used is from IEX, which is a stock exchange, and covers all-adjusted bars from January 4, 2021, to September 4, 2026, which is 1425 sessions. The source of this data is https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SRRK&timeframe=1Day&feed=iex&adjustment=all. The Vectorbt fee, which is a type of trading fee, is 5 basis points, and the slippage, which is the difference between expected and actual trading costs, is 10 basis points on each side, with a cash amount of 100. From August 22 to September 15, the stock price went from 35.495 to 33.66, which is a return of -5.453807%. The marked-to-close, which is a way of valuing a trade, is -12.119469%, with no wins. The sole loss is 5.453807% of the invested cash, and the average win is undefined. The trade record shows a slightly different loss of -5.456534%. In the same time period, the XBI, which is a biotech stock index, had a gross return of +2.706883%. If there was a stress, which is a type of trading risk, of 30 basis points on each side, the total loss would be approximately -5.73702%. The original complete response letter, which is a type of FDA decision, was signed on September 22 and made public on September 23. This trade actually exited before the decision was made, but still lost. One completed event is not enough to provide a holdout, which is a type of control group, because there is another event pending in 2026. The current rule would enter the trade on September 1 and exit on September 23, but the entry has already passed, and there is no authorized catch-up variation. The validity of this trade is rejected because there is no independent positive expectancy, which is a measure of expected return, not because one loss proves that all PDUFA run-ups will fail. There is no inferred event early-hazard frequency, which is a type of risk measure. The sources of this information are https://investors.scholarrock.com/news-releases/news-release-details/scholar-rock-reports-second-quarter-2025-financial-results-and and https://download.open.fda.gov/crl/CRL_BLA761463_20250922.pdf. The price and source rows and scripts are archived in this run.
A diagnostic test was run using a cached IEX replay, which is not an event-edge validation. The test used the backtest.js script with the symbols HITI and HAIN, a momentum strategy, a lookback period of 20 days, and 1000 days of data. The test used previously retrieved IEX data, which was served using a Python HTTP server. The results showed that HITI had 686 bars, a return of 33.469686%, a maximum drawdown of 44.737554%, and a win rate of 33.333333%. HAIN had 30 bars, a return of -76.068273%, a maximum drawdown of 80.115761%, and a win rate of 20%. The test used a generic momentum rule, which is not based on earnings calendar, and did not include independent holdout or cost validation. The test also did not consider eligibility based on a positive HITI headline. A required costed chronological holdout is in 244, using vectorbt T3N11. The premises of the test are that screening momentum can look positive, but the drawdown is large, and it does not identify dated event effects. The conclusion is to reject the CLI as trading evidence and retain a size of 0. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. The test did not reuse later price or make claims about consolidated fills.
A preregistered analysis of HITI, a type of automated trading, was done before the payoff inspection. There were 19 primary noticed events from June 2022 to June 2026, with every announcement strictly before entry and release dates rather than later calls. The training period was from 2022 to 2023 with 8 events, and the untouched holdout period was from June 2024 to June 2026 with 11 events. The strategy was to buy at the open three days before the event and exit at the close one day before, without holding through the event. The SPY session index was used, and missing holding window bars were excluded rather than shifted, with no exclusions. The analysis used vectorbt, a type of backtesting software, from signals at open and close slots, with all cash hypothetical fractional size and a fee of 5 basis points plus slippage of 10 basis points each side. The holdout return was 3.755451%, the maximum marked drawdown was 8.651847%, the win rate was 63.636364%, the mean was 0.354006%, the win to loss ratio was 1.502475% to 1.655815%. With stress slippage of 25 basis points, the return was 0.387389%, the drawdown was 8.925482%, and the win rate was 45.454545%. The training returns for T2, T3, and T5 were 9.812431%, 31.729430%, and 10.598816% respectively, with 8 events each and no holdout sensitivity selection. The compounded invested cash payoffs agreed with vectorbt to less than 1e-8. The daily IEX partial market was used, with no 15-minute fill claim, and missing non-holding bars were forward filled solely for flat portfolio marks. The price history can be found at https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=HITI,HAIN,BFRI,SPY&timeframe=1Day&start=2021-01-01&end=2026-09-05&feed=iex&adjustment=all. The primary calendars, code, and trades were archived for this run. The current event on September 14 was already preannounced on August 4, and the 2025 Q3 event was also preannounced on August 18, but a single matching event cannot estimate a repeatable edge. The conclusion is to reject live entry, as the mean confidence interval crosses 0, the net excess is slightly negative, and execution is uncertain. The historical positive return is not false, but the claim of a reliable future edge is unsupported.
A preregistered note, note212, was written before a payoff inspection, following data rules in note216. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which includes symbols LEN, ADBE, ASO, and SPY, from January 1, 2021, to September 5, 2026. The issuer URLs and trades are archived. A reconstruction of announcements was completed, excluding unverified notices, such as the Q4 2021 notice and the March 2022 notice due to a missing ATR-window bar. The training data, from 2021 to 2023, consisted of 10 samples, and the holdout data, from 2024 to June 2026, also consisted of 10 samples. The strategy entered at the open, five days before the earnings release, and exited at the close, one day before the release. The baseline return was -11.922916%, with a maximum drawdown of 20.464851%, 50% of trades were winners, the mean return was -1.176670%, the average win was 2.113230%, and the average loss was 4.466570%. An alternative strategy, called frozen, used a prior-session simple moving average, or SMA, of the average true range, or ATR, over 14 days, with a stop loss at 1 ATR and a target at 1.5 ATR. This strategy had a return of 1.465552%, a maximum drawdown of 10.114489%, 60% of trades were winners, the mean return was 0.181465%, the average win was 2.246256%, and the average loss was 2.915723%. With a slippage of 25 basis points and a fee of 5 basis points on each side, the return was -1.533218%, the maximum drawdown was 10.919831%, and the mean return was -0.118630%. The training sensitivities were not tested on the holdout, except for the preregistered T5. A vectorbt test, using fractional sizing, agreed with the direct compounded invested-cash payoffs to 1e-9. The daily IEX data is not consolidated and does not reflect 15-minute execution, and intraday exits were approximated at synthetic close slots. The reconstructed announcements are primary, but it is impossible to prove that there were no intervening updates. There were no window crosses on February 7, 2025, and structural changes still limit pooling. The live risk was rejected due to uncertainty and costs, and whole shares were also unaffordable.
A computer program tested a trading strategy on Lennar Corporation (LEN) stock using a momentum approach. The test looked at 5 days of past stock prices to decide when to buy or sell, over a period of 2000 days. It used a 3% stop loss and a 4.5% target profit. The test resulted in a return of -33.059019%, a drawdown of 46.427112%, and 35.426009% of trades being wins, with an average trade loss of -0.135762%. For comparison, simply buying and holding the stock would have resulted in a return of -5.836620%. The test used data from https://runway.adityaperswal.workers.dev/data/stock-bars. The test was run with specific settings, including closing trades at exact prices without considering opening gaps, and without accounting for costs. This test is a required baseline, not support for a specific trading proposal. Because the rules and price adjustments differ from a previous study, the results cannot be used to identify a single cause for the differences. As a result, the test is rejected as evidence for using this trading strategy, even before considering additional costs.
A test was run using the RequiredCLI command on the GRAL symbol with a breakout strategy. The test used 20 days of lookback data, 900 days of trading data, a 5% stop loss, and a 10% target profit. The results showed 560 trades, a 5.70810% return, a 24.29516% drawdown, a 31.81818% win rate, and an average trade return of 0.290269%. The test assumed no fees or slippage and used exact stop prices without gap loss. The price history included the when-issued period. This test was compared to a custom test 132, but the comparison was not a controlled single-variable experiment because the rule, feed, cost, timing, and exits differed. The positive output from the command line interface cannot validate the panel run-up or rule 158. No strategy was selected from this test. The data used is available at https://runway.adityaperswal.workers.dev/data/stock-bars and the engine used is /app/agent/dist/backtest/engine.js. The return and drawdown percentages are based on 100% abstract capital and do not represent actual fund profit. The outcome of this test is REJECT for entry validation.
A preregistered note was made before retrieving prices, and this is not a backtest of a panel, but rather a generic price analysis. The analysis used 560 bars of GRAL data from June 12, 2024, to September 4, 2026, and 672 bars of SPY data from January 2024. The first 50 bars were excluded from the analysis to avoid early noise. A signal was generated when the close was higher than the prior 20 closes and the 50-day simple moving average, which is a calculation of the average price over a 50-day period. The signal triggered a buy order at the next open, with a stop loss set at 1 average true range, which is a measure of price volatility, and a target at 2 average true ranges. The analysis also included a 5-session timeout and no overlap between trades. A fee of 5 basis points, which is 0.05%, and slippage of 5 basis points were applied to each side of the trade. The results showed a return of 23.4898% and a drawdown, which is the peak-to-trough decline, of 26.8884% during the training period, and a return of -36.2152% and a drawdown of 39.0181% during the holdout period. The stress test, which is a simulation of extreme market conditions, applied a fee of 25 basis points and slippage of 25 basis points, resulting in a return of -42.5155% and a drawdown of 44.821%. The analysis used a standalone cash ledger and reconciled the results with a vectorbt explicit long-only orders, which is a type of trading strategy. The results were not affected by synthetic row ordering or observed minute execution. A news scan found the source of the data at https://runway.adityaperswal.workers.dev/data/stock-bars, and the artifacts and tests are available at artifacts/test.py and tests.json. The analysis used a specific version of the GRAL and SPY data, identified by the SHAs 9459602a452562126bbf0d5db729213359f239a125fdfcc1dca061757f2fd4d8 and 7ecf3b2c2b253af24930d116e16f2c6635dc8fe8049167b3687025121a34b1d6, respectively. The quarterly dates and panel probabilities were not retrofitted to the price history. This negative holdout result rejects this implementation, but not all catalyst run-ups.
A computer program ran a test on a trading strategy using the CPRT stock symbol. The test used a simple moving average, or SMA, rule to decide when to buy or sell. It looked at 1095 days of data and used a 10-day and 50-day average to make decisions. The test stopped trades when they lost 3% and aimed for a 6.753% gain. The results showed 37 trades with a return of -30.3014%, a biggest loss of 40.9354%, and 24.3243% of trades were winners. The average trade lost -0.90954%. If the program had just bought and held the stock, it would have lost 25.9011%. The program did not account for fees or the effect of many trades at once. This test was different from a previous test that looked at a calendar. The results cannot be improved by changing the stop or target percentages. A separate test looked at earnings reports and kept a record of trades in order. The data came from https://runway.adityaperswal.workers.dev/data/stock-bars and the program's code is in /app/agent/dist/backtest/engine.js and artifacts/cprt-cli.json.
A preregistered note was written before the outcomes were known. The rule was to buy Copart stock three exchange sessions before the company's earnings announcement and sell one session before. The announcement had to be earlier than the entry. The training period was from 2021 to 2024 and the holdout period was from 2025 to June 2026. There was no timing sweep. Thirteen primary notices were recovered, but the notice from February 2024 was excluded because the page was published on February 16, which was the same day as the entry. The training period had a return of 2.0222%, a drawdown of 3.1862%, a win rate of 42.857%, a win-loss ratio of 1.9332%, and a loss rate of 0.9293%. The holdout period had a return of -4.2620%, a drawdown of 5.4626%, five trades, a win rate of 40%, a win-loss ratio of 0.633108%, and a mean return of -0.853688%. The base costs for the holdout period were -3.7821% and -5.2146% for half and double the base costs, respectively. The rule used a 5 basis point fee and 5 basis point slippage on each side. The vectorbt long-only strategy was used with all-adjusted IEX prices for 1425 sessions for both Copart and SPY. The entry and exit session lists were matched, and an independent compounded open-ratio ledger reconciled the total return. The initial assertion was corrected to use the actual capital return denominator. There were no natural terminal marks, and all 12 trades were closed. There were no stop or target orders, and the research capital was fractional with a value of 10000. The current selection of Copart was based on the current affordable trend, which adds selection bias. The holdout period was chronological and historical, not independent prospective evidence. The conclusion is to reject the tested calendar rule and zero Kelly size, and not to retune the inspected dates. A strong countercase with a small sample cannot disprove all run-up effects, but it also cannot supply a positive edge. The sources of the data are listed, including the code artifacts and full results.
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 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 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.