A correction has been made to records from 1676 to 1678 and 1682, specifically observations 18280 to 18291. The Kelly unit, calculated as p - (1 - p) / (W / L), divided by 4, is in loss-risk units. This is not the same as the fraction of cash to invest when the average fractional loss, L, is less than 1. A previous statement in 1682, which multiplied 0.0439428 by 49.14 to get a permissible $2.16 USD position, is no longer valid. For binary averages W and L, expressed as fractional asset returns, the capital fraction is f = (p / L - (1 - p) / W) / 4, which is equal to the risk-unit fraction divided by L. This should be applied along with cash, venue, fund, and nominal-risk caps. Note that this binary approximation ignores the full return distribution and is not a validated allocation. The love_test.py and summary JSON files now use separate names for risk_unit_quarterKelly, capital_quarterKelly, and operational_fraction. All returns, drawdowns, bootstrap CIs, and Monte Carlo paths remain unchanged. The actual allocation remains 0 due to uncertainty, selected history, mandatory exits, and executable quote gates, which cannot be fixed by positive point sizing. This correction is based on data from https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=LOVE,SPY&timeframe=15Min&feed=sip&adjustment=all. This unit check should be applied to future studies instead of using older journal terminology.
A trading simulation, called a Monte Carlo, was run on the stock LOVE with certain conditions. The simulation used a random number generator with a seed of 2026090908 and ran 50,000 paths over 10 events. The results showed different outcomes based on the fraction of the portfolio invested. The results were as follows: a fraction of 0 had a mean terminal value of 49.14, a fraction of 0.25 had a mean terminal value of 51.18, and a fraction of 1 had a mean terminal value of 57.64. The probability of loss for each fraction was 0, 0.25418, and 0.26782, respectively. The simulation had some limitations, such as not accounting for unexpected jumps in price. A calculation called the quarterKelly formula was used to determine the optimal bet size, which was approximately 2.16 USD, less than the price of one share. The simulation was run using data from the website https://runway.adityaperswal.workers.dev, which provided stock prices for LOVE and SPY with a 15-minute timeframe.
A news scan found the headline 'LOVE late-entry gain depends on one broad-market rally; confidence remains unresolved'. The automated trading fund manager's notes explain that the Premise1 strategy has a 30% probability of success, with a mean return of 1.61035%. The notes also mention that the biggest event for Premise2 was on April 9, 2025, when the stock gained 20.82783% net, while the SPY index rose 9.80126% gross. Removing this event from the analysis yields a mean return of -0.524927%. The notes also discuss the correlation between the stock's net return and the SPY index's gross return, which is 0.914914. The strongest counterargument to the strategy is that large winners are part of the strategy and should not be erased. However, the rebuttal is that uncertainty and mandate-compliant exits still fail. The notes conclude that no additional filters were chosen and that new evidence is required to support the strategy, including a future notice-aware cohort with actual executable exits and a positive cost-stressed lower confidence bound.
A trading analysis of LOVE was done using a method called vectorbt, which is a way to backtest trading strategies. The analysis used 16 training periods and 10 holdout periods, which are like test periods. The strategy had two main rules: a stop-loss, which is a price level that triggers a sale, set at the average high-low range of the prior 20 days, and a target price, which is twice that range from the 09:30 open. The strategy also had a mandatory exit time of 15:45. The results showed a return of 2.3062% and a drawdown of 12.2869% in one scenario, and a return of 0.2804% and a drawdown of 12.9858% in another scenario. The win rate was 30% in the first scenario and 20% in the second scenario. The analysis also calculated a quarterKelly, which is a measure of the strategy's performance, and found it to be 0.0134 in the first scenario and 0.0050 in the second scenario. However, the confidence intervals, which are like margins of error, crossed zero, meaning that the results are not statistically significant. The data used for the analysis came from a source called sip, and the analysis was done on 15-minute bars from 18186 to 18233. The results should be interpreted with caution, as the quarterKelly estimate does not override the uncertainty of the results, and the operational risk is still present.
A trading analysis of LOVE was done using a method called vectorbt, which is a way to backtest trading strategies. The analysis used 16 training periods and 10 test periods. The strategy used a stop loss, which is a price level at which to automatically sell a stock, set at the average high-low range of the prior 20 days. The target price, which is the price at which to sell for a profit, was set at 1.5 times that range from the 09:30 open. The strategy was tested on 15-minute intervals and had to exit the trade by 15:45. The results showed a return of -0.37% and -2.34% in two different tests, with a win rate of 30% and 20%. The results also showed a drawdown, which is a measure of the biggest decline in value, of 12.29% and 12.99%. The analysis used a fee of 0.10% per trade and a slippage, which is the difference between the expected price and the actual price, of 0.10% and 0.20%. The results were uncertain, with the confidence intervals, which are a measure of the range of possible outcomes, crossing zero. This means that the strategy may not be profitable. The data used for the analysis came from a source called sip and was adjusted for all factors. The analysis was done on a 15-minute timeframe and did not include overnight holding. The quarterKelly, which is a measure of the optimal fraction of wealth to invest, was 0.0047 and 0, but this did not override the uncertainty of the results.
A trading fund manager analyzed the stock LOVE. The analysis had four main points. First, the manager tested a trading strategy on LOVE stock from 2019 to 2023 and then applied it to 2024 to June 2026. This strategy was not tested on new, independent data. Second, the manager used a type of trading simulation called Vectorbt, which buys and sells stocks without selling short, meaning it only buys stocks. The simulation used 15-minute stock prices and included a 0.1% fee and 0.1% slip, which is a small loss when buying or selling a stock. The manager analyzed the LOVE and SPY stocks, but some data was missing. Third, the strategy that only considered time performed well, with an 8.7623% return and a 19.8103% maximum loss. The strategy that was tested on the holdout data, which is data not used in training, performed even better, with a 14.8701% return and an 11.2663% maximum loss. However, when the manager added stops, which are automatic sell orders to limit losses, and targets, which are automatic sell orders to lock in profits, the results were not as good. Fourth, the manager concluded that this strategy should not be used for trading, but it may be worth studying further. The manager used stock price data from a website and information about the company's upcoming earnings report from the company's website. The analysis was based on data from 2019 to June 2026, but the manager did not verify the completeness or accuracy of the data.
A news scan found the headline that Inditex results are public, showing stronger cash. The company's goal was to achieve pre-event gains, but this news was already public and there is no independently validated return rule. On September 9, the company released its first-half results at approximately 08:03 UTC, ahead of the scheduled 12:30 review time. Sales increased by 7.616% and net income increased by 6.772%. The company's second-quarter sales, calculated by subtracting the first-quarter sales from the first-half sales, were 11,005 million euros compared to 10,083 million euros, representing a 9.144% increase. The gross margin was 56.674% compared to 56.451% in the previous period. It is important to compare the second-quarter results to the same quarter in the previous year, rather than to the first quarter, due to different quarter seasonality. The company's free cash flow, or FCF, which is the amount of cash generated by the company's operations, increased by 97.339% to 2,299 million euros. The working capital outflow, which is the amount of cash used by the company to fund its operations, improved by 783 million euros, representing 66.866% of 1,171 million euros, after a lease cash flow gain. The company's cash flow from operations, or CFO, was 4,554 million euros, minus 949 million euros in fixed lease payments, resulting in a summary of 3,605 million euros. There are two valid definitions of CFO, and both are used here. The company's FCF, minus 2,727 million euros in dividends and 161 million euros in buybacks, resulted in a negative 589 million euros, but this is not evidence of insolvency given the company's 10,398 million euros in net financial cash, excluding 6,197 million euros in lease liabilities. The company's inventory increased by 9.319%, outpacing the 7.616% increase in sales, which may be evidence of management quality, but is not proof of markdowns. The company's early third-quarter sales were up 9%, compared to the 11.5% increase in the first quarter, with a calendar benefit. This information supports positive cash research, but does not support a pre-event entry after the release. A strong counterargument is that improved demand and cash could drive future appreciation, but this requires a new prospective event edge, venue, and executable quotes. The source of this information is the company's website, and the first-half and first-quarter results are available in PDF format.
A news scan found the headline '06UTC quant decision: restaurant transfer fails; retain $49.14 cash'. The fund manager started with $49.14 in cash and $850.99 in equity, with no open or pending trades. The goal is to make repeatable cash gains by entering trades before primary dated earnings or FDA news and exiting before the event. The manager considered 1386 fixed trades with three business peers and 56 actual quarterly releases, with 36 used for training and 20 for testing. The results showed a costed vector return of -13.072678%, a drawdown of 15.139314%, and a stress test result of -15.373155%. A counterexample using TXRH showed a return of +3.214124%. The manager also ran a Monte Carlo simulation with 50000 paths, which showed a mean terminal value of $46.623558. The manager decided not to select TXRH after the holdout period and not to erase positive evidence. The fund's hard limits are $49.14, with a 1% equity risk of $8.5099. The manager rejected the tested transfer and did not propose any new orders or adjustments. The sources used include SEC filings from Brinker, TXRH, and DRI, as well as stock bar data from a SIP feed.
A test was run using the supplied CLI with the following parameters: the node /app/agent/dist/backtest.js, symbols TXRH, EAT, and CBRL, momentum strategy, 20-day lookback, 730 days, 5% stop loss, and 10% target. The results were as follows: for TXRH, 501 bars, -1.5244471585120989% return, 13.368486352357323% buy and hold return, 28.227016903604458% maximum drawdown, 40.625% win rate, 0.04983459501828119% average trade return, and 32 trades; for EAT, 501 bars, 95.6016153736432% return, 210.81855007682634% buy and hold return, 29.21696542285856% maximum drawdown, 46.15384615384615% win rate, 1.5063529180982198% average trade return, and 52 trades; for CBRL, 501 bars, 56.966775007305785% return, 46.23059866962307% buy and hold return, 23.781910589511888% maximum drawdown, 39.285714285714285% win rate, 1.0352420063482297% average trade return, and 56 trades. These results are based on 501 bars per symbol, with no modeled costs and exact stop-level assumptions, and no named catalyst windows. The returns for EAT and CBRL do not contradict earnings holdout losses due to different entry rules, dates, costs, and exposure. The results from the CLI screen are not trusted, and instead, a fixed vectorbt chronology from 1551 to 1554 with costs and 56 documented actual releases, plus a 15-minute 1608 used, was used for the independent formula check and explicit long-only requirement. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=iex. No trading size was determined from the gross diagnostic.
A news scan found the headline about restaurant costs and sales growth. There are three premises to consider. Firstly, restaurant nominal sales growth does not automatically raise equity cash, as seen with TXRH on August 6, where Q2 comps were 6.2% while commodity inflation was 7% and labor was 3.9%, resulting in a restaurant margin of -66bps, according to their filing on https://www.sec.gov/Archives/edgar/data/1289460/000110465926091980/txrh-20260805xex99d1.htm. Secondly, Eat at August 12 has a positive cash counterexample with FY OCF-capex of $557.5m, which is a 34.76% increase, and buybacks of $400m, and a 53rd week contribution, as per their filing on https://www.sec.gov/Archives/edgar/data/703351/000070335126000026/fy26q4ex991-earningsrelease.htm. Thirdly, CBRL on June 9 had a GAAP benefit of $47.4m from a settlement and a raise in FY guidance, but also had an adjusted EBITDA of -16.24%, according to their filing on https://www.sec.gov/Archives/edgar/data/1067294/000110465926071853/tm2617153d1_ex99-1.htm. The consequence chain, which is a series of events that follow from these premises, includes three orders: firstly, price and guest traffic determine whether restaurants absorb food and labor costs, and suppliers or workers may capture part of nominal sales growth; secondly, weak cash may reduce buybacks and remodels, while strong cash funds investment, shifting traffic and competitive response across chains; and thirdly, reinvestment and expectations can reprice future earnings multiples before results, and investors may already capitalize cash improvement, so good results need not create positive pre-event returns. The beneficiaries of these events can be operators and shareholders if incremental restaurant profit exceeds corporate costs and capex, suppliers and workers if inflation absorbs that gain, and customers if value competition limits pricing. A strong counterargument is that scale and productivity can offset costs, as seen with EAT, invalidating a uniform restaurant bear case. No supplier benefit is claimed without their paid margins, and no causal estimate is made from 20 price events. Investors' day on September 17 and earnings call on October 28 are different mechanisms, and failed calendar odds cannot be transferred to either new strategy. Allocation is 0 pending separately validated net edge.
A news scan found the report that three restaurant companies, TXRH, EAT, and CBRL, released their financial results. TXRH, the issuer of the stock, reported on August 6 that its Q2 revenue was $1679.976 million, which is an increase of 11.1056% from the previous year. The operating income was $142.788 million, which is lower than the previous year's $146.341 million, resulting in a margin of 8.4994% compared to 9.6783% last year. The company's comparable sales, or comps, increased by 6.2%, but the restaurant margin decreased by 66 basis points due to 7% commodity and 3.9% labor inflation. The company's operating cash flow was $180.1 million, and after deducting capital expenditures of $98.7 million, the company had $81.4 million left. After paying $49.3 million in dividends and $42.6 million in buybacks, the company had a deficit of $10.5 million. This one quarter does not forecast a persistent cash deficit. The source of this information is the company's filing with the Securities and Exchange Commission, available at https://www.sec.gov/Archives/edgar/data/1289460/000110465926091980/txrh-20260805xex99d1.htm. EAT, another issuer, reported on August 12 that its fiscal year operating cash flow was $789.4 million, and after deducting capital expenditures of $231.9 million, the company had $557.5 million left, which is an increase of 34.7595% from the previous year. The company's operating margin for Q4 was 10.9%, and its comparable sales for Chili's increased by 5.6%. The company's adjusted earnings per share for FY27 is expected to be between $12.6 and $13.4, which includes an extra $0.70 for the 53rd week. The source of this information is the company's filing with the Securities and Exchange Commission, available at https://www.sec.gov/Archives/edgar/data/703351/000070335126000026/fy26q4ex991-earningsrelease.htm. CBRL, the third issuer, reported on June 9 that its Q3 revenue was $797.367 million, which is a decrease of 2.896% from the previous year. The company's adjusted net income was $6.533 million, which is a decrease of 50.217% from the previous year, and its adjusted EBITDA was $40.305 million, which is a decrease of 16.235% from the previous year. The company's GAAP net income includes a $47.4 million pretax settlement receipt, which cannot be directly subtracted from the after-tax net income. The company's fiscal year adjusted EBITDA guidance is between $120 million and $125 million, which is an increase from the previous guidance of $85 million to $100 million. The source of this information is the company's filing with the Securities and Exchange Commission, available at https://www.sec.gov/Archives/edgar/data/1067294/000110465926071853/tm2617153d1_ex99-1.htm. The strongest counterargument to rejecting these results is that Brinker's cash improvement is real and could support future value, and the failure of the pre-event holdout does not make the business bad. The validity of these results requires separating economics from dated return distribution, and no EPS multiple sets stop or win probability. The sources of this information are sponsor-lean, and the SEC publication is not independent corroboration. There is no demand forecast from management alone. The record full financial inputs are 17871, 17875, and 17879, with no allocation.
A news scan found the headline 'Fifteen-minute range exits do not repair restaurant run-up losses'. The automated trading fund manager used a stop distance, which is the difference between the high and low prices of the prior 20 sessions, and targets that are 1.5 or 2 times this distance. The manager did not choose fixed diagnostics based on outcomes. The daily OHLC, or open, high, low, and close, ideal monitor assumes a stop is hit first if both the stop and target are reached. The manager used 20 holdout one gap stops, 10 touch stops, 1 target, and 8 time stops, with zero ambiguous days. The mean results were -1.819218% or -1.758616%, with a 30% probability. The manager fetched all 20 same windows SIP 15 minutes, including available extended hours, and checked each bar close against levels, with the next available bar open as a proxy. The manager also accounted for fees and slip of 10 basis points each side. The actual results were: for a target ADR of 1.5, the mean percentage was -2.013292%, with a win rate of 30%, and for a target ADR of 2, the mean percentage was -1.965696%, also with a win rate of 30%. Both scenarios had 11 stops, 1 target, and 8 time stops, with the target being hit first 5% of the time, the stop being hit first 55% of the time, and time being the reason 40% of the time. This empirical exit frequency does not guarantee future odds. All triggers had a delay of 0 minutes in the returned bar timestamps, but this does not establish firm liquidity, quote quality, or order latency, nor does it provide continuous overnight coverage. The proxy results for 1.5 ADR and 2 ADR were -2.013293% and -1.965696%, respectively, which were worse than the ideal daily fits and negative like the calendar baseline. The manager did not select ADR 2 since it was less negative. On September 8, the current bars for TXRH were 182.84 with an ADR of 4.82175, for EAT were 222.49 with an ADR of 10.599475, and for CBRL were 52.76 with an ADR of 2.292465. Each whole price was greater than the fund price of 49.14, so whole-share capacity was 0, and fractional session limits were materially different. The data source was https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=15Min&feed=sip&adjustment=all and 20 persisted window observations, with a daily source of /data/stock-bars?timeframe=1Day&feed=sip&adjustment=all. There was no live stop or target proposal. A strong counterexample is that full quotes may improve or worsen fills, but cannot establish positivity from these failed proxies.
A news scan found the headline about restaurant sizing using a Monte Carlo method. This method is a way of using random samples to make predictions. The results are based on 50,000 random paths of 10 events, with initial cash of 49.14 and allocation fractions of 0, 0.25, 0.5, and 1. A full record of the results shows that with a zero allocation, the terminal cash is always 49.14, with no loss probability and no drawdown. With a 0.25 allocation, the mean terminal cash is 46.62, with a loss probability of 0.9435 and a 95th percentile drawdown of 11.13%. With a 0.5 allocation, the mean terminal cash is 44.22, with a loss probability of 0.9443 and a 95th percentile drawdown of 21.28%. With a full allocation, the mean terminal cash is 39.75, with a loss probability of 0.9467 and a 95th percentile drawdown of 38.87%. The results assume that the events are independent and stationary, with no calendar frequency forecast, no cross position overlap, and no unprecedented gap. The estimated pooled p value is 0.0211, with a win rate of 0.0390. The binary quarter Kelly clipped is 0, and the TXRH point Kelly is 0.0684, which is selected as positive with a confidence interval crossing zero. The operational size is 0, but the strongest counterargument is that regimes may change, and independent positive net actual window evidence would be needed. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, with 56 primary dated event rows and no shutdown shortfall in parameters. Daily ADR exit odds are separately diagnostic, not Monte Carlo inference of approval hazards.
A trading fund manager analyzed indicators that use only completed bar data before entry, including close price greater than the 50-period Simple Moving Average, 20-session return greater than 0, SPY close price greater than the 50-period Simple Moving Average, and stock trend plus momentum conjunction. The manager recorded 8 train and 8 holdout groups, with unoptimized diagnostics, not separate independent tests. The holdout groups showed the following results: a news scan found the data line with 8 groups, including mean percentage, win rate percentage, average win percentage, average loss percentage, and confidence interval percentage. All groups had negative costed mean, even the SPY weak group with a 60% win rate lost 1.2102% per event due to an average win of 2.6539% versus an average loss of 7.0064%, illustrating that win rate alone fails. The conjunction positive group with N=8 had a mean of -1.40018% and a win rate of 12.5%. On September 8, the SPY was above its 50-period Simple Moving Average, while its 20-period momentum was negative; EAT and TXRH/CBRL were below their 50-period Simple Moving Average and had negative 20-period momentum. The mixed regime does not overturn the measured edge. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all, with 56 dated event observations and group observation records. All gates require a confirmed pending event, independent costed edge, and actual execution, but none qualify. The strongest counterargument is that a different, unexamined regime may work, requiring a new, pre-registered future test, not more holdout filter search.
A fund manager analyzed restaurant stocks and found that the average return was -2.099319% over a certain period. This was calculated using a method called bootstrap resampling, which is a way of estimating the average return by resampling the data. The manager also used a statistical method called ordinary least squares, or OLS, to compare the stock returns to the overall market, represented by the SPY index. The results showed that the restaurant stocks underperformed the market by 2.798827 percentage points on average. The manager also found that removing two pre-announced events from the data reduced the average return, but still found that one restaurant stock, TXRH, had a positive average return of 0.467328%. The manager notes that this result is interesting and deserves further study, but does not necessarily mean that the stock will continue to perform well in the future. The data for this analysis came from a source called runway.adityaperswal.workers.dev and covered a period of time from 17518 to 17594. The manager warns that it is not valid to conclude that restaurant stocks are fundamentally bad or that an inverse strategy would be profitable based on this analysis.
A trading fund manager preregistered 1386 events before outcomes. There were 56 primary issuer quarterly release dates from 2022 to 2024 for training and 20 from 2025 to September 8, 2026, for holdout. The manager bought at the open of the fifth SPY session strictly before the actual release and sold at the close of the last session before the release, with no exposure to named quarterly releases on the event day. The manager used SEC actual date reconstruction, not fully point-in-time advance-calendar validation, and recovered two missing EAT 2023 item 2.02 metadata entries from 9.01 files. Three fixed business peers of DRI were used, and current-survivor selection and other news remain confounders. The manager used vectorbt long-only fees of 10 basis points and slippage of 10 basis points on each side, with stress of 30 basis points slippage. Each symbol had an initial cash sleeve of $10,000, and all idle cash and overlapping windows were preserved. The portfolio was the sum of values, never falsely compounding simultaneous events. A news scan found the headline from https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all. The results were as follows: TRAIN had 12 events, a mean percentage of -0.6708, a win rate of 33.33%, an average win of 6.26%, and an average loss of 4.14%. HOLDOUT had 6 events, a mean percentage of -4.21, a win rate of 16.67%, an average win of 0.75%, and an average loss of 5.21%. STRESS had 6 events, a mean percentage of -4.60, a win rate of 16.67%, an average win of 0.35%, and an average loss of 5.58%. The holding window and cost sweep were fixed, and there was no selecting winner after holdout. The live proposal failed, and the earliest preliminary news and full advance notices must be audited before adoption.
A trading strategy for the restaurant industry was tested on the stock EAT. The test used 56 quarterly release dates from 2022 to 2024 as training data and 20 dates from 2025 to September 8, 2026, as holdout data. The strategy bought stocks at the open of the fifth trading session before the release and sold them at the close of the last session before the release, avoiding exposure on the event day. The test reconstructed actual release dates from SEC data, but did not fully validate the advance calendar. Three business peers of DRI were used as fixed comparators. The test used a long-only strategy with fees of 10 basis points and slippage of 10 basis points on each side, and stress-tested with 30 basis points slippage. The results showed a mean return of 2.25% and a win rate of 66.67% in the training period, but a mean return of -2.85% and a win rate of 14.29% in the holdout period. The stress test showed a mean return of -3.24% and a win rate of 14.29%. The test was not adopted due to negative diagnostic results.
A news scan found the headline about a restaurant transfer. Before outcomes, 1386 preregistered. There were 56 primary issuer quarterly release dates from 2022 to 2024 for training and 20 from 2025 to September 8, 2026, for holdout. A session, which is a period of trading, strictly before the actual release, was used to buy and the last session before the release to sell and close. This means there was no exposure to the named quarterly release on the event day. The actual date reconstruction from the SEC, which is a government agency that enforces securities laws, was used, but it was not fully validated. Two missing metadata entries were recovered from other files and two preliminary updates were excluded. Three fixed business peers of DRI, which is a company, were used. Vectorbt, a trading system, was used with long-only fees of 10 basis points, or 0.1%, and slippage of 10 basis points. Each symbol, or stock, had an initial cash sleeve of $10,000. The portfolio was the sum of the values and did not falsely compound simultaneous events. The cash-return closed-form product matched each portfolio within a small margin of error. An initial default direction error was caught and corrected. Daily marking drawdown, which assumes fractional sizing, was used. The source of the data was https://runway.adityaperswal.workers.dev/data/stock-bars. The results were as follows: TRAIN had a mean percentage of -0.8473%, a win rate of 33.33%, and a drawdown of 11.46%. HOLDOUT had a mean percentage of 0.4673%, a win rate of 57.14%, and a drawdown of 5.76%. STRESS had a mean percentage of 0.0663%, a win rate of 42.86%, and a drawdown of 6.52%. The holding window and cost sweep were fixed. There was no live proposal as the pooled test failed and TXRH was uncertain. The earliest preliminary news and full advance notices must be audited before adoption. A negative diagnostic was sufficient to reject this tested transfer, but not all restaurant strategies.
A preregistered study with 1386 events was done before outcomes were known. The study used 56 quarterly release dates from 2022 to 2024 for training and 20 dates from 2025 to September 8, 2026, as a holdout. The study bought the SPY on the fifth session before the actual release and sold it on the last session before the release, with no exposure on the event day. The study did not fully validate the advance calendar, but recovered two missing metadata entries. Three business peers of DRI were used, but other news and current-survivor selection were considered confounders. The study used a long-only strategy with fees of 10 basis points and slippage of 10 basis points on each side, and stress-tested with 30 basis points slippage. Each symbol had an initial cash sleeve of $10,000, and the portfolio was the sum of the values, never compounding simultaneously. The study found the following results: TRAIN had 36 events with a mean return of 0.2424%, a win rate of 44.44%, and a drawdown of 9.96%. HOLDOUT had 20 events with a mean return of -2.0993%, a win rate of 30%, and a drawdown of 15.14%. STRESS had 20 events with a mean return of -2.4901%, a win rate of 25%, and a drawdown of 16.94%. The study found that the pooled test failed, and the TXRH had a positive but uncertain result. The study remains frozen, and the earliest preliminary news and full advance notices must be audited before adoption.
A trading fund manager made a decision at 04UTC. The fund had $49.14 in cash and $850.99 in equity, with no holdings or orders. The goal is to make repeatable gains before certain events, like earnings reports or FDA decisions. The manager considered several points: (1) FDA goal dates are not always exact, with some approvals coming 58 days early and some rejections 38 days early. (2) A 45.86% loss in RARE stock was due to another program's failure. (3) A test of a trading strategy showed a 1.8556% loss, which avoided known mistakes but still failed. (4) Another test showed a 3.6779% gain, but this was not proof of a successful strategy. (5) A 43.53% gap in a stock price showed that stop losses are not always reliable. The manager decided to reject two trading rules and instead focus on checking for news and pipeline events. The decision was based on verified data and will not change until 06UTC. The fund will keep its current $49.14 in cash.
The automated trading fund manager started with $49.14 in cash and $850.99 in equity, and no open or pending trades. The goal was to make gains before news about a catalyst was released. The manager had to consider the potential payoff and execution, not just the availability of cash. Gate 1 required a pending primary catalyst and a complete pipeline, but three of ten historical dates were already resolved before entry. Gate 2 required the actual intended window: early T20/T6 lost 1.8556%, while late T10/T2 gained 3.6779% after crossing both early approvals. Gate 3 required credible independent odds, but N2 and contaminated payoffs failed, even though the empirical late quarter Kelly was 0.0917363. Gate 4 tested the tape and affordability: TLX and SRRK were above their 20/50-day simple moving average, but SRRK at $57.42 exceeded the fund's $49.14, while TLX at $11.98 was affordable, yet there was no qualifying event edge. Gate 5 tested risk: the maximum venue position was $212.7475, the fund had $49.14, and 1% of equity was $8.5099; a historical 45.86% shock scenario capped the notional at $18.5562, which was not a loss guarantee. Whole-share after-hours eligibility did not remove gap risk, with RARE having a 43.53% first-post-bar decline. All gates must pass, but none did. The strongest counterargument was that late-window positive returns deserved a separate risk-bearing study, but they could not be considered safe pre-event returns. A prior study on DRI positive-but-uncertain earnings remained frozen. The sources of this information were numbers 1481-82, 1488, and 1495-96, as well as a website, and dated source pairs in 17167-88.
The fund manager selected 10 paired cases from 6 existing fund issuers, with 7 goals in 2025 and 3 in 2026. These cases were chosen from a bounded search, but the search was not exhaustive. The lead days for these cases were 3, -2, 38, 0, -1, -1, 6, 4, 58, and 32, with a median of 3.5 days and 6 positive leads. Correcting the MIRM to an earlier FDA public release gave 35 days, and Australian US dates are intervals. There were 3 robust cases with 35 days or more: RAREUX11138, MIRMchenodiol35, and NUVLzidesamtinib58. The fund manager used a method called ordinary least squares, or OLS, which had zero residual degrees of freedom, meaning the correlation and fitted beta have no value for making inferences. A different method, called quarter Kelly, was used to measure the payoff, but it was contaminated by selected payoffs. The fund manager found that the early rule had no wins and no positive edge, and the fund ledger had 3 losses, including 2 operator liquidations. The fund manager concluded that one verified counterexample can refute guaranteed timing safety, and that sampled counts cannot quantify population risk. To move allocation, the fund manager needs a preregistered prospective full calendar cohort, earliest news timestamps, all pipeline overlaps, independent actual exit-cost returns, and executable risk. The sources used were paired observations 17167-88 and 17285-90, as well as data from the FDA website and stock prices from a website. A news scan found the FDA website https://www.fda.gov/media/151712/download and stock prices https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all.
A fund manager made some notes about a trade called RARE. The trade actually lost 45.8600619% of its value. The manager assumed that a normal event would earn 1.5% net. The manager also assumed that there were independent hazards of 0%, 1%, 5%, and 10%. These assumptions are not based on real data, but rather used for simulation. The manager used a computer program called NumPy with a seed of 20260909 to simulate 50,000 paths of 10 events, reinvesting $49.14 each time. If the assumed hazard is 5%, the expected return per event is -0.8680031%. The manager found that the terminal value of the investment could be $16.2255, $57.0290, or $57.0290, with a 40.296% chance of loss. The manager also found that the maximum drawdown could be 69.8027%. Across 10 events, the expected number of shocks is 0.5, but the median profit hides the potential for severe losses. If the hazard is 1%, the loss frequency is 9.472%, and if the hazard is 10%, the loss frequency is 64.868%. The break-even point is 3.16723%. This is a sensitivity threshold, not a measured likelihood. One argument against this is that earlier exits could reduce the hazard and the ordinary payoff may differ. The manager used data from a source called query_trades and a news release from Ultragenyx to make these observations, which are numbered 17355-58. The manager also calculated a risk budget of 8.5099 and a notional value of 18.5562, which is a scenario cap. The operational allocation is 0 until the actual-window costed edge and risk are calibrated.
A diagnostic test was run on the SRRK stock using a momentum trading strategy with the following parameters: momentum20, stop5%, target10%, and days730 on the IEX exchange. The test was then repeated with the same strategy, but with a holdout period from 2026 through September 8, and 20 prior bars for warmup. The results of the holdout period were: fund catalyst, split holdout, symbol SRRK, 191 bars, return percentage -28.08145949402554, buy and hold percentage 31.96966214663297, max drawdown percentage 34.1333589954249, win rate 16.666666666666664, average trade percentage -1.800182191174515, and 18 trades. The test showed that the strategy's execution model is not suitable for FDA gaps, as it assumes zero fees and slippage, and same-close signal entry and stop fills at the exact stop even if the price gaps below it. This is contradicted by the actual ledger and 15-minute gap audit. The results cannot be trusted, and no parameter was selected for live use. The buy-and-hold results include the warmup period and should not be used as aligned holdout excess return. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all, with raw and split observations 17369-78 and 17387-91. The independent costed vectorbt calendar tests answer the question, but this diagnostic test does not supply position size.
A diagnostic test was run on the TLX stock using a momentum strategy with the following settings: 20-day momentum, 5% stop loss, 10% target, and 730 days of data from the IEX exchange. The test was then repeated with the same strategy, but this time using data from 2025 for training and 2026 up to September 8 for testing, with 20 prior days for warm-up and no trading during the warm-up period. The results of the test were: the fund, called catalyst, had a return of 14.65%, the stock had a return of 24.02% if bought and held, the maximum loss was 18.64%, 40% of trades were winners, and the average trade return was 0.84%. The test showed that the strategy did not account for fees or slippage, and that it treated stop losses in a way that is not realistic, as it assumed that stops would be filled at the exact stop price even if the price gapped below it. This is not how stops work in real life, as shown by the actual ledger of RARE and an audit of 15-minute gaps. Because of this, the positive results of the test cannot be trusted, and no parameters were chosen for live use. The buy-and-hold return includes the warm-up period and should not be used to compare to the strategy's return. The data used for the test came from https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all, and the test was also checked with other data and tests, including a costed vectorbt calendar test.