A news scan found the headline that Oddity Tech reported its second-quarter 2026 results. The results for the quarter ending June 30 show that the company's revenue decreased and it had an operating loss, even though it had a positive net income. This was because the company had income from financial sources and a gain from paying off debt, which makes it seem like the company's operations are not recovering as well as they might be. The company's operating cash flow minus money spent on property, equipment, and intangibles is -$15.557 million. It has $561.162 million in cash and investments, but owes $550 million on a note, leaving $11.162 million before considering other obligations. Management expects revenue to decrease by 5% in the third quarter and by 19% for the full year, but believes the advertising problem can be solved. However, it is possible that the company's improvement could be better than expected. The conclusion is that the idea that earnings will improve is only a hypothesis and does not give the company a price advantage. There is no independent verification of this hypothesis. The information comes from the company itself, specifically from its September 9 report on the quarter and first half of the year ending June 30, and management is in favor of the idea that things will return to normal.
The notes from January 1724 to 1729 are frozen. The fund still has one-third of its initial capital in Bitcoin and Ethereum each, but the planned investment in Solana was not made because 418 calendar bars are missing, which prevents valid training. This is a reduced-coverage diagnostic, not a replication of a three-asset strategy. The monthly fraction of positive returns is one-third, and the next calendar day's open is used. The fund has no leverage, and there is a fee of 25 basis points plus 10 to 35 basis points per side. The data from 2023 to 2024 is used for training, and 2025 and 2026 are used for holdout. There are no missing calendar prints in Bitcoin and Ethereum. The final period's close is used for costed liquidation, and the sample includes complete position episodes with partial resizes aggregated, not independent monthly bets. The training data contains no losses and does not estimate losing tails. A mandatory short-horizon range with 15-minute exits and hourly robustness is unvalidated, and negative holdout rejects live adaptation. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The results show a return of 18.0028%, a drawdown of -9.674%, and a win rate of 50% over 4 samples, with an average win percentage of 13.6349% and an average loss percentage of -14.7976%.
A news scan found the headline 'QWIDE91002 crypto reduced-coverage monthly holdout 60bp'. Frozen notes 1724/1729 indicate that Bitcoin and Ethereum each retain one-third of the initial capital, while Solana stays in cash due to 418 missing calendar bars that prevent valid training. This is a reduced-coverage diagnostic, not a 3-asset replication. The strategy has a monthly positive-return fraction of 1/3/12, with the next calendar day's open, inactive cash at 0%, no leverage, and a fee of 25 basis points plus 10/35 basis points slip per side. The training period is from 2023 to 2024, and the holdout period is from 2025 to 2026, with no missing calendar prints in Bitcoin and Ethereum. The final period close is costed liquidation, and the sample is complete position episodes with partial resizes aggregated, not independent monthly bets. Training N2 contains no loss and does not estimate the losing tail; the 0.25 arithmetic Kelly is not validated. Mandatory short-horizon range and 15-minute exits, as well as hourly robustness, are unvalidated, and negative holdout rejects live adaptation. The source is https://runway.adityaperswal.workers.dev/data/crypto-bars, which provides the following data: the family is crypto_monthly, the signal is mix, the era is holdout, the cost side is 0.006, the return is -9.194244272417295, the drawdown is -18.2231279353811, the win rate is 33.33333333333333, the sample is 3, the average win percentage is 2.159762501145231, the average loss percentage is 18.741634168681625, the mean trade percentage is -11.774501945406007, the quarter Kelly diagnostic is 0, the calendar days are 365, and the symbols are BTC/USD and ETH/USD.
A news scan found the notes from a crypto trading fund manager. The notes are about a reduced-coverage monthly training session that resulted in a 60 basis point return. The fund had 1724 out of 1729 notes frozen. The fund kept one-third of its initial capital in Bitcoin and Ethereum, but the planned investment in Solana was kept in cash due to 418 missing calendar bars that prevented valid training. This training session was not a replication of a 3-asset portfolio. The training had a few rules: the fund would invest one-third of its capital on the first day of each month, the next day's opening price would be used, no cash would be kept inactive, and no leverage would be used. The fund also paid a 25 basis point fee and a 10-35 basis point slip per trade. The training data was from 2023 to 2024, and the holdout data was from 2025 to 2026. There were no missing calendar prints in the Bitcoin and Ethereum data. The final period's close was used to calculate the cost of liquidation, and the sample included complete position episodes with partial resizes aggregated, not independent monthly bets. The training data, called N2, had no losses and did not estimate the losing tail. A mandatory short-horizon range and 15-minute exits and hourly robustness were not validated. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The results of the training session were: return of 86.84%, drawdown of -27.53%, win rate of 100%, sample size of 2, average win percentage of 24.20%, average loss percentage of 0%, mean trade percentage of 24.20%, quarter Kelly diagnostic of 0.25, calendar days of 731, and symbols of BTC/USD and ETH/USD.
A news scan found the headline 'QWIDE91002 crypto reduced-coverage monthly 2026 35bp'. Frozen notes 1724/1729 indicate that Bitcoin and Ethereum each retain one-third of the initial capital, while Solana stays in cash due to 418 missing calendar bars that prevent valid training. This is a reduced-coverage diagnostic, not a 3-asset replication, which means it is a limited test, not a full copy of a three-asset investment. The monthly positive-return fraction is one-third, with the next calendar day's open, inactive cash at 0%, no leverage, and a fee of 25 basis points plus 10/35 basis points per side, where basis points refer to a unit of measurement for interest rates and fees. The training period is from 2023 to 2024, the holdout period is from 2025 to 2026, and there are no missing calendar prints in Bitcoin and Ethereum. The final period close is costed liquidation, and the sample is complete position episodes with partial resizes aggregated, not independent monthly bets. Training N2 contains no loss and does not estimate the losing tail, where the losing tail refers to the potential for large losses. The 0.25 arithmetic Kelly is not validated, and mandatory short-horizon range and 15-minute exits and hourly robustness are unvalidated, where the Kelly criterion is a formula for bet sizing. Negative holdout rejects live adaptation. The source of this information is https://runway.adityaperswal.workers.dev/data/crypto-bars, which provides data on cryptocurrency prices. The data includes a family of crypto monthly signals, a mix signal, and an era of 2026, with a cost per side of 0.0035, a return of 0.548747461410759, a drawdown of -9.563996719218704, a win rate of 50, a sample size of 4, an average win percentage of 14.19247202665343, an average loss percentage of 14.397035091432208, a mean trade percentage of -0.10228153238938872, a quarter Kelly diagnostic of 0, 252 calendar days, and symbols including BTC/USD and ETH/USD.
A news scan found the headline 'QWIDE91002 crypto reduced-coverage monthly holdout 35bp'. Frozen notes 1724/1729 show that Bitcoin and Ethereum each retain one-third of the initial capital, but Solana stays in cash because 418 missing calendar bars prevent valid training. This is a reduced-coverage diagnostic, not a 3-asset replication, which means it is a test with limited data, not a copy of a typical 3-asset investment. The monthly positive-return fraction is 1/3, the next calendar day open is used, inactive cash is 0%, there is no leverage, and a fee of 25 basis points plus a 10/35 basis points slip per side is applied. A basis point is a unit of measurement for interest rates and fees. The data is split into training from 2023 to 2024, holdout from 2025 to 2026, with no missing calendar prints in Bitcoin and Ethereum. The final period close is costed liquidation, and the sample is complete position episodes with partial resizes aggregated, not independent monthly bets. Training data N2 contains no loss and does not estimate the losing tail, which refers to the part of the data that shows potential losses. The arithmetic Kelly, a formula used to determine the optimal bet size, is not validated at 0.25. Mandatory short-horizon range and 15-minute exits, as well as hourly robustness, are unvalidated, and negative holdout rejects live adaptation. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The provided data includes a return of -8.644818908014573, a drawdown of -18.14438187989059, a win rate of 33.33333333333333, a sample size of 3, an average win percentage of 2.640778186404199, an average loss percentage of 18.351569692690266, a mean trade percentage of -11.354120399658777, a quarter Kelly diagnostic of 0, 365 calendar days, and the symbols BTC/USD and ETH/USD.
A news scan found the headline 'QWIDE91002 crypto reduced-coverage monthly train 35bp'. Frozen notes 1724/1729 show that Bitcoin and Ethereum each retain one-third of the initial capital. Solana was planned to be the third, but it stays in cash because 418 missing calendar bars prevent valid training. This is a reduced-coverage diagnostic, not a 3-asset replication. A monthly train has a 1/3/12 positive-return fraction, next calendar-day open, inactive cash 0%, no leverage, and a fee of 25 basis points plus 10/35 basis points slip per side. The train data is from 2023-24, the holdout data is from 2025 and 2026, and there are no missing calendar prints in Bitcoin and Ethereum. The final period close is costed liquidation, and the sample is complete position episodes with partial resizes aggregated, not independent monthly bets. Training N2 contains no loss and does not estimate the losing tail; the 0.25 arithmetic Kelly is not validated. Mandatory short-horizon range and 15-minute exits and hourly robustness are unvalidated, and negative holdout rejects live adaptation. The source is https://runway.adityaperswal.workers.dev/data/crypto-bars, which provides data on crypto bars. The results include a return of 89.53%, a drawdown of -27.30%, a win rate of 100%, a sample size of 2, an average win percentage of 24.77%, an average loss percentage of 0%, a mean trade percentage of 24.77%, a quarter Kelly diagnostic of 0.25, 731 calendar days, and symbols including BTC/USD and ETH/USD.
The QWIDE91002 swing trading strategy, which is 3-of-3 for the year 2026 with a cost of 16 basis points per side, has been analyzed. The strategy, also known as Frozen1720/1721, uses a same universe approach with 50 ETFs, where the first monthly session enters only when all 1/3/12 excess signs are positive, with a prior 20-day average daily range stop at 1 and target at 1.5, and half of the position is added with a 1R, or risk-reward ratio, which is a measure of potential profit versus potential loss. The strategy aims to break even the next day, with a maximum of 5 sessions. Daily open gap ordering is known, otherwise, the stop is at the first bar with ambiguity, which refers to a situation where the market direction is unclear. Opening profits are filled conservatively at a certain level. Fees are 1 basis point plus 5/15 slip, which refers to the cost of buying or selling a security. The vectorbt amount order ledger reconciles custom cash per asset to less than 1e-8, which is a very small amount. Synthetic 9/12/16 hour slots encode ordering, but observed fill times are not considered. Drawdown, which is the decline in value from a peak, is sampled and marked with an approximation, with 15-minute overnight unvalidated data. The strategy is equally funded in 50 independent sleeves, not pooled, with 100% per trade compounding. The data sources are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is swing, era is 2026, cost side is 0.0016, return is -1.6855682639409064, drawdown is -2.805276757007169, win rate is 51.162790697674424, sample is 215, average win percentage is 1.1626302801816983, average loss percentage is 2.021004709127899, mean trade percentage is -0.3921682029695003, quarter Kelly diagnostic is 0, monthly clusters are 9, mean holding is 2.353488372093023, and vectorbt reconciled is true.
The QWIDE91002 swing trading strategy had a holdout period with the following results: it used a 3-of-3 rule, meaning three conditions had to be met to make a trade, and had a cost of 16 basis points per side, which is a measure of the cost of buying or selling a security. The strategy was backtested on data from 1/3/12, with a stop loss of 1 times the average daily range and a target of 1.5 times the average daily range, with half of the position sized at 1 times the average true range, a measure of price volatility. The strategy had a breakeven point the next day and a maximum of 5 sessions. The daily open gap ordering was known, otherwise the strategy used a stop-first bar ambiguity, meaning it would stop trading if the first bar of the day was ambiguous. Opening profits were filled conservatively at a certain level. The fees were 1 basis point plus 5/15 slip, which refers to the cost of buying or selling a security. The vectorbt amount order ledger reconciled custom cash per asset to less than 1e-8, which is a very small number. The strategy used synthetic 9/12/16 hour slots to encode ordering, but did not observe fill times, which refers to the time it takes to execute a trade. The drawdown was sampled using a mark approximation, with 15 minute overnight unvalidated data. The strategy was equally funded in 50 independent sleeves, not pooled, with 100% per trade compounding. The data sources used were https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics of the strategy were: family was swing, era was holdout, cost side was 0.0016, return was -8.228292501395773, drawdown was -9.15806359987802, win rate was 43.82566585956417, sample was 1239, average win percentage was 0.9965922379632003, average loss percentage was 1.39147340747158, mean trade percentage was -0.3448877371962889, quarter Kelly diagnostic was 0, monthly clusters were 72, mean holding was 2.4624697336561745, and vectorbt reconciled was true.
A news scan found the headline 'QWIDE91002 swing 3-of-3 train 16bp/side'. This refers to a trading strategy, where 16bp/side means 16 basis points per side, a basis point being one hundredth of a percent. The body of the note mentions Frozen1720/1721, which is a specific trading period, and Same50ETFuniverse, referring to a trading universe with 50 equally traded assets, or etfs, which are exchange-traded funds. The strategy enters a position only when all signs are positive, with a prior stop at 1 and target at 1.5 times the average daily range, or adr, with half of the position plus one times the risk, or r, and breakeven the next day, with a maximum of 5 sessions. Daily open gap ordering is known, otherwise the strategy stops at the first bar with ambiguity, and opening profits are filled conservatively at a certain level. Fees are 1 basis point plus 5 to 15 slip, which refers to the difference between the expected and actual price of a trade. The note also mentions vectorbt, a backtesting library, which reconciles the amount order ledger with custom cash per asset to less than 1e-8, a very small number. Synthetic 9 to 12 and 16 hour slots encode ordering, but observed fill times are not considered, and drawdown, or loss, is sampled and marked with an approximation every 15 minutes overnight, but is unvalidated. The strategy is equally funded in 50 independent sleeves, not pooled, with 100 percent per trade compounding. The note cites two sources, including a stock bars data feed and a research paper from AQR. The full metrics of the strategy are also provided, including a return of -5.3426 percent, a drawdown of -5.3643 percent, and a win rate of 37.09 percent.
The QWIDE91002 swing trading strategy, which is 3-of-3 for the year 2026 with a cost of 6 basis points per side, has been analyzed. This strategy, also known as a trading system that buys and sells stocks based on certain rules, started with a frozen position in 1720/1721 and used the same universe as the 50 ETFs, with the first monthly session only entering long positions when all signs were positive, with a stop loss at 1 times the average daily range and a target at 1.5 times, and half of the position was closed at a profit of 1 times the risk, with a breakeven stop on the next day and a maximum of 5 sessions. The daily open gap was used for ordering, and if there was ambiguity, the stop was placed at the first bar, with opening profits filled conservatively at the specified level. The fees were 1 basis point plus a 5 to 15 slip, and the vectorbt amount order ledger was reconciled with a custom cash per asset of less than 1e-8. The strategy used synthetic 9, 12, and 16 hour slots to encode ordering, but did not observe fill times, and used a drawdown sampled mark approximation, with 15 minute overnight unvalidated data. The strategy was equally funded in 50 independent sleeves, and was not pooled with 100 percent per trade compounding. The data was sourced from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics for the strategy are: family is swing, era is 2026, cost per side is 0.0006, return is -0.837390516077452, drawdown is -2.6756370346532887, win rate is 52.093023255813954, sample size is 215, average win percentage is 1.3435491263306152, average loss percentage is 1.863295064902291, mean trade percentage is -0.19275297458561413, quarter Kelly diagnostic is 0, monthly clusters is 9, mean holding period is 2.353488372093023, and vectorbt reconciled is true.
The QWIDE91002 swing trading strategy had a return of -3.5853043981237565% during the holdout period. This strategy, also known as a 3-of-3 holdout, uses a 6 basis points per side cost structure. A basis point is a unit of measurement, one hundredth of a percent. The strategy was tested on data from the period starting 1/3/12, with a prior 20-day average daily range stop at 1 and target at 1.5, and half of the average true range plus 1. The daily open gap ordering was known, otherwise the strategy used a stop at the first bar's ambiguity, which refers to uncertainty in the market. Opening profits were filled conservatively at a certain level. The fees were 1 basis point plus 5 to 15 slip, which refers to the difference between expected and actual trading costs. The vectorbt amount order ledger was used to reconcile custom cash per asset to less than 1e-8, a very small number. The strategy used synthetic 9, 12, and 16 hour time slots to encode ordering, but did not observe fill times, which refer to the times when trades are executed. The drawdown, or peak to trough decline, was sampled using a mark to market approximation, which estimates the value of a portfolio, and 15-minute overnight unvalidated data. The strategy was equally funded in 50 independent sleeves, not pooled, with 100% per trade compounding. The source data came from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and a research paper from AQR, https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics of the strategy are available, including a win rate of 46.48910411622276%, average win percentage of 1.1366843295470295%, and average loss percentage of 1.2592056317905695%.
A news scan found the headline 'QWIDE91002 swing 3-of-3 train 6bp/side'. The Frozen1720/1721 strategy, also known as a trading plan, uses a Same50ETFuniverse approach, which is a type of trading universe, meaning a set of rules to select stocks. This strategy enters a position only on the first monthly session with all excess signs positive, with a prior 20-day average daily range stop at 1 and a target at 1.5, and half of the position size plus one risk unit. The position is set to breakeven the next day, with a maximum of 5 sessions. The daily open gap ordering is known, otherwise a stop is set at the first bar with ambiguity, which means uncertainty. Opening profits are filled conservatively at a certain level. Fees are 1 basis point plus 5 to 15 slip, which refers to a small percentage of the trade value. The vectorbt amount order ledger reconciles custom cash per asset to less than 1e-8, which is a very small number. Synthetic 9, 12, and 16 hour slots encode ordering, but observed fill times are not considered. Drawdown, which is a measure of loss, is sampled and marked with an approximation, with 15-minute overnight unvalidated data. The strategy is equally funded in 50 independent sleeves, which are like separate accounts, and not pooled at 100 percent per trade compounding, which means the returns are not combined. The source of this information is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is swing, era is train, cost side is 0.0006, return is -2.635043849171137, drawdown is -2.943814217606766, win rate is 42.73624823695346, sample is 709, average win percentage is 0.7446188077803125, average loss percentage is 0.8841538956012193, mean trade percentage is -0.18807754986835026, quarter Kelly diagnostic is 0, monthly clusters are 34, mean holding is 2.3413258110014104, and vectorbt reconciled is true.
A news scan found the headline 'QWIDE91002 monthly passive 2026 16bp/side'. This refers to a monthly report on an automated trading fund, QWIDE91002, with a 16 basis points per side fee. The fund uses a strategy called 'passive', meaning it does not actively try to beat the market. The fund has 50 predetermined cash ETF sleeves, which were set up in 2016 and trained from 2017 to 2019, and then held from 2020 to 2025. The fund's performance is measured against the SHY index, which is a type of bond index. The fund's return is calculated at the end of each month, and it is a mix of positive and negative returns. The fund does not select specific stocks, but instead uses a full sleeve approach, meaning it invests in all the available ETFs. The fund has fees of 1 basis point and slippage of 5 to 15 basis points. The fund also has a feature called'vector bt position episodes', which refers to the way the fund adjusts its positions over time. The fund does not use a risk balancing strategy, unlike some other funds, such as those offered by AQR. The fund's metrics are as follows: it has a return of 13.72%, a drawdown of -8.12%, a win rate of 86%, and an average win percentage of 17.07%. The fund's annual return is 20.67%, and its mean daily percentage is 0.0788%. The fund's volatility is 14.22% per year, and its Sharpe ratio, which measures risk-adjusted return, is 1.40. The fund currently has no open positions. The data for this report comes from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf.
The QWIDE91002 monthly passive holdout had a return of 16 basis points per side. The fund uses 50 predetermined cash ETF sleeves, which were warmed up in 2016, trained from 2017 to 2019, and held out from 2020 to 2025, with a separate period in 2026. The month-end return is calculated by subtracting the SHY return, which is a type of bond fund. The fund has a mix of positive fractions and components diagnostics, but no selection is made, and the full sleeve is held passively. The next session will open with a monthly target weight within independently compounding sleeves, with 1 basis point fees and 5 to 15 slippage, and the final era will close with costed liquidation. The fund also has vector bt position episodes that aggregate partial resizes, and correlated positions that are not independent and identically distributed, which means they are not random and do not have the same distribution. Inactive cash is zero, and expenses are embedded in prices. This is a long cash adaptation, not a risk-balanced short futures strategy like the one used by AQR, and it does not have mandatory ADR exits. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is passive, cost per side is 0.0016, era is holdout, return is 77.39577595673421, drawdown is -29.686335533123042, win rate is 94, sample is 50, average win percentage is 82.91660269182037, average loss percentage is 8.96818326635472, quarter Kelly diagnostic is 0.23337761383573194, annual return is 10.031989087023785, mean daily percentage is 0.04343735205773449, annual volatility percentage is 16.481427168339913, Sharpe ratio with zero cash is 0.6641544210186046, and open positions is 0.
The QWIDE91002 monthly passive train had a return of 16 basis points per side. The system used 50 predetermined cash ETF sleeves, with data from 2016 used for warmup, 2017-2019 for training, 2020-2025 for holdout, and 2026 for separate testing. The system's performance was measured by month-end returns minus the SHY index. The results showed a positive mix of fractions with no selection, using a full sleeve passive approach. For the next session, the system will open with a monthly target weight within independently compounding sleeves, with 1 basis point fees and 5-15 slippage, and final era close costed liquidation. The system also used vectorized backtest position episodes with aggregate partial resizes, and correlated positions that were not independent and identically distributed, giving it an edge. Inactive cash was set to zero, with expenses embedded in prices. This is a long cash adaptation, not a risk-balanced short futures strategy like AQR's, and does not require mandatory ADR exits. The system's metrics are as follows: family is monthly, signal is passive, cost per side is 0.0016, era is train, return is 33.04901076986295, drawdown is -17.822829741004465, win rate is 98, sample size is 50, average win percentage is 33.94444098527958, average loss percentage is 10.661824731701195, quarter Kelly diagnostic is 0.24342951829191636, annual return is 10.021907782507732, mean daily percentage is 0.039939242162321704, annual volatility percentage is 10.187884991733197, Sharpe ratio with zero cash is 0.987907601339425, and open positions is 0. The data sources are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf.
A news scan found the headline 'QWIDE91002 monthly passive 2026 6bp/side'. The fund manager's notes are about a trading strategy. The strategy uses 50 predetermined cash ETF sleeves, which were set up in 2016 and tested from 2017 to 2019, and then not changed from 2020 to 2025. In 2026, the strategy is separate. At the end of each month, the strategy looks at the return minus the return of SHY, which is a type of low-risk investment. The strategy mixes the positive fractions of the return and does diagnostics on the components, but does not select any. The strategy is passive and uses the full sleeve. The next session will open with a monthly target weight within independently compounding sleeves, with 1bp fees and 5/15 slippage, and the final era close will be costed with liquidation. The strategy also looks at vector bt position episodes and aggregate partial resizes, and notes that correlated positions are not independent, which is called an IID edge. Inactive cash is zero, and expenses are embedded in prices. This strategy is a long cash adaptation, not a risk-balanced short futures strategy like AQR's. There are no mandatory ADR exits, so it is unapproved. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is passive, cost side is 0.0006, era is 2026, return is 13.952572096657988, drawdown is -8.119084725235936, win rate is 86, sample is 50, average win percentage is 17.302009882450044, average loss percentage is 6.612579607424165, quarter Kelly diagnostic is 0.20162350282815392, annual return is 21.024684009024732, mean daily percentage is 0.07995452287652759, vol annual percentage is 14.214391569886033, Sharpe zero cash is 1.4174746534751963, and open positions is 0.
The QWIDE91002 monthly passive holdout had a 6 basis points per side performance. The system uses 50 predetermined cash ETF sleeves, which were warmed up in 2016, trained from 2017 to 2019, and are being held out from 2020 to 2025, with 2026 being separate. The month-end return is calculated by subtracting the SHY return, and the mix has a positive fraction with three components and diagnostics, but no selection is made. The passive full sleeve is used. For the next session, the open monthly target weight is within independently compounding sleeves, with 1 basis point fees and 5 to 15 basis points slippage, and the final era close is costed with liquidation. The system also considers vector bt position episodes with aggregate partial resizes, and correlated positions are not independently and identically distributed, which means they are not random and unconnected. Inactive cash is zero, and expenses are embedded in prices. This is a long cash adaptation, not an AQR risk-balanced short futures strategy, and there are no mandatory ADR exits, hence it is unapproved. The source data can be found at https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is passive, cost side is 0.0006, era is holdout, return is 77.75092292197321, drawdown is -29.68633553312302, win rate is 94, sample is 50, average win percentage is 83.28282239156854, average loss percentage is 8.785917230151693, quarter Kelly diagnostic is 0.23341757574169797, annual return is 10.06869769989811, mean daily percentage is 0.04356934609810567, volatility annual percentage is 16.48038216244716, Sharpe zero cash is 0.6662148431084862, and open positions are 0.
A news scan found the headline 'QWIDE91002 monthly passive train 6bp/side'. The fund manager's notes mention Frozen1720/1721/1724, which refers to a set of frozen or fixed positions. The manager uses fifty predetermined cash ETF sleeves, which are groups of exchange-traded funds that track a specific index or sector, with a warm-up period in 2016, training from 2017 to 2019, a holdout period from 2020 to 2025, and a separate period in 2026. The manager calculates the return minus the SHY, which is a type of bond fund, and signs, or indications, of positive or negative returns. The manager mixes the positive fraction, or part, of the return with three components and diagnostics, without selecting specific securities, and uses a passive full sleeve approach, meaning all securities are included without active management. The next session will open with a monthly target weight within independently compounding sleeves, with 1 basis point fees, which is a unit of measurement for interest rates or fees, and 5 to 15 basis points slippage, which refers to the difference between expected and actual prices. The manager also mentions vector bt position episodes, which refers to a type of statistical analysis, and aggregate partial resizes, which means adjusting the size of positions. The manager notes that correlated positions, or positions that move together, are not independent and identically distributed, or IID, which is a statistical concept, and that there is no edge, or advantage. Inactive cash is set to zero, and expenses are embedded in prices. This is a long cash adaptation, not an AQR risk-balanced short futures strategy, which means the manager is focusing on long-term cash investments rather than short-term futures contracts. The manager also mentions that there are no mandatory ADR exits, which refers to American Depositary Receipts, and that this strategy is unapproved. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is passive, cost side is 0.0006, era is train, return is 33.315375355762036, drawdown is -17.822829741004387, win rate is 98, sample is 50, average win percentage is 34.21261824564346, average loss percentage is 10.482949371648012, quarter Kelly diagnostic is 0.2434679703119502, annual return is 10.095532305485566, mean daily percentage is 0.04020459105128569, volatility annual percentage is 10.187861252772898, Sharpe zero cash is 0.9944733927512432, and open positions is 0.
The QWIDE91002 monthly report for December 2026 shows a 16 basis points per side performance. The fund uses 50 predetermined cash ETF sleeves, which were warmed up in 2016, trained from 2017 to 2019, and held out from 2020 to 2025, with a separate period in 2026. The month-end return is calculated by subtracting the SHY return, with a mix of positive fractions and component diagnostics, but no selection, and a passive full sleeve approach. The next session will open with a monthly target weight within independently compounding sleeves, with 1 basis point fees and 5/15 slippage, and final era close costed liquidation. The fund also uses vector bt position episodes with aggregate partial resizes, and correlated positions that are not independently and identically distributed, which means they are not random and unrelated. Inactive cash is set to zero, and expenses are embedded in prices. This is a long cash adaptation strategy, not a risk-balanced short futures strategy like AQR, and it does not require mandatory ADR exits. The source data is from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is 12, cost per side is 0.0016, era is 2026, return is 12.652638875446364, drawdown is -7.9762544242265925, win rate is 65, sample is 60, average win percentage is 18.27104928228229, average loss percentage is 3.786189909436414, quarter Kelly diagnostic is 0.1443679479236617, annual return is 19.012928221907654, mean daily percentage is 0.07311567997987553, volatility annual percentage is 13.917123248215363, Sharpe ratio with zero cash is 1.323919536121903, and open positions is 0.
A news scan found the headline 'QWIDE91002 monthly 12 holdout 16bp/side'. This refers to a trading fund's monthly report, where 12 represents the signal and 16bp/side is the cost per side, with bp meaning basis points, or a unit of measurement for interest rates and investment returns. The fund has a frozen set of 50 cash ETF sleeves, which are groups of investments, and uses data from 2016 as a warm-up, 2017-2019 for training, 2020-2025 as a holdout, and 2026 as a separate period. The fund's return is calculated at the end of each month, minus the return of the SHY investment, which is a type of bond. The fund has a mix of positive fractions and diagnostics, but does not select specific investments. Instead, it uses a passive full sleeve approach. The next session will open with a monthly target weight within independently compounding sleeves, with 1bp fees and 5/15 slippage, which refers to the difference between the expected and actual price of a trade. The fund also has vector bt position episodes, which are periods of buying or selling, and aggregate partial resizes, which means adjusting the size of the investments. The fund's expenses are embedded in the prices, and it is a long cash adaptation, meaning it focuses on holding cash for an extended period. This is different from an AQR risk-balanced short futures approach, which involves selling investments to reduce risk. The fund does not have mandatory ADR exits, which are rules for exiting investments. The source of this information is https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is 12, cost side is 0.0016, era is holdout, return is 31.297330195582873, drawdown is -15.795237483747048, win rate is 43.66197183098591, sample is 284, avg win pct is 22.28359889309118, avg loss pct is 6.8880149180692545, quarter kelly diagnostic is 0.06561874178027188, annual return is 4.646044077827849, mean daily pct is 0.01977521205153735, vol annual pct is 9.295417883603923, sharpe zero cash is 0.5361085966643298, and open positions is 0.
The system known as QWIDE91002 has been analyzed. It uses a method called Frozen1720/1721/1724, which involves fifty predetermined cash ETF sleeves. The data was divided into periods: 2016 for warmup, 2017–19 for training, 2020–25 for holdout, and 2026 for a separate evaluation. The return at the end of each month is calculated by subtracting the return of SHY, which is a type of low-risk investment. The system mixes a positive fraction of the components with diagnostics but does not select any particular component; instead, it uses a passive full sleeve approach. For the next session, the system will open with a monthly target weight within independently compounding sleeves, considering 1 basis point fees and 5/15 slippage, and will finalize with a costed liquidation at the end. The system also aggregates partial resizes of vector bt position episodes and acknowledges that correlated positions do not have independent and identically distributed edges. Inactive cash is set to zero, with expenses embedded in prices. This system is a long cash adaptation and does not involve short futures like AQR risk-balanced strategies, and it does not require mandatory ADR exits. The data sources include https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and a research paper from AQR. The full metrics of the system are provided, showing details such as return, drawdown, win rate, and average win and loss percentages.
A news scan found the headline 'QWIDE91002 monthly 12 2026 6bp/side'. The fund manager's notes for QWIDE91002 mention that certain positions, referred to as Frozen1720/1721/1724, were frozen. The fund has fifty predetermined cash ETF sleeves, which were set up in 2016 as a warmup, trained from 2017 to 2019, held out from 2020 to 2025, and are separate for 2026. The fund's return at the end of each month is compared to the return of SHY, a type of bond fund, and the results are used to adjust the mix of investments. The fund's components are diagnosed, but no selection is made, and the full sleeve is used passively. For the next session, the monthly target weight will be set within independently compounding sleeves, with fees of 1 basis point and slippage of 5 to 15 basis points. The final liquidation will be costed at the end of the era. The fund's vector bt position episodes will be aggregated and partially resized. Correlated positions are not independent and identically distributed, which means they are not random and do not have the same distribution, and are referred to as an edge. Inactive cash is set to zero, and expenses are embedded in prices. This is a long cash adaptation, not a risk-balanced short futures strategy like the one used by AQR, and does not require mandatory ADR exits. The data sources for this information are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics for the fund are also provided, including a family of monthly, a signal of 12, a cost side of 0.0006, an era of 2026, a return of 12.9164524179755, a drawdown of -7.969869188682088, a win rate of 66.66666666666666, a sample of 60, an average win percentage of 18.049197418516112, an average loss percentage of 3.781317519453858, a quarter Kelly diagnostic of 0.14920828411473333, an annual return of 19.420340712904572, a mean daily percentage of 0.07447154968361418, a volatility annual percentage of 13.907936465750502, a Sharpe ratio of 1.3493612489879947, and zero open positions.
The QWIDE91002 monthly report for the 12 holdout period with a 6 basis points per side fee is as follows. The system uses 50 predetermined cash ETF sleeves, with data from 2016 used for warmup, 2017-2019 for training, 2020-2025 for holdout, and 2026 for separate testing. The month-end return is calculated by subtracting the SHY return, with a mix of positive fractions and component diagnostics, but no selection. The system uses a passive full sleeve approach. For the next session, the open monthly target weight is determined within independently compounding sleeves, with 1 basis point fees and 5/15 slippage, and final era close costed liquidation. The system also considers vector bt position episodes and aggregate partial resizes, with correlated positions not being independently and identically distributed, which is referred to as an edge in trading, meaning an advantage. Inactive cash is set to zero, with expenses embedded in prices. This is a long cash adaptation strategy, not a risk-balanced short futures strategy like the one used by AQR. There are no mandatory ADR exits, hence it is unapproved. The data sources are https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is 12, cost per side is 0.0006, era is holdout, return is 32.714881480437356, drawdown is -15.527755926989506, win rate is 44.36619718309859, sample size is 284, average win percentage is 22.170973223244573, average loss percentage is 6.7865232234357515, quarter Kelly diagnostic is 0.06834179862516034, annual return is 4.833632844144531, mean daily percentage is 0.020486728282477263, annual volatility percentage is 9.293483328543937, Sharpe ratio with zero cash is 0.5555135081942556, and open positions is 0.
The QWIDE91002 monthly report for 12 train 6 basis points per side is as follows. The system uses 50 predetermined cash ETF sleeves, with data from 2016 used for warmup, 2017-2019 for training, 2020-2025 for holdout, and 2026 for separate testing. The month-end return is calculated by subtracting the SHY return, which is a type of bond fund. The system mixes the positive fraction of the return with three components for diagnostics, without selecting any particular component, and uses a passive full sleeve approach. For the next session, the system will open with a monthly target weight within independently compounding sleeves, with 1 basis point fees and 5-15 basis points slippage, and will account for final era close costed liquidation. The system also aggregates partial resizes of vector bt position episodes and accounts for correlated positions that are not independent and identically distributed, or IID, which refers to a statistical concept where events are independent and have the same distribution. Inactive cash is set to zero, and expenses are embedded in prices. This system is a long cash adaptation, not a risk-balanced short futures strategy like AQR, and does not have mandatory ADR exits. The source data comes from https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all and https://www.aqr.com/-/media/AQR/Documents/Insights/Journal-Article/AQR-JPM-Fall-2017.pdf. The full metrics are: family is monthly, signal is 12, cost side is 0.0006, era is train, return is 12.64417830055451, drawdown is -13.779417138856044, win rate is 61.48648648648649, sample is 148, average win percentage is 9.956674741798178, average loss percentage is 4.926073466009103, quarter Kelly diagnostic is 0.10607973066214499, annual return is 4.062784104918227, mean daily percentage is 0.01686899490862431, volatility annual percentage is 7.360692195381743, Sharpe ratio with zero cash is 0.577525401706172, and open positions is 0.