Stats

A daily cross-check did not find an exceptional Bitcoin-specific incident move. The analysis is based on three premises: (1) using 615 daily bars from January 2025 to August 31, the log return of Bitcoin can be compared to Ethereum and Solana, (2) the residuals on September 6 and 7 were -0.51944 and -0.70685 log percent, which are within the pre-period central 95% range of -2.33508 to +2.13082 log percent, and (3) daily observations contain multiple stages and unrelated forces, and Ethereum and Solana may have shared the shock. The conclusion is consistent with hourly data: these data do not show a distinct favorable Bitcoin recovery premium and do not prove that there is no causation. This study is limited to one event and a broad historical regime, and its results cannot be used to make a buy or short rule. This is an event study, not a backtest, and the two days are not independent incidents. The sources of the data are https://runway.adityaperswal.workers.dev/data/crypto-bars and https://blockstream.info/api/tx/a6d697a25266ce3c78774fd1d75f896b7af522ada209b0f6228ea497bc49a46d, and the method and results are preserved in local files daily_check.py and daily_check.json.

The QREV90804 parent return is uncertain and mostly matched market exposure. Here are the key points: 1. The primary total 5-day parent baseline from 2021 to 2025 showed a return of 41.613514%, with 192 training data points showing a return of 27.193277%, but the stress holdout showed a loss of 20.393113%. The entry and exit points were compared to the EFA gross mean return of 0.212457% per trade, versus the strategy net mean return of 0.237819%, a difference of 0.025362 percentage points. This comparison mixes net strategy and gross benchmark and does not claim formal alpha. The continuous gross EFA return was 50.789600%, and the SPY return was 94.645163%, which preserves the opportunity cost of cash. 2. There were 261 calendar weekly returns, with a 20000 circular 13-week block bootstrap, seed 90804, showing a parent mean return of 0.176101%/week, with a 95% confidence interval of [-0.084090, 0.474054]%. The managed mean return was -0.209919%, with a confidence interval of [-0.422146, -0.001114]%. The US intraday parent confidence interval was [-0.156326, 0.468542]%, and the managed confidence interval was [-0.447964, -0.033861]%. These intervals are conditional on historical samples, not future guarantees. 3. Daily OLS versus EFA with HAC5 showed a parent beta of 0.733053, with a zero-risk free annual intercept of 2.336558%, and a confidence interval of [-13.397885, 18.071002]%. The managed beta was 0.263032, with an intercept of -13.656251%, and a confidence interval of [-25.775216, -1.537287]%. This is not a proper excess return CAPM alpha. 4. Weekly cross-section regressions of 260 formations showed a mean next 5-day return slope on prior 5-day return of -0.029300, with a confidence interval of [-0.080490, 0.021891]. The US intraday showed a slope of -0.084993, with a confidence interval of [-0.172683, 0.002697], using HAC13. The Spearman means were -0.003712 and -0.025202, showing a weak relation, with neither confidence interval excluding zero. Uncontrolled country risk and currencies, 12 survivor wrappers, and previously inspected periods limit identification. 5. The parent return in 2022 was 40.594921%, versus a loss of 15.915858% in 2024. The managed return lost 4 out of 5 years. There were no rule parameter changes from these diagnostics. The source of this data is https://runway.adityaperswal.workers.dev/data/stock-bars, with full figures recorded as QREV90804 observations, reproducible using primary_diagnostics.py.

A news scan found the headline 'No allocatable event edge from three censored live losses or clinical significance'. This is a report from the Directquery fund, also known as the catalyst fund, as of September 8. The report can be found at https://runway.adityaperswal.workers.dev, specifically at the query_trades page for the catalyst fund. The report states that the total loss from three exact price and quantity losses is $31.9373 USD, with a mean loss of 16.3806%. The MIRM and IONS operator planned exits due to liquidation and censoring. One notable loss was 6.440R, using an entry price of 25.84 and an original stop price of 24, with an exit price of 13.98976. The report notes that there is no claim of precise order slippage without contemporaneous quotes. The fact that there are zero observed wins does not prove that the population has a zero win rate. The report also mentions that the average win is unidentified because there are no winning trades. The EGTX independent generic trend holdout has 56 observations, with a p-value of 0.375, a win rate of 7.383%, and a loss rate of 6.091%, resulting in a negative Kelly criterion that is clipped to 0. This means that the EGTX trend cannot be pooled with FDA trades. The report includes a descriptive one-day autocorrelation of 0.0460. A survival cohort analysis found that 228 out of 484, or 47.107%, of the cohort was retained. The selected older cohort had a hazard ratio of 0.28, which cannot be converted to FDA approval odds, as referenced in a study found at https://www.endocrine-abstracts.org/ea/0101/ea0101op-03-04. The report concludes that the premises mismatch, censoring, and negative expectancy prevent calibration, resulting in a conclusion size of 0. The evidence bar for change requires an independent dated-event holdout that is positive after costs, with a mean uncertainty above zero, plus venue and gap validation, and this is not a single rally.

A study of Bitcoin returns found that after a certain event, the return on Bitcoin was ordinary after adjusting for the overall market. The study used data from June 1 to September 8, with 2380 hourly bars and 615 daily bars, and no gaps or duplicates in the data. It used a statistical method called ordinary least squares to fit the data, and found that the relationship between Bitcoin and other cryptocurrencies, Ethereum and Solana, was strong. On September 7, between 17 and 21 UTC, the return on Bitcoin was 0.264963%, which is not unusual. The study also looked at other time periods and found that the returns on Bitcoin were all within the normal range. The conclusion is that there is no distinct edge or effect from the event, but this does not prove that the event had no effect. The study has limitations, including that it only looked at one event and the data may be affected by other factors. The data and code used in the study are available online.

A fund manager analyzed the stock PATH. The manager considered three premises: a jump in the stock's price may or may not happen again, but it can signal a change in volatility, a useful alternative to the current method must improve future forecasts, and forecast accuracy is not the same as the expected value of traded options. The manager used a formula to calculate the stock's price changes, where r is the log of the current price divided by the previous price, g is the log of the current option price divided by the previous stock price, and i is the log of the current stock price divided by the current option price. The manager clipped g at +/-3 times the prior 60-day gap root mean square and reconstructed r. The manager then calculated forecasts using a formula that combines the mean of the squared price changes and an exponentially weighted moving average. The manager used data from April 21, 2021, to September 4, 2026, and found that the current sigma base is 87.80%, the clipped sigma base is 71.85%, and the realized volatility over 60 days is 68.75%. The manager trained the model on data from 2022 to 2025 and tested it on data from 2026. The results showed that the clipped model performed better on the training data, but not on the holdout data. The manager also found that the clipped model performed better on negative events, but the results were sensitive to influential events. The manager concluded that historic downward shocks often decay, but there is no evidence that universal clipping is superior. The manager also found that the relationship between log future variance and log base/jump is sparse and not suitable for estimating a causal jump premium. The manager recorded all forecast rows, events, and inputs as observations and noted that the prior remains unresolved.

A low-risk beta exists for QLV90802, but its alpha and net swing edge are unproven or negative. The daily returns from 2021 to 2025 were analyzed using simple-return regressions with 1255 observations each. The beta values were 0.493767 for SPLV, 0.620644 for USMV, 0.655538 for EFAV, and 0.576398 for EEMV, compared to SPY, SPY, EFA, and EEM. The annualized intercepts, assuming a risk-free rate of zero, were 0.300782%, -0.251437%, 0.324073%, and 1.215566%. The 95% confidence intervals using HAC Newey West lag 5 were [-7.594128, 8.195692]%, [-5.945486, 5.442612]%, [-4.101764, 4.749910]%, and [-2.180548, 4.611681]%. These are not proper excess-return CAPM alphas, as they do not account for cash funding costs. The correlations were 0.673501, 0.846462, 0.905256, and 0.932317, indicating lower volatility supports defensive exposure, but not abnormal return certainty. A fixed 10-day swing from 2021 to 2025 resulted in returns of -13.295242% for SPLV, -13.522852% for USMV, -10.970767% for EFAV, and -6.334328% for EEMV, all of which were negative. A circular 3-month block bootstrap with 20,000 seeds and 60 monthly cohorts, including cash months, resulted in mean 95% confidence intervals of [-0.471476, -0.002576]% for SPLV, [-0.478138, -0.008777]% for USMV, [-0.352405, -0.038367]% for EFAV, and [-0.297287, 0.073146]% for EEMV. All 48 low-risk holdout variants were negative, with the best being EEMV 5-day managed at -5.591951%. A positive counterexample was found for EEMV 2026 managed, with a return of 1.087344% and a 95% confidence interval of [-0.840583, 1.131843]%, but this failed for longer windows. The conclusion is to allocate zero to QLV90802. The source of this data is https://runway.adityaperswal.workers.dev/data/stock-bars, with 877 frozen metrics and 252 rows recorded in diagnostics.py. These results are based on historical model outputs, and forward probabilities are unidentified.

A news scan found the headline '49-chain CRR and forecast audit: 41 baseline surpluses narrow to four sensitive legs'. This refers to an automated trading fund manager's notes. The premises of the audit are: 1) the model value must cover the ask, exit spread, and fees, 2) forecast variance is uncertain and completed jumps may not recur, and 3) American exercise must be modeled, while dividends and deliverables still need verification. The audit used 49 selected chain feeds from September 11 to October 16, which included 36,908 standard-root contracts, 3,209 asks less than or equal to 0.08, and 1,252 positive bids with both sizes. The audit found 41 baseline passes, with 4 at 0.8 sigma. All evaluated inputs, including IV, Greeks, timestamps, and values, were saved in observation batches. The audit adjusted SIP daily bars from 2024 to September 4, 2026, and forecasted annual variance using a sample variance of 0.5 and an exponentially weighted moving average of 0.94 of squared log returns. The audit assumed a risk-free rate of 0.04 and a dividend yield of 0. The audit checked the forecast using 10 non-overlapping session windows and found that only 17 of 49 chains were better, with a mean loss difference of 0.016576. The audit also found that PATH, RIOT, and ETHA had losses of 0.039230, 0.027439, and 0.124452, respectively, while EWJ had a gain of 0.003464. The audit used a nominal chronological holdout and selected names using the September 4 census. The audit found that PATH had a sigma of 88.8646% and a fair value of 0.291626, while RIOT had a sigma of 94.7884% and a fair value of 0.201134. The audit also found that removing a single log return from the data lowered the sigma and fair values, but this was not an authorized new forecast. The audit advanced the clock to Tuesday, 13:30 UTC, and found that the short PATH low fair value and RIOT failed. The sources of the data include https://runway.adityaperswal.workers.dev/data/stock-bars?feed=sip&adjustment=all, https://runway.adityaperswal.workers.dev/data/option-chain/PATH?feed=indicative, and https://docs.alpaca.markets/us/docs/historical-option-data.

The automated trading fund manager made some notes about the stock market. They started with some premises: widening the scope of the market requires a denominator, not just popular names. Listings, returned snapshots, current observations, and optionability are distinct concepts. IEX-only liquidity is incomplete and cannot be trusted as a consolidated ranking. The Nasdaq primary directories were created on September 4 at 21:31 and include 13,155 unique non-test symbols, including ETFs, units, warrants, and preferreds. A parser was used to preserve NA tickers via keep_default_na=False, and 384 unsupported syntax were explicitly excluded. The manager requested syntax-supported symbols, but initial SIP 403 and unthrottled IEX 429/special-symbol errors invalidated partial 1497 ranking. Recovery reused successful responses and throttled missing requests, resulting in a final 12,646 returned snapshots, with 12,488 positive close, 8,068 September 4 daily bars, and 4,578 other/missing bar dates. There were 125 syntax-supported unreturned symbols and 509 total not returned, including unsupported ones. The final retry batches had 0 HTTP failures. All measured census rows were recorded in compact record_observation batches, and scripts research/census.py/recover_final.py and raw snapshots were retained locally. The manager noted that missing rows are not zero liquidity, and old bars are not current signals. Exploratory IEX dollar volume and range surfaced IOT, PATH, SNDK, SNPS, and IREN, plus semiconductors. A country, metal, sector, and crypto-linked ETF option cohort extends outside that selection. Chain coverage includes 49 names, not all 13,155. Eight international provider quotes include Tokyo, Hong Kong, Germany, London, indexes, and FX with unaltered currencies and timestamps. The code chart Previous Close anchors 5-day range, so reported change Pct must not be called daily return. Direct foreign listed-option venue eligibility is unavailable, so US-listed country/crypto ETF proxies are used for mandate, not spot directional diversion. Resolution is in cents/individual contracts, and there is no inference that uncovered markets lack mispricing. This is a data/coverage audit, not a validated alpha screener. The sources used include https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt, https://www.nasdaqtrader.com/dynamic/SymDir/otherlisted.txt, https://runway.adityaperswal.workers.dev/data/stock-snapshots?feed=iex, and https://query1.finance.yahoo.com/v8/finance/chart/7203.T?range=5d&interval=1d.

The QLV90802 broad US, international and crypto census is based on several premises. Firstly, NasdaqTrader primary nasdaqlisted and otherlisted files stamped September 4, 2022, 21:31, retrieved on September 8, enumerate 13,155 unique non-test issues. Symbol NA is preserved as text. 384 unsupported formats, mainly due to preferred notation, are excluded with named rows, and 12,771 standard formats are requested, with 12,682 returning September 4 daily SIP bars and 89 having no return. Initial SIP snapshots are 403, then historical batches failed because ABR$D poisoned 400 symbol requests, but a repaired format filter gives 0 batch errors. Secondly, completed day close greater than or equal to 5, price times volume greater than or equal to 10 million, and security description filter yields 3,261 results. Bounded history top 600 by dollar volume, plus fixed ETFs, equals 605 requested, with 600 having greater than or equal to 100 observations, and 5 short-history excluded. 1,019 symbols are cached, including partial pass 414 extras, but not 1,019 final ranked. Derive 63 session momentum and annual volatility, 20-day median price times volume, ATR 14 and MA 100, with a 100 observation minimum and liquidity of 10 million. All 600 rows recorded 100 packed observations, with partial 594 rows retained as diagnostic only. Thirdly, 29 prelisted non-US candidates are considered, with 28 returned and ROG.SW missing, and local currency 63-day simple price returns are not FX converted or total return, with dates September 4 and 7, and not all non-US listings. Fourthly, Alpaca support table has 36 USD pairs, with all 36 daily series returned, and stablecoins are separate from directional edge. Venue volume is not global volume. Fifthly, the current census excludes delisted history and selects recent liquidity, with none of these ranks backfilled as an unbiased strategy. The original all-US request is listing filter coverage, not complete price history for every listing. The sources used are https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt, https://www.nasdaqtrader.com/dynamic/SymDir/otherlisted.txt, https://runway.adityaperswal.workers.dev/data/stock-bars, https://alpaca.markets/support/what-cryptocurrencies-does-alpaca-currently-support, https://runway.adityaperswal.workers.dev/data/crypto-bars, and https://query1.finance.yahoo.com/v8/finance/chart/. In conclusion, watch ranks only, with 0 validated entries, and missing formats, delistings, FX, venue, and limit 600 named gaps.

A direct query of trades for the Catalyst fund on September 8 found three losses and two censored exits. A Python calculation of exact entry, exit, and quantity found a mean loss of 16.3806% and a total loss of 31.9373 USD. This means that 93.322% of losses were rare, with a loss of 6.440 versus the original entry of 25.84 and stop of 24. The MIRM and IONS operators had liquidations that censored their planned horizons, which means their planned exit points were not reached. Because of this, it is not possible to infer the true win probability or the efficacy of the original targets. There is no average winning trade. The premises of this record are that it is biased and censored, and therefore cannot be used to calibrate a positive edge. As a result, the live-record Kelly is unidentifiable, and 0 was adopted. A new test, called NewPLAY, was run with a holdout of 6, and it found 2 wins out of 3, with a yield of 5.972 and a loss of 7.795, which is a positive mechanical Kelly. However, the interval crosses zero, and the training fails. It is not possible to pool different catalysts to pretend they are homogeneous. Slippage by order type and session is unavailable without decision quotes, so the return is calculated as exit minus entry, rather than execution slippage. It is possible that the results are due to bad luck, but this does not establish a probability.

The automated trading fund manager made some observations based on data from a website. They looked at 9893 out of 9897 independent trade returns from https://runway.adityaperswal.workers.dev/data/stock-bars. Using a method called bootstrap, which involves randomly selecting data, they found the following results, seeded on September 8, 2026: for ATI, the mean return was 61.9048%, with a win rate of 3.61227% and a loss rate of 2.63703%, and a 95% confidence interval of -0.481668% to 3.058334%; for IREN, the mean return was 33.3333%, with a win rate of 2.77049% and a loss rate of 6.21184%, and a confidence interval of -6.556397% to 0.179634%. They also calculated a Quarter Kelly point estimate, which is a measure of the optimal fraction of wealth to bet, of 8.523625% for ATI and 0% for IREN. However, ATI is not authorized for 8.52% because it failed training, selection, and confidence interval tests, and is missing a signal. Both ATI and IREN have a 0% authorization. The manager also looked at their own ledger and found two operator liquidation wins, with a mean return of 2.278710% and no losses, but noted that this is not a representative sample of the entire population. The manager's reasoning is that there is no robust positive edge, which means there is no reliable advantage, so they are allocating zero funds. However, the soundness of this decision is uncertain due to limitations in the data analysis, such as not accounting for regime clusters, using current ticker selection, and having few trades and unobserved future losses. A strong counter case is that ATI's 2026 profit could survive higher costs, but the manager cannot retroactively select 2026 as the sole training period. They require prospective evidence from a held-out regime, rather than re-tuning the model.

A study of 249 daily closing returns from January 2 to September 7, 2026, found that seven liquid crypto exposures share market risk. The study used daily bars from a venue and calculated descriptive distributions using pandas and numpy, as well as ordinary least squares (OLS) regression, which is a statistical method to model the relationship between variables. The daily standard deviation (SD), which measures volatility, was 2.4425% for Bitcoin (BTC), 3.3584% for Ethereum (ETH), 3.3426% for Solana (SOL), 2.5925% for Litecoin (LTC), 3.3512% for Chainlink (LINK), 4.3700% for Uniswap (UNI), and 3.4700% for Ripple (XRP). The correlation, which measures how much two variables move together, between BTC and other cryptos was: ETH 0.9123, SOL 0.8796, LTC 0.8017, LINK 0.8603, UNI 0.6637, and XRP 0.8706. The beta, which measures systematic risk, of UNI was 1.1875, with an OLS daily intercept of +0.17985% and an R-squared (R2) of 0.4405, but this alpha estimate is not suitable for trading due to non-stationarity and multiple testing. The 5/95% daily returns for UNI were -5.694/+7.731, and for BTC were -3.512/+4.236. The study concludes that: premise 1, multiple majors are not independent risk; premise 2, UNI has lower co-movement but higher own volatility; and premise 3, there is no predictive causation from regression. Therefore, simultaneous positions must share risk budget and UNI requires separate test. However, it is possible that future correlations may be lower, and covariance is period-specific. Prior fast fees were likely binding, and should only be changed if an independently costed edge is positive. All observations were recorded from the source https://runway.adityaperswal.workers.dev/data/crypto-bars.

A study on the QTM90800 calendar premium found that it lacks reliable evidence for its existence later on. The study used a statistical method called ordinary least squares, or OLS, on daily adjusted returns, which is a way to measure the relationship between variables. The results showed that for the SPY stock, the average daily return was 0.0154245% on certain days, compared to 0.0694268% on other days, a difference of -0.0540023%, but this difference was not statistically significant, with a HAC t-value of -0.7563. For the EEM stock, the difference was +0.0517793%, with a t-value of 0.6756, but again, this was not statistically significant. The study also used a method called bootstrap, which involves randomly selecting samples of data, to test the results. The bootstrap results showed that the average return for the SPY stock was -0.191696%, with a 95% confidence interval of -0.523982% to 0.159359%, and for the EEM stock, the average return was +0.246076%, with a 95% confidence interval of -0.196141% to 0.680338%. The study found that the correlation between the SPY and IWM stocks was 0.766949, and between the EFA and EEM stocks was 0.540052, indicating that these stocks are related. The study concluded that the results do not support the existence of the QTM90800 calendar premium, and that more research is needed to understand the relationship between these stocks. The data used in the study came from https://runway.adityaperswal.workers.dev/data/stock-bars, and the code used to analyze the data is available in the diagnostics.py file, with observations from 9801 to 9805.

A news scan found the following report: Premises used Python to analyze daily log returns through September 7. The report found that the correlations between Ethereum, Solana, and Litecoin with Bitcoin were 0.88435, 0.79804, and 0.71898, respectively. This means that these cryptocurrencies tend to move together with Bitcoin. The report also used a statistical method called Ordinary Least Squares, or OLS, which is a way to measure the relationship between variables. The OLS results showed betas of 1.27648, 1.15520, and 0.84627, indicating how much each cryptocurrency moves when Bitcoin moves. The report analyzed 90 days of data from OKX, a cryptocurrency exchange, and found that the open interest, or OI, which is the number of outstanding contracts, declined from 21:15 to 00:03 by 0.65494%, 0.45312%, and 0.76039% for Bitcoin, Ethereum, and Solana, respectively. The report also looked at the current funding forecast, which is a prediction of the cost of holding a position, and found it to be positive at 0.293149%, 0.494613%, and 0.057946% after settlement. However, the report concluded that it cannot confirm that alternative positions are diversified, nor can it infer that there is fresh leverage expansion or spot carry income. The report's validity is limited to descriptive statistics, which describe the characteristics of the data, and is only based on 90 days of data from one venue, OKX. The report also noted that the relationships between the cryptocurrencies are contemporaneous, meaning they happen at the same time, and are based on a survivor universe, which means that only the cryptocurrencies that have survived are included in the analysis. One counterexample to the report's findings is that an idiosyncratic, or unique, catalyst for Litecoin could decouple it from the other cryptocurrencies. The beneficiaries of more turnover, or trading activity, are revenue and liquidity providers, who earn fees and have less risk capacity. The report retained the existing frozen hourly rule expectancy of -0.633%, -0.707%, and -0.573% from a previous analysis, but did not make any new claims about these estimates being current. The sources of the data used in the report are https://runway.adityaperswal.workers.dev/data/crypto-bars, https://www.okx.com/api/v5/public/open-interest, and https://www.okx.com/api/v5/public/funding-rate. The report did not provide a confidence interval, which is a measure of the reliability of the results, for the serially correlated OLS, and did not make any causal trade, or claim that one variable causes another.

The automated trading fund manager has analyzed BWEN's trades. Two sets of data, Observation9305 and backtest749, contain 12 closed trades from 2026. These trades were used to calculate the possible range of returns, which is between -9.561459% and 1.242761%. This range is based on 20,000 random samples of the 12 returns. However, this range is not a guarantee of future performance due to the small sample size and other factors. The probability of a win is estimated to be 0.25, with an average win of 8.465371% and an average loss of 8.003927%. The Kelly criterion, a formula used to determine the optimal bet size, is 1.05765. The live supply ledger shows two wins averaging 2.278710%, but no losses. The prior mean interval of -0.874362% to 1.958541% remains unresolved. The goal is to achieve a repeatable positive net return, but this has not been achieved yet. The fund manager requires more evidence of independent payoff before making any decisions.

The audit of a fund's performance shows a significant loss. The fund had 3 closed trades, with no wins and no open or pending trades. A calculation using Python found that the total loss was $31.93, with an average loss of 16.38%. One loss of $29.80 made up 93.32% of the total losses, and the actual loss was 6.44 times the planned loss. The stop loss was overshot by $10.01 per share. Two trades were closed by the operator, which affected the intended exit strategy. The conclusion is that the sample does not show a positive edge in terms of sizing, and the average winner is not available. The validity of the results is limited by the small sample size and mixed exit strategies. It is noted that future trades can still be winners, even if none were observed in this sample. The risk of stop-level losses is not the same as the maximum gap loss. To improve the results, a full event map and holdout period should be used instead of just deadline-only entries. The price data used is based on actual venue records, not quotes or modeled fills.

The manager of an automated trading fund has made some premises. The first premise is that prior forecasts may be wrong and rejection is conditional rather than proof of efficiency. The second premise is that a revised volatility must clear costs and be independently validated. The third premise is that European sensitivity is not American executable valuation, meaning that the way to calculate option values in Europe is different from the way it is done in America, where an option can be exercised at any time. A new Python code using the scipy Brent inversion method has been used in the research, which is located in the file research/hurdles.py, and it has 12 rows of observations from 9266 to 9277. The analysis was done on September 4, and it looked at the prices of two securities: TLT, which had a price of $82.22, and EEM, which had a price of $68.72. The bid and ask prices for TLT were $79.50, $0.02, and $0.03, and the volatility was 0.094090586. The bid and ask prices for EEM were $73.50, $0.03, and $0.07, and the volatility was 0.178580666. The Black-Scholes model, which is a mathematical model used to calculate the value of an option, was used to calculate the values of the options, with assumptions of a risk-free rate of 0.04, a dividend yield of 0, and a multiplier of 100. The model was solved for different hypothetical round-trip dollars, and the results showed that the baseline calendar values were below the zero-fee hurdles. The economic sensitivity of the results was potentially material, but the prior underforecast of the variance of EEM and the post-hoc volatility exceeded the root, which means that the results may not be reliable. The hurdle is a diagnostic fair value proxy, not an attainable bid or profit expectation. The required net target bid was calculated, and the results showed that the zero-fee calendar roots were not enough to rescue the frozen values. A news scan found the sources of the prior venue chains and underlying stock prices, which were deliberately not refreshed during the holiday, and all source quotes were dated September 4.

An audit was done at 19:40. The results show that the net expectancy and quarter Kelly remain negative. The audit was based on two premises. First, past experiments with codes 593-598 were reviewed, but not repeated or adjusted. The results of these experiments showed that the returns on investments in Bitcoin, Ethereum, and Solana were -52.1851%, -58.5744%, and -49.2381% per hour, with losses of 52.2339%, 66.9159%, and 60.9755% at their worst points. The average wins were 0.9911125444%, 4.1772817551%, and 5.200698903%, while the average losses were 1.1805209889%, 1.2598696402%, and 1.302715925%. Second, a calculation using the formula E=p*w-(1-p)*l, where E is the expected value, p is the probability of winning, w is the amount won, and l is the amount lost, showed that the expected value per trade was -0.6328916631%, -0.7069389899%, and -0.5733609910% for Bitcoin, Ethereum, and Solana respectively. The quarter Kelly, a measure of the optimal fraction of the portfolio to invest, was -0.1596417245%, -0.0423085532%, and -0.0275617278%, but was set to zero since it was negative. The third premise was that daily samples were not enough to establish a reliable edge across different frequencies, and one of the trades was stopped by the operator, which affected the results. The conclusion is to keep the current rejection and zero allocation. The audit results are valid, but only apply to these specific tests and not all cryptocurrency strategies. A new rule might work, but there is no current evidence to support it. The inputs are recorded in Observation9254. The sources used were https://runway.adityaperswal.workers.dev/data/crypto-bars, https://docs.alpaca.markets/us/docs/crypto-fees, and query_trades fund=crypto. No new backtest is claimed.

A complete fresh query of the quant fund on September 7 at 19:41 UTC found 3 closed operator liquidations and 0 live orders. A recalculation using Python numpy showed a win rate of 1/3, with an average win of 7.939941%, an average loss of 5.261432%, and a mean return of -0.860974%. The gross realized loss was -1.046053 USD. The quarter Kelly calculation, which is a formula used to determine the optimal bet size, was 0.3. However, due to truncated 4-6 week planned horizons and missing arrival quotes, it was not possible to calibrate forward probability, exit frequency, or slippage. As a result, the conclusion is that there is zero empirical allocation, and the N3 strategy edge is not reliable. A repeated ledger audit did not find any new backtest results.

A review of the frozen strategy at 16:30 found that the negative net expectancy kept the Kelly criterion at zero. The review looked at archived backtests 593-598 for Bitcoin, Ethereum, and Solana, which showed average wins and losses. For Bitcoin, the average win was 29 out of 115, or 29.11%, with an average loss of 1.18%. For Ethereum, the average win was 12 out of 118, or 4.18%, with an average loss of 1.26%. For Solana, the average win was 12 out of 107, or 5.20%, with an average loss of 1.30%. Using the formula E = p*w - (1-p)*l, where E is the expectancy, p is the probability of winning, w is the average win, and l is the average loss, the expectancy was calculated to be -0.63% for Bitcoin, -0.71% for Ethereum, and -0.57% for Solana. The raw quarter Kelly criterion was also calculated to be -0.16 for Bitcoin, -0.04 for Ethereum, and -0.03 for Solana. The total returns were -52.19% for Bitcoin, -58.57% for Ethereum, and -49.24% for Solana, with drawdowns of 52.23% for Bitcoin, 66.92% for Ethereum, and 60.98% for Solana. The daily number of trades was insufficient, with 0 trades for Bitcoin, 2 trades for Ethereum, and 1 trade for Solana. The review concluded that the strategy should be preserved without re-tuning the parameters. The inference is valid under the required cross-frequency and net edge gate, but it does not prove that every future crypto strategy will be unprofitable. The bullish partial liquid return and funding changes are a new context, but not independent validation. The sources used were https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees, with a query of trades from the crypto fund.

The automated trading fund manager reviewed the quant ledger for QTA9071630. The ledger shows 3 trades that were liquidated by the operator and 0 live or queued orders. The results of these trades are as follows: the fund won 1 out of 3 trades, with an average win of 7.93994% and an average loss of 5.26143%, resulting in a mean return of -0.860974% and a gross profit and loss of -1.046053 USD. The quarter Kelly, a measure of the optimal fraction of the portfolio to invest, is calculated as 0.3. All trades were closed within 4-6 weeks, which is shorter than the intended horizon. Since there are no stop or target exits, it is not possible to estimate their future frequencies. Additionally, missing arrival quotes prevent the measurement of slippage. The conclusion is that there is no calibrated forward edge and no positive empirical allocation. This analysis is based on existing ledger data and is not a new backtest. The source of this information is a query of trades from the quant fund on September 7 at 16:31 UTC.

A review of frozen edge found negative expectancy, which keeps the Kelly criterion at zero. Premises from vectorbt/CLI experiments 593–598 and analysis 714 have shown that Bitcoin has 29 wins out of 115 with an average win of 9.911125444% and an average loss of 1.180520989%, Ethereum has 12 wins out of 118 with an average win of 4.177281755% and an average loss of 1.259869640%, and Solana has 12 wins out of 107 with an average win of 5.200698903% and an average loss of 1.302715925%. Using the formula E=p*w-(1-p)*l, where E is the expectancy, p is the probability of winning, w is the average win, and l is the average loss, the expectancy is -0.63289% per trade for Bitcoin, -0.70694% per trade for Ethereum, and -0.57336% per trade for Solana. The raw quarter Kelly criterion is -0.159642 for Bitcoin, -0.042309 for Ethereum, and -0.027562 for Solana, which clips to 0. The holdout returns are -52.1851% for Bitcoin, -58.5744% for Ethereum, and -49.2381% for Solana, and the drawdowns are 52.2339% for Bitcoin, 66.9159% for Ethereum, and 60.9755% for Solana. Daily N is 0 for Bitcoin, 2 for Ethereum, and 1 for Solana, which is inadequate. The own trade 10 operator liquidation cannot calibrate strategy exits or wins. The conclusion is to retain rejection, with no new backtest and no retuning of already inspected holdout. The validity failed required net edge, which implies zero approved size. The soundness is limited to these strategies and historical data, and is not proof that all future crypto edges are negative. A strong counterexample with a new regime or new rule could work, but would require separately specified independent net daily/hourly validation and 15-minute execution before change. Existing simulations 605 and 606 remain scenarios, not edge. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and internal 593–598 and query_trades fund crypto. The costs are listed at https://docs.alpaca.markets/us/docs/crypto-fees.

The QTA9071350 live calibration remains unchanged with a zero empirical quarter-Kelly. A recent review of the quant13 fund's trades, which are trades made using a quantitative approach, at 13:51 UTC found 3 closed trades. The results showed that 1 out of 3 trades were winning trades, with a mean winning return of 7.93994%, a mean losing magnitude of 5.26143%, and a mean return of -0.860974%. The total gross realized loss was -1.046053 USD. All trades were closed by the operator on September 4, before the 4-6 week signals were completed, and there were no stop or target exits, which are predetermined prices to buy or sell a security. Since there are no quotes available when trades are entered, the slippage, which is the difference between the expected price and the actual price, is unknown. The quarter-Kelly formula, which is used to calculate the optimal bet size, is max(0, 0.25 * (p - (1-p) / (avg win / avg loss))) and equals 0, indicating no forward statistically calibrated edge. The proposed size will be retained at 0. This information comes from the primary internal query_trades source, based on saved observation 9172, and is a recalculation of the ledger, not a new backtest, which is a simulation of how a trading strategy would have performed in the past.

A review of archived experiments 596–598 shows the following results: for Bitcoin, 29 out of 115 trades were winning, with a win percentage of 99.11125444% and a loss percentage of 1.180520989%; for Ethereum, 12 out of 118 trades were winning, with a win percentage of 4.177281755% and a loss percentage of 1.259869640%; for Solana, 12 out of 107 trades were winning, with a win percentage of 5.200698903% and a loss percentage of 1.302715925%. Using the formula E = p * w - (1 - p) * loss, where E is the expectancy, p is the probability of winning, w is the win percentage, and loss is the loss percentage, the expectancy per trade is -0.63289% for Bitcoin, -0.70694% for Ethereum, and -0.57336% for Solana. The raw quarter Kelly is -0.159642 for Bitcoin, -0.042309 for Ethereum, and -0.027562 for Solana, which clips to 0. The total holdout returns are -52.1851% for Bitcoin, -58.5744% for Ethereum, and -49.2381% for Solana, with drawdowns of 52.2339% for Bitcoin, 66.9159% for Ethereum, and 60.9755% for Solana, based on 115, 118, and 107 trades respectively. Daily calendar sensitivity from experiments 593–595 shows that 0, 2, and 1 trades cannot establish robustness against hourly failure. The ledger shows that one Bitcoin operator liquidation is a censored strategy exit, with no average win, slippage, or stop calibration. The conclusion is to maintain rejection, as this is an arithmetic review of archived vectorbt/CLI research, not a new backtest or independent holdout. A counterexample may be that a different future rule may work, but fresh funding or quotes do not supply its conditional payoff. Older bar provenance, 15-minute stop fill stress, and independent loss samples remain binding. Existing simulation sensitivity from experiments 605 and 606 is not new evidence, and there is no need to retune, repeat CLI, or run new Monte Carlo simulations to change a negative edge. The source of this information is internal experiments 593–598 and 659, as well as query_trades fund crypto, with additional data from https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. Observation 9171 carries exact arithmetic.

A news scan found the report of a live record calibration for QTA9071230. The report states that a query of trades for a fresh fund returned only three trade IDs, 9, 7, and 1, all of which were closed on September 4 by the operator. The report includes the exact arithmetic for entry, exit, and quantity, which was saved as observation 9144. This observation showed a win rate of 1 out of 3, with an average winning return of 7.93994% and an average loss of 5.26143%, resulting in an unweighted mean return of -0.860974% and a total realized gross profit and loss of -1.046053 USD. The report also notes that the holds, which lasted 20.9794, 43.2302, and 47.2386 hours, were shorter than the intended 4-6 weeks. Additionally, the stop-hit and target-hit counts were both 0, as all exit reasons were discretionary, meaning they were chosen by the operator. The report concludes that there is no live-record sizing support due to the empirical quarter Kelly being 0, which is a measure of the optimal fraction of wealth to bet, calculated as max(0, 0.25 * (p - (1-p) / (avg win / avg loss))). The source of this information is a primary internal query of trades for the fund, retrieved on September 7 at 12:32 UTC, and is based on observation 9144.