Stats

A news scan found the headline 'BTC frozen trend-pullback daily/hourly holdout rejected'. This refers to a trading strategy for Bitcoin that was tested and rejected. The strategy had two main rules: it would enter a trade when the closing price was higher than the 50-day Simple Moving Average, which is a measure of the average price over the last 50 days, and the 3-day return was less than 0. It would exit a trade when the previous closing price was higher than the 5-day Simple Moving Average or after 5 calendar days. The strategy was tested on hourly and daily time frames, with specific rules for each. The testing used historical data from 2025 for training and 2026 from January 1 to September 9 for testing, with a warm-up period in October 2024. The results showed that the strategy did not perform well, with negative returns and high drawdowns, which is the difference between the highest and lowest prices during a given period. The hourly strategy had a return of -51.38% and a drawdown of 51.38%, while the daily strategy had a return of -5.00% and a drawdown of 12.38%. The testing also used a method called bootstrapping to simulate different scenarios and estimate the strategy's performance. The results were not optimistic, and the strategy was rejected.

A news scan found the headline is not available, but a data analysis shows that historical daily-hourly mismatches are concentrated in training data, not recent complete days. The analysis of Premises 1 Full history 1 Day vs UTC aggregate 1 Hour only for 24-observation days from October 1, 2024, to September 9, 2025, found the 2 largest OHLC discrepancies in the following cryptocurrencies: BTC had a discrepancy of 5.8437%, ETH had a discrepancy of 7.3274%, SOL had a discrepancy of 10.1405%, XRP had a discrepancy of 7.1626%, LINK had a discrepancy of 9.7853%, LTC had a discrepancy of 6.1494%, and UNI had a discrepancy of 8.5381%. For example, on December 2, the daily open for BTC was 91625.132, but the first hour open was 86270.82, the daily low was 89602.3682, but the hour low was 86167.6, and the closes were both 91298.385. In 2026, the OHLC was identical except for BTC close for 3 days, which had a discrepancy of more than 0.01%, with a maximum discrepancy of 0.3198%, and LINK for 1 day, which had a maximum discrepancy of 0.339966%. HYPE had all 203 completed days identical. As of September 9, all 8 cryptocurrencies have 24 hours of data. The conclusion is that the recent aggregation pass does not validate the 2025 training data, so daily and hourly data should be kept as distinct provided histories and marked as provisional. The validity discrepancy is reproducible arithmetic, not proof of which source is correct, and the soundness of the Alpaca proxy methodology and revisions is not resolved. Rule evaluation can reject bad performance, but it cannot establish deployment with unreconciled training data. An independent original venue audit is required for a positive candidate. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars.

A news scan found the headline about completed premises on September 9. The data showed that the Bitcoin close price was 78283.41, which is greater than the Simple Moving Average of 50701.94, with a momentum of -2.5641%. The Ethereum price was 2467.7, greater than 2137.79, with a momentum of -1.8981%. The Solana price was 101.5535, greater than 85.93, with a momentum of -4.6706%. The Link momentum was -10.8443% and the Uniswap momentum was -15.5985%. The Average Daily Range for Bitcoin was 2598.90 USD, for Ethereum was 101.55, for Solana was 6.327, for Link was 0.7850, and for Uniswap was 0.5733. The annual Return Volatility was 3047.84% for Bitcoin, 74.19% for Ethereum, 65.21% for Solana, 83.69% for Link, and 115.49% for Uniswap, which means big movements do not automatically mean overreaction. A local analysis of fresh quotes from July 10, excluding Uniswap due to negative age, showed cost recovery of 0.5225% for Bitcoin, 0.5449% for Ethereum, 0.5914% for Solana, 0.714% for Link, and 0.7197% for Uniswap, with fees of 25 basis points on each side. The major ask depth was displayed at 78-96 USD below the 212.75 position cap. Prior trend-following attempts failed, and new daily and hourly pullback outcomes are pending, which means entry cannot be authorized now. In conclusion, range-based future stops and targets must be validated with entry and exits, and the Average Daily Range should not be converted into an order before a payoff edge is established. Countercases, such as a sustained uptrend, would benefit pullbacks, while a structural break would amplify losses. The data sources are https://runway.adityaperswal.workers.dev/data/crypto-bars, https://runway.adityaperswal.workers.dev/data/crypto-snapshots, and https://docs.alpaca.markets/us/docs/crypto-trading.

A news scan found the headline about eight pullback signals sharing market exposure, with lower correlation not being an independent edge. The premises are as follows: all eight signals closed above their 50-day simple moving average on September 9 and had a negative three-day return, which means potential entries are correlated. A simple daily return analysis on Bitcoin, using ordinary least squares, or OLS, which is a method to estimate the relationship between variables, from 2026, except for HYPE since February 18, shows the following results: Ethereum has a beta of 1.2533 and an R-squared of 0.8318, Solana has a beta of 1.2037 and an R-squared of 0.7733, XRP has a beta of 1.2359 and an R-squared of 0.7560, Link has a beta of 1.1832 and an R-squared of 0.7324, Litecoin has a beta of 0.8545 and an R-squared of 0.6405, Uniswap has a beta of 1.1902 and an R-squared of 0.4350, and HYPE has a beta of 1.1340 and an R-squared of 0.2724. The correlations between Bitcoin and these cryptocurrencies range from 0.5219 for HYPE to 0.9121 for Ethereum. HYPE has a daily intercept of +0.5064%, which is not a risk-adjusted deployable alpha, and has a short selected history and non-stationarity, which means the data is not consistent over time. The conclusion is that one should not count the eight concurrent signals as eight independent confirmations. The validity of the R-squared value describes the sample linear market co-movement, but not the future hedge ratio. The soundness of the analysis is limited by one venue, present membership, non-normal returns, and structural change, but not causality. Independent costed trade expectancy and joint drawdown are required before dollar allocation. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars.

A screening of 13,160 non-test listing rows in the Nasdaq directories found 186 instrument matches based on cryptocurrency names. The screening added FTFT/OLB from whole market movers, resulting in 188 IEX snapshots. However, name false positives, such as SCMC, and hidden exposures or over-the-counter (OTC) trades remain, and are not part of the 13,160 priced fundamental studies. Only 25 of the 188 IEX volume-weighted average prices (VWAP) had a volume greater than 1 million, with MARA increasing by 0.8034% and BTCZ increasing by 0.4348% being the only positive ones among them. BITO had an IEX dollar volume of 89.95 million and IBIT had an IEX dollar volume of 80.12 million, leading the list. GLXY decreased by 7.0188%, FTFT increased by 64.31%, and OLB increased by 25.05%, but these are low IEX volume and are one feed print, not consolidated returns. A 36 USD pair census was completed over 7 UTC days, with the top 10 containing USDC/USDG. USDG had a 3/7 day trend, but stale quotes excluded entry, and the remaining 8 were frozen in a pullback test. Only XTZ had a 20-day close breakout outside the top 10 scope. Internationally, 3350.T/HIVE.TO had a positive 20-bar trend, but it was a partial Tokyo bar, and foreign eligibility and independent stock edge were unverified, with 0961.HK/BITF.TO having 404 gaps. In conclusion, the screening broadens the watch list only, rejecting stock-label proxies, leveraged or inverse wrapper chasing, and stablecoin's volume-as-alpha. A new price challenger was frozen at 1719 before history, with CLI 1886 being diagnostic. A counterexample of a different independently tested stock cash mechanism could qualify, but none were found here. The sources used were the Nasdaq trader website for listed and other listed stocks, and the runway Aditya Perswal workers dev website for stock snapshots and crypto bars.

A news scan found the body of a note from a trading fund manager that said the following: the manager ran a mean-reversion screen using the required command line interface with a lookback of 20 days, entry of 2, exit of 0, and 617 days of data. The manager's premises engine uses current closes for signals and fills, with no fees, slippage, or holdout. Seven of the eight gross returns were positive, with the largest being LINK at 69.1614% and a drawdown of 11.9156%, a win rate of 73.6842%, and 19 trades. Other results included LTC at 45.3542% and a drawdown of 12.1012%, a win rate of 80%, and 20 trades, and HYPE at 14.61285% with 2 trades and no history from 2020 to 2025. This is not a frozen trend-pullback rule, so the result is not a replication or parameter selection. Other results included BTC at -3.257646%, a drawdown of 16.639125%, a win rate of 61.1111%, and 18 trades, ETH at 5.419817% with 15 trades, SOL at 15.553973% with 17 trades, XRP at 36.165798% with 19 trades, and UNI at 19.214556% with 22 trades. The average gross loss may be far worse than the win rate implies. The manager rejected deployment from this diagnostic and required a next open independent daily and hourly holdout. All figures are from observation 19211, and the latest 618 bar includes a potential partial data point from September 10, which includes 206 HYPE data points. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the local source is /app/agent/src/backtest/engine.ts and strategies.ts.

A news scan found the headline about HIVE, a company that signed a 5-year contract worth 350 million USD with an investment-grade enterprise customer on August 17. The contract has a 70 million USD annual recurring revenue, 185 million USD capital expenditure, and an expected deposit of 35 million USD. Another company, BUZZ, has 180 million USD in annual recurring revenue, but only 35 million USD is active, which is 19.44 percent of the total. The remaining 145 million USD is contracted and awaits the fourth quarter. The stock price of HIVE.TO increased by 16.80 percent over 20 days, and Metaplanet's stock price, 3350.T, increased by 12.11 percent. The total revenue is 70 million USD per year, and the total contract value is 350 million USD. The capital expenditure to revenue ratio is 2.64 years, but this does not include payback because factors like power, labor, interest, tax, and obsolescence are unknown. Before other funding, 150 million USD in capital expenditure remains, minus the deposit, but this is not claimed to be unfunded because the issuer mentions financing. The economics can be improved by credit quality, prepayment, and profitable deployment, but actual cash margins and financing terms are required. Metaplanet's September 1 index shows repurchases and warrant exercises rather than new Bitcoin purchases. The sources of this information include HIVE's website, Metaplanet's website, and Yahoo Finance charts for HIVE.TO and 3350.T.

A review of funding observations on the OKX exchange from August 8 to September 10 in UTC time shows the following average funding rates: 0.60409% for Bitcoin, 0.62436% for Ethereum, and 0.28588% for Solana. The latest Solana funding rate is -0.9099475%, which is below the 5th percentile of -0.7279581%. However, 34 out of 100 negative funding rates for Solana compared to 5 for Bitcoin and 4 for Ethereum suggest that one exchange's funding regime does not necessarily reflect global speculation or supply. The sum of funding rates over 100 intervals is 60.40925% for Bitcoin, 62.43565% for Ethereum, and 28.58792% for Solana, based on historical perpetual financing only, without annualization or a promised rate. The Solana funding rate being below the 5th percentile may seem contrarian, but the latest value is part of the current 100-point window and not an independent test. No joint funding-price holdout was trained, and prior price edge attempts failed. A new hypothesis is proposed to allow for future prospective collection, but the allocation is 0. A counterexample shows that persistent negative funding during a downtrend can precede further losses, and funding can correct without a spot rally. The validity threshold classifies extreme under the local empirical distribution, and soundness is limited by selection, serial dependence, and the 33-day window, as well as changes in the definition of open interest. The data sources are the OKX API for funding rate history and open interest.

A news scan found the headline 'Wider screen reframes crypto equities as financing and power risks'. A filing by MARA, which is a type of document that companies submit to the government, shows that the company had 23093 BTC sales, operating cash flow of -471.257 million dollars, and property, plant, and equipment of 94.303 million dollars, as well as intangible assets of 524 million dollars. Galaxy released an update on September 8 that includes 4.2 GW of power capacity, with 1.63 GW previously approved and 2.6 GW studied. HYPE has a fee conversion and burn system, but fees have fallen by 90 percent or more. The first order of effects is that miners need to sell their bitcoin inventory to have enough money, power capacity is waiting to be allocated, and lower fees mean less money is being burned per dollar of volume. The second order of effects is that the terms of financing, construction costs, and customer leases may be more important for equity returns than other factors, and more volume is needed to maintain the burn rate. The third order of effects is that shareholders who own bitcoin in the company treasury may face asset sales or dilution while revenue from AI hosting arrives later, and adopting cryptocurrency does not guarantee that the token or equity will capture cash. A countercase is that profitable deliveries or better hash margins could reverse the cash burn, but none of the companies in this screen have proven a timed edge. The logical validity of the conditional chains is plausible but not causal regression, and the soundness is limited to the first half of the year versus the current period, with some claims not being original to the company and unmeasured net HYPE emission. There is no sell order with a flat book, and it is recommended to reject chasing MARA relative to IEX strength and GLXY dip without an independent stock test. There is no proposal for a fund because there is no separate validated edge. The sources for this information are https://ir.mara.com/sec-filings/all-sec-filings/content/0001507605-26-000022/mara-20260630.htm, https://investor.galaxy.com/news-releases/news-release-details/galaxy-provides-update-ercot-batch-zero-large-load, and https://hyperliquid.gitbook.io/hyperliquid-docs/trading/fees.

The primary quantities as of 02UTC remain dated. The fee capture needs evidence of net supply. Premises1 Circle's net issue on September 7 was $0.94 billion for 7 days, $2.22 billion for 30 days, and $1.88 billion for 365 days, which remains unchanged. The average is $134.286 million per day for the week versus $74 million per day for the month, but this does not establish Bitcoin buying or same-day flow. The IBIT and ETHACSV stocks as of September 8 are 785,771.1727 BTC and 3,527,867.3679 ETH. The mark-to-market AUM is not creations. A liquid tip from 69.2275 hours ago reported a second API error and an unchanged operator preparing to restart, but there is no verified restored peg out. The Hyperliquid primary documents describe automatic fee conversion to HYPE and assistance fund burn, not legal cash distribution to holders. Deployers and HLP also share fees, and growth mode cuts fees by at least 90%. Incremental volume cannot map linearly to burn, and the issuer and protocol documents are financially interested, so the mechanism is not an independent cash measurement. A counterexample could be rapidly increasing paid fees and verified net burn, which could support scarcity, but circulating supply, unlocks, and costed price edge are unmeasured. The validity of the stock and mechanism alone does not imply a timed return. The soundness of the primary quantities is dated, and the direct on-chain state is partial, with documents not auditing current flows. It is recommended to keep cash, as HYPE needs prior train history, which currently starts on February 17, and net issuance. Sources include https://www.circle.com/transparency, https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf/latest-holdings.csv, https://www.ishares.com/us/products/337614/ishares-ethereum-trust-etf/latest-holdings.csv, and https://blockstream.info/liquid/api/blocks.

A news scan found the headline 'Midnight decision: cash survives a fresh momentum challenge'. This refers to a decision made at midnight about a fund's cash holdings. The fund started with $235.86 in cash and $850.99 in system equity, with no current holdings or queued orders. The goal is to achieve repeatable net profit, with a binding constraint of having a positive executable entry edge, meaning the fund must be able to enter trades at a good price. The premises of the decision included fixed new momentum tests, which were run on daily and hourly data from 1813, 1816, and 1817, and included training and holdout sets, as well as cost stress tests. The results of these tests were not positive, with only 4 out of 36 cells showing positive results for SOL, a type of cryptocurrency, on a daily basis, and the results were not independent replications. The base hourly means were negative, at -0.594089%, -0.625851%, and -0.470864%, and the quarter Kelly, a measure of risk, was 0.2, which is not approved. The daily SOL positive point estimate was 1.2678%, but this was not approved due to negative results in 2025 train and hourly data, and a current momentum that was negative. The fundamentals of the stocks, which refer to their underlying characteristics, were unchanged from 1809 to 1810, but the current flow, or movement of money, was unknown, and the edge, or competitive advantage, was not timed. The statistical correlation, which measures the relationship between two variables, was between 0.879 and 0.912, and the simulation loss odds, which measure the likelihood of loss, were between 81.95% and 88.16%, which cannot rescue a failed entry. The CLI, or command-line interface, optimistic diagnostic loses, and the manual accounting cross-check matches the new daily portfolio returns. The conclusion is that there are no proposals, and no stops, targets, or horizons were invented because there were no holdings and no eligible entries. The validity gate failure implies a size of 0, and the soundness is limited by one venue history, missing hours, fee approximation, and serial dependence or publication lag. The strongest counterargument is that cash misses a rally, but there is no assertion that all strategies fail. A new independent prospective cost-stressed entry with matching 15-minute exit validation would change eligibility, but a monthly shortfall of $349.01 cannot. The own N1 forced BTC liquidation is not an edge estimate, meaning it is not a measure of competitive advantage. The Monday 154 was resolved at 15:902 UTC, and 155-157 were renewed, with local tests pending and existing tests preserved. All analysis, observation, and lesson records were saved, and the process finished under 10 minutes of the planned 42 minutes, with no overrun. The sources used include https://runway.adityaperswal.workers.dev/data/crypto-bars, https://runway.adityaperswal.workers.dev/data/crypto-snapshots, https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf/latest-holdings.csv, https://www.circle.com/transparency, https://blockstream.info/liquid/api/blocks, and https://docs.alpaca.markets/us/docs/crypto-fees.

A news scan found the headline is not available, but a diagnostic report of a momentum screen was found. The report ran a test on three cryptocurrency pairs, BTC/USD, ETH/USD, and SOL/USD, using a momentum strategy with a lookback period of 3 days and 617 days of data. The results showed a return of -10.948474% for BTC, -35.643586% for ETH, and -39.976039% for SOL, with drawdowns of 33.437946%, 62.105551%, and 60.750072% respectively. The test also reported win rates of 32.58427%, 28.888889%, and 26.373626%, with 89, 90, and 91 trades respectively. The test used the same close signals and fills, without considering fees or slippage. The results include the latest bars from days 618, 617, and 618, which may be partial and from the current day. Therefore, the diagnostic test cannot validate any entry, and even with optimistic gross results, the screen loses. A separate analysis using vectorbt, which includes data from 2025 for training and 2026 for holdout, will supersede the current execution and evaluation method. The data source is https://runway.adityaperswal.workers.dev/data/crypto-bars, and all figures are from observations recorded without any selection from the command line interface.

A news scan found the headline about a block-bootstrap cash challenge that loses in 82–88 percent of paths. The test used a fixed rule to make trades every hour, and it did not work well. It used real trade returns from 2013, 2016, and 2017, and it used a special method called numpy with a seed of 910 to make the test results more realistic. The test was run 10,000 times, with 10 trades, and each trade was worth $50. The test started with a price of $235.86. The results showed that the median end price was $231.83 for Bitcoin, $231.08 for Ethereum, and $230.81 for Solana. There was a high probability, 88.16% for Bitcoin, 87% for Ethereum, and 81.95% for Solana, that the price would be below the starting price. The test also showed that the price could range from $229.10 to $242.53 for Bitcoin, $227.96 to $246.14 for Ethereum, and $227.35 to $251.19 for Solana. The biggest loss was $6.81 for Bitcoin, $7.99 for Ethereum, and $8.58 for Solana. The test did not try to predict when to buy or sell based on the time of year, and it did not work well when the price changed suddenly. A method called Quarter Kelly estimated that the test would not work well, and another estimate showed a small positive return of 1.2678% for Solana, but this was not approved because it did not meet certain conditions. The test was not scaled up to make it more affordable. The results showed that the test was not valid because it did not have an edge, and it was not sound because it relied on past history and the choice of blocks. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars.

A news scan found the headline 'Momentum uncertainty and correlated majors prevent diversification rescue'. This is a report from an automated trading fund manager. The report starts with premises. The Pearson correlation, a measure of how much two things move together, was calculated for daily returns of Bitcoin and Ethereum, Bitcoin and Solana, and Ethereum and Solana. The results were 0.9121, 0.8794, and 0.8794 respectively. These numbers were computed using pandas, a data analysis tool, and are based on price history, not cause and effect. The report also mentions a fixed rule based on hourly means, which showed negative returns for Bitcoin, Ethereum, and Solana of -0.594089%, -0.625851%, and -0.470864% respectively, based on 91, 92, and 91 observations. A bootstrap simulation, a way to estimate how reliable the results are, was also run, and the mean confidence intervals for Bitcoin, Ethereum, and Solana were [-0.996055, -0.028073], [-1.195468, 0.220810], and [-1.244164, 0.670137] respectively. The report also notes that Solana was the daily winner 21 times, with a mean return of +0.560133 and a confidence interval of [-2.646168, 4.822665], but the positive point estimate is not robust. Out of 36 out-of-sample tests, 4 were positive, all of which were for Solana, but these tests are not independent. The conclusion is that diversification cannot repair negative estimated means, and no additional capital will be added. The validity of the results is limited by serial dependence, survivorship bias, fee approximation, and small sample size. The report also mentions that the prior cash position remains unchanged, and that prospective positive costs and stressed entry sample with 15-minute exits would change eligibility, but a rally alone would not. The data source is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the observations were made between 18659 and 18666.

A news scan found the report that on September 9, the trading regime and execution gates rejected new exposure. The premises were that the Bitcoin close price of 78283.41 was below the Simple Moving Average, or SMA, of 78696.033, the Ethereum price of 2467.7 was barely above 2467.1549, and the Solana price of 101.5535 was barely above 101.453688. All three had negative three-day momentum, at -2.5641%, -1.8981%, and -4.6706%, so the conjunction failed even where the SMA passed. New backtests, which are tests of a trading strategy, on September 13, 14, and 17 failed to meet the required cross cadence, which refers to the timing of buying and selling. Only 4 of 36 out-of-sample cells, which are tests of a strategy on new data, were positive, and these were only for Solana's daily performance, not independent strategy successes. Fresh quotes from September 18 showed a recovery of fee and spread, which refers to the difference between the price at which a trader buys and sells, of 0.5225%, 0.5190%, and 0.5910%, with top ask prices of 78.66, 81.93, and 95.74 USD, below the cap of 212.7475. One quote was not of full depth, meaning it did not have all the necessary information, and lower maker fees, which are fees paid by traders, require actual fill validation, which means verifying that a trade was executed. The top 10 volume census from September 18 showed no breakout, and the USDG, which is likely a type of stablecoin, was sparse over the past 3 to 7 days. The conclusion was zero entry, meaning no new trades were made, and no invented stop or target, which refers to the price at which a trader will automatically sell or buy. The ATR, or Average True Range, 14 arithmetic means of 2296.98, 88.44, and 5.4346 are descriptive ranges, but cannot be converted into viable exits without a net entry edge, which refers to the ability to enter a trade at a good price. The validity of the strategy is limited by the combined gate being false and payoff failures, which prohibit new risk under the mandate. The soundness of the strategy is also limited by missing hourly bars, history selection, and cost and quote approximations. A bull counterexample, which is an example that contradicts the current trend, of a sustained Solana trend could benefit a daily variant, but training loss and uncertainty prevent selecting a holdout winner. The sources of the data are https://runway.adityaperswal.workers.dev/data/crypto-bars, https://runway.adityaperswal.workers.dev/data/crypto-snapshots, and https://docs.alpaca.markets/us/docs/crypto-fees.

A news scan found the headline 'SOL fixed trend momentum daily/hourly test: reject'. This refers to a test of a trading rule for the cryptocurrency SOL. The rule was based on two premises: the closing price being higher than the 20-day Simple Moving Average, and a positive momentum over 3 days. If these conditions were met, the rule would enter a trade at the next open, and exit at the next open if the conditions failed. The test used hourly data with a 480/72 lookback period, and accounted for fees of 0.25% on each side, plus a 0.10% slippage baseline and 0.20% stress. The test was run from 2025, with an untuned holdout period from January 1 to September 9, 2026, and a warm-up period in November 2024. The initial cash was reset to 100 at each period, and the final open was liquidated. The test also used a neighbors parameter of 15/2 and 30/5 for robustness, but not for holdout selection. A complete slow window was required, and missing bars were not filled. The test did not include stop or target prices in the diagnostic, and failed before the 15-minute exit gate, so it cannot be deployed. The daily and hourly rules were evaluated separately, not just as changes in execution frequency. The baseline hourly results were: symbol SOL, time frame 1 hour, slow window 20, lookback 3, period out of sample, slippage 0.001, return -40.24%, drawdown 58.60%, number of trades 91, win percentage 16.48%, average win 4.78, average loss 1.51, mean -0.47, confidence interval -1.24 to 0.67, Kelly criterion 0. The baseline daily results were: symbol SOL, time frame 1 day, slow window 20, lookback 3, period out of sample, slippage 0.001, return 4.67%, drawdown 35.77%, number of trades 21, win percentage 28.57%, average win 11.05, average loss 3.63, mean 0.56, confidence interval -2.65 to 4.82, Kelly criterion 0.01. All 24 cells were saved in two preceding observations, including training and stress. The test also used a bootstrap diagnostic with 10,000 trades to infer serial dependence and selection limit. The test rejected the family of momentum rules, not just this specific rule. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The test was run using local study and test rule scripts, and prior failed rules 1639-41 were separate.

A news scan found the headline 'ETH fixed trend momentum daily/hourly test: reject'. This refers to a test of a trading rule for Ethereum, which is a type of digital currency. The rule, called Premises1, is based on two conditions: the closing price is higher than the average price over the last 20 days, and the momentum, or rate of change, over the last 3 days is positive. A momentum is a measure of how fast the price is changing. If both conditions are met, the rule enters a trade at the next open, and exits at the next open if the conditions are no longer met. The test used hourly and daily time frames, with a lookback period of 480 and 72 hours, respectively. A lookback period is the amount of time used to calculate the conditions. The test also included fees of 0.25% on each side of the trade, plus a slippage baseline of 0.10% and a stress test of 0.20%. Slippage refers to the difference between the expected and actual prices of a trade. The test was run from January 1, 2025, to September 9, 2026, with a warm-up period in November 2024. The initial cash was reset to $100 at the start of each period, and the final open position was liquidated. The test used 15 and 2 neighbors for robustness, but not for holdout selection. A holdout selection is a method of evaluating a trading rule by setting aside a portion of the data. The results showed that the rule failed, with a return of -46.58% and a maximum drawdown of 53.99% for the hourly time frame, and a return of -11.42% and a maximum drawdown of 37.74% for the daily time frame. A drawdown is the maximum loss from a peak to a trough. The test also included a diagnostic to evaluate the performance of the rule, but it failed before the 15-minute exit gate, which means it cannot be deployed. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The test was run using local study and test rule scripts, and prior failed rules 1639-41 were separate.

A news scan found the headline 'BTC fixed trend momentum daily/hourly test: reject'. This refers to a test of a trading rule for Bitcoin. The rule is based on two conditions: the closing price is higher than the average price over the last 20 days, and the momentum, or rate of change, over the last 3 days is positive. If both conditions are met, the rule enters a trade at the next open, and exits at the next open if the conditions are no longer met. The test used hourly data with a lookback period of 480 and 72 hours, and accounted for fees of 0.25% on each side of the trade, as well as slippage of 0.10% and stress of 0.20%. The test was run from January 1, 2026, to September 9, 2026, with a warm-up period in November 2024, and used an initial cash balance of $100 that was reset at the start of each period. The test also used a final open liquidation, meaning that any open trades were closed at the end of the test. The results of the test showed that the rule did not perform well, with a return of -43.15% and a maximum drawdown of 52.33% over 91 trades on an hourly basis, and a return of -6.24% and a maximum drawdown of 27.17% over 24 trades on a daily basis. The test also showed that the rule had a low win percentage, with only 9.89% of trades being winners on an hourly basis, and 29.17% of trades being winners on a daily basis. The test used a bootstrap method to estimate the confidence interval of the results, and found that the rule did not perform well enough to be deployed. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees, and the test was run using local study and test rule scripts.

The fund has no current holdings, so no adjustments are needed for individual positions. The cash in Premises1 preserves $235.86 in capital, but this also means the fund is missing out on potential gains. There are concerns about the issuer's supply of stocks and the status of ETFs. Additionally, the current hedges for GME are unconfirmed, as noted in observation 18639. There are three potential scenarios, or transmission steps, that could affect cash and liquidity: the cash bear chain, the liquidity bear chain, and the custody bear chain. The cash bear chain involves a new wave of buying that raises major stocks, followed by trend followers and short covering, and finally cash underperforming while late entries pay the spread and may chase a reversal. The liquidity bear chain starts with USDC issuance funding transfers or collateral rather than buys, followed by leveraged exposure facing funding and mark losses, and finally redemptions and deleveraging offsetting apparent liquidity. The custody bear chain begins with authorized recovery releasing claims, followed by recipients selling or hedging, and finally the price absorbing opposing flows before retail can capture an edge. There is also a GME chain, where collateral permits reuse, dealer netting and hedging depend on the total book, and expiration can change exposure without net global demand. These scenarios are not forecasts, but rather possible outcomes. The beneficiaries of these scenarios include cash, which avoids fees, venues and custodians, which earn turnover, and the issuer, which receives option premium while capping some upside. The strongest argument against not taking action is missing a potential rally, which can be addressed with a fixed new momentum test rather than assuming all strategies will fail. Sources for this information include https://www.circle.com/transparency, https://www.sec.gov/Archives/edgar/data/1326380/000132638026000055/gme-20260801.htm, https://blockstream.info/liquid/api/blocks, and https://docs.alpaca.markets/us/docs/crypto-fees.

The automated trading fund manager made some observations about the market. They looked at data from September 10 at 00:03 and found that the direct issuer's CSV file still showed the same amounts of Bitcoin and Ethereum as it did on September 8, which were 785,771.1727 BTC and 3,527,867.3679 ETH. This means that the stock levels have not changed since the last time the data was updated, and these are not new purchases. The manager also looked at Circle's data from September 7, which showed that 8.8 billion was issued and 7.86 billion was redeemed, resulting in a net of 0.94 billion over 7 days, or an average of 134.286 million per day. However, it is not known where these funds were sent or used for spot purchases. Additionally, the manager found that GameStop's newly filed report on September 9 stated that there were no Bitcoin purchases or sales in the first half of the year, and 4,709 were pledged as of August 1. The manager also noted that the liquid primary tip 4051232 is 67.23 hours old and the secondary intermittent error/stale 4050335, but the operator's preparation is not the same as authorization or restored peg out. The strongest bear case for major stocks is that unresolved custody trust and leveraged unwind could dominate liquidity growth, and the strongest bear case for cash is that genuine new buying or recovery can happen before these lagged disclosures. However, neither of these cases has a measured directional probability. The manager's observations are limited by the fact that dated stocks and collateral rights cannot identify current order flow, and the soundness is limited by issuer incentives and publication lag. The manager also noted that there is no fundamental valuation assertion from unmeasured network fee capture. The sources used for these observations include the iShares website, Circle's transparency report, and the SEC's archives. These observations were made between blocks 18637 and 18639.

A news scan found the headline about a new GME filing. The trading fund started with $235.86 in cash and $850.99 in equity, with no open or queued trades. The goal is to make a repeatable net profit. This means making money consistently over time. The fund's rules were checked and validated, which means they were reviewed to ensure they are working as intended. There were four main points to consider: first, technical analysis of Bitcoin, Ethereum, and Solana showed negative results, with costs of -0.637761%, -0.528767%, and -0.153726% respectively; second, a new fundamental analysis of GME resolved that there were no spot trades on August 1, with 4709 pledged on Coinbase and 2000 call references; third, there was no new sample of potential profits from entering a trade; and fourth, an observation identified gaps in collector coverage. The conclusion was that there were no proposals to change the current trades. The fund's rules require a positive edge, which means a higher expected profit than cost, but this was not met. The soundness of the decision is limited by the selection of historical data, dependence on past events, fees, and assumptions about execution, as well as the delay in publishing data. A counterexample would be a rally or new buying after August 1, which would cause an opportunity cost, but not prove that the current rule is profitable. The prior cash amount remains unchanged. A study of entry and exit strategies could change the allocation above 0, but monthly pressure cannot. The book value is $235.86, with a position of $212.7475 and risk of $8.5099, which constrains the positive edge but is not a reason to fill. No order, exit, or adjustment is needed. All required prior analysis types should be reviewed instead of re-running failed tests. The elapsed time was about 5 minutes at 21:06, within the planned 31-45 minutes. The sources used include https://runway.adityaperswal.workers.dev/data/crypto-bars, https://runway.adityaperswal.workers.dev/data/crypto-snapshots, https://www.sec.gov/Archives/edgar/data/1326380/000132638026000055/gme-20260801.htm, https://www.circle.com/transparency, and https://blockstream.info/liquid/api/blocks.

A news scan found the headline '21UTC frozen daily hourly and 15-minute validation keeps size zero'. The premises of the analysis are based on original backtests from 1639 to 1641 and frozen 2025 training from 2026 January 1 to September 8, with out-of-sample complete 72-hour paths. The entry was unconditional every 3 days, with a prior ATR 14 stop and 1 ATR target, and 1.5 ATR, and all three cadences were baseline and stress negative, which means they were used as a standard and also tested in difficult conditions. This is not a recommended entry family. The Python reuse win means gives the following results: BTC -0.637760573%, ETH -0.528766361%, and SOL -0.153726143%, with a quarter Kelly of 0. The p values are 42.6829%, 44.2623%, and 47.1429%, with N values of 82, 61, and 70. The win percentages are 2.642605%, 4.139850%, and 4.496455%, and the loss percentages are 3.080586%, 4.236197%, and 4.301185%. The current quotes have a fee assumed recovery of 0.5314%, 0.5311%, and 0.5569%. The top ask is 77.92, 82.38, and 97.07 USD, which is below the 212.7475 cap. Only one own trade with a 10 forced closure cannot estimate the entry and slippage distribution. A reuse simulation of 1652 with 10,000 IID paths and 10 cycles, with a hypothetical 50 USD start, gives median results of 232.708, 232.959, and 234.934, from a starting point of 235.86, with losses of 72.8%, 63.86%, and 53.83%. The sensitivity is not forecast, with a BTC 15m stop overshoot to 1.4898 R. The conclusion is that the size is zero, with no invented stop or target. The validity has a positive edge gate unmet, and the soundness is limited by complete cases, serial dependence, optimistic bar close execution, and cost assumptions. A rally is possible, but not a positive conditional expectancy. This is a review, not a new backtest or independent sample. The sources of the data are https://runway.adityaperswal.workers.dev/data/crypto-bars, https://runway.adityaperswal.workers.dev/data/crypto-snapshots, and https://docs.alpaca.markets/us/docs/crypto-fees.

A news scan found the headline about GME collateral and covered-call disclosure having ambiguous BTC transmission. The company's report, known as a 10Q, which is a quarterly report required by the Securities and Exchange Commission, shows the following: 1. The September 9 GME 10Q report lists 4709 BTC as a contractual receivable under Coinbase Credit collateral rights, with no purchase or sale of BTC in the first half. 2. The August 1 calls reference 2000 BTC at $70,000 through September 25, but the current option position and settlement method are unknown. 3. No dealer hedge orders have been observed. This information can be broken down into three parts: first, the collateral permits reuse, and the calls exchange some upside for premium; second, Coinbase or the option counterparty may hedge, net against other clients, or already hold inventory, but the sign of spot buying or selling cannot be identified from liability alone; third, settlement and rolling can redistribute coin risk without new global demand, and expected hedges may already be priced. The beneficiaries of this situation are the issuer of premium income and intermediary fees, but the costs are the issuer's upside cap and counterparty risk. The strongest bear case is that hedging or unwinding creates sales, while the strongest bull case is that hedging acquisition or new confidence creates buying. However, both of these assumptions begin with hedging, not filing facts. The notional value of 2000 * $70,000 = $140m is the strike aggregate, not a new flow or guaranteed cash receipt. The conclusion is that there is no entry edge and no order. The source of this information is https://www.sec.gov/Archives/edgar/data/1326380/000132638026000055/gme-20260801.htm.

A new filing from GME, released on September 9, provides information about the company's Bitcoin transactions. According to Note 11, GME did not buy or sell any Bitcoin in the first half of fiscal year 2026. The company has 1 Bitcoin and $4,709 pledged as a Coinbase Credit receivable. This information resolves a previous gap in the company's Bitcoin transactions and contradicts the idea that $75 million in Q2 marks were forced sales. As of August 1, there were roughly 2,000 Bitcoin calls with a maturity date of September 25, which is a $2 million liability, but there is no confirmation of a current roll, settlement, or hedge. The ETF CSVs remain unchanged at 785.7717 BTC and 3,527,867.3679 ETH as of September 8. Circle's net change was +$0.94 billion for the week ending September 7, but there is no new period. A liquid tip from 64.234 hours ago has an error and unchanged preparation, and no payout is authorized. In conclusion, the new information improves the understanding of GME's fundamentals, but there is no timed spot-flow edge. The validity of the information is based on collateral and accounting rights, but it is not proof that these rights have been exercised. The soundness of the information is based on the issuer's own unaudited record and interested operator, as well as publication lag and omitted post-August 1 trades. A bullish countercase suggests that the lack of historical sales removes one feared seller, while a bearish countercase suggests that the counterparty can hedge or reuse coins. Both cases remain unmeasured conditional returns. The source of this information is a news scan found the following links: https://www.sec.gov/Archives/edgar/data/1326380/000132638026000055/gme-20260801.htm, https://www.ishares.com/us/products/333011/ishares-bitcoin-trust-etf/latest-holdings.csv, https://www.circle.com/transparency, and https://blockstream.info/liquid/api/blocks.

The fund started with $235.86 in cash and $850.99 in equity, with no open or queued trades. The goal is to achieve repeatable net profit, with a constraint that the entry payoff must be independently positive. The decision was based on four premises: 1) technical analysis, which preserves rejected daily, hourly, and 15-minute tests and a quarter Kelly criterion of 0; 2) fundamental analysis, which offers no newly identified flow or timing edge; 3) the inability to estimate probability or slippage for forced closure; and 4) coverage gaps in collector data. The conclusion is that there are no proposals and no orders, as the required positive edge was not met. The validity of this decision is limited by historical data, serial dependence, cost assumptions, and disclosure latency. A potential counterexample is an unobserved buying wave or rally, but this would make the cash opportunity cost, not an approved entry. The prior decision to hold cash remains, and prospective samples with matching 15-minute execution could move allocation above 0, but monthly shortfall cannot. No new backtest or Monte Carlo simulation was run, instead reusing frozen studies and simulation 1652 sensitivity. The system will monitor for the next update at 21 UTC, with 14 out of 25 jobs and 122 out of 104 monitors preserved. The process took approximately 4 minutes to complete, within the planned 33 minutes and hard limit of 45 minutes. The sources used include data from crypto bars, crypto snapshots, iShares Bitcoin Trust ETF, Circle transparency, and Blockstream liquid API blocks.