A news scan found the text: Numpyseed2026090502; each symbol independently draw three complete 24h log return blocks from 2026 history, 10k 72h paths. XRP/LINK/UNI 245 usable day blocks, PAXG 193; previous close-to-current close hour returns; no unseen intra-hour movement or 15min venue checks. Reference Sep 4 close with 1 ADR 20 stop, 2 ADR 20 target widths respectively 7.609/15.218%, 5.912/11.823%, 7.126/14.251%, 3.685/7.370%. Observed hourly close target before stop frequencies 3.71/5.67/8.51/.60%; stop first 9.53/22.25/24.30/13.93%; neither 86.76/72.08/67.19/85.47%. Unconditional resampling, not actual stop-touch probabilities, signal-conditioned forecast or weekend-specific sample. 72h terminal 5th/95th XRP -7.764/+10.303%, LINK -8.471/+10.095%, UNI -10.435/+13.670%, PAXG -4.236/+4.179%. A measured 2 ADR target can still be unlikely inside 3 days; do not stretch holding window merely to fit. Nominal 1% equity risk 8.5108 yields notional ceilings 111.86/143.97/119.44/212.77, not guaranteed loss limits. Actual size 0 due uncertain net edge. Validity supports caution about attainability; soundness limited by resampling past blocks, separate assets, stationarity and close only paths. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; artifacts/stat_risk.py/stats-risk.json.
A data analysis from the source https://runway.adityaperswal.workers.dev/data/crypto-bars found that from January 1 to September 4, 2026, the net natural-exit samples for the cryptocurrencies XRP, LINK, UNI, and PAXG were 6, 14, 13, and 8, respectively. Using 20,000 independent and identically distributed resamples with a seed of 2026090502, the mean trade 95% intervals were calculated as XRP [-1.194, 18.765]%, LINK [-1.726, 2.184]%, UNI [-2.933, 2.634]%, and PAXG [-1.402, 0.419]%. This analysis assumes that the trades are independent, but the current selection of trades and market conditions may limit the interpretation of the results. The frequencies of positive resamples were 94.375% for XRP, 51.585% for LINK, 38.65% for UNI, and 13.8% for PAXG, but these are not posterior probabilities. A daily ordinary least squares analysis on BTC with an intercept found betas of 1.237, 1.176, 1.181, and 0.299 for XRP, LINK, UNI, and PAXG, respectively, with correlations of 0.871, 0.874, 0.677, and 0.446. The sample size was 247 for each cryptocurrency except PAXG, which had a sample size of 204. The similar betas for XRP and LINK weaken the narrative of diversification. The raw binary mean-payoff quarter Kelly notional formula gave capital fractions of 3.889, 0.281, -0.468, and -7.550, which are clearly unstable and constrained before caps. Since the lower mean bounds are negative, the authorized allocation is 0. The premises of a positive XRP return and a win rate of 66.67% depend on only 6 trades, and the strongest bull case is a large payoff, but no precise population edge has been established. There is a need for prospective fixed rule evidence and tested venue exits, and one should not pick the best lookback from the inspected holdout. The UNI terminal open mark was excluded from payoffs rather than counted as a winner.
A news scan found the note: Registered note125 before results. The signal completed when the daily close was greater than the previous 20 closes, and the desired action was to go long at the next open, while this condition was true, and exit at the next open when this condition was false; there was no stop or target overlay. The baseline included a 25 basis point fee and 5 basis points of slippage on each side. The test used vectorbt with all cash of $10,000, long only, and an independent cash and share ledger, with a maximum error of less than 1e-6. The training period was from September 2024 to December 2025, and the diagnostic chronological holdout was from January 1 to September 4, 2026. The present-day universe selection means this is not independent prospective evidence. PAXG began on February 12, 2026, with no pre-2026 training, and the holdout label was not justified for this symbol. The daily return was -4.103506577337702%, and the drawdown was 5.412997437937084%. An identical daily signal on hourly returns matched with a drawdown of 6.3723479589041006%. A separate 20 observed hour rule had a return of -78.88693847013198%, a drawdown of 78.97980465858033%, a win rate of 14.393939393939394%, and 264 observations. The high turnover cost was rejected, and gaps meant that 20 observed bars were not exactly 20 hours. The double cost daily was -8.597865801961252%. There was no terminal open position. There was no current completed September 4 breakout, and mandatory venue exits and +1R management, and 15-minute fills were untested. XRP had 6 positive trades, which was insufficient for size, and others were unstable or negative. The source was https://runway.adityaperswal.workers.dev/data/crypto-bars, and the artifacts and test.py and backtests.json files, with all metrics and observations saved. The failed initial truncated hourly outputs were discarded, and the current figures are final.
A news scan found the note: Registered note125 before results. The signal completed when the daily close was greater than the previous 20 closes, and the desired action was to go long at the next open as long as the condition was true, and exit at the next open when the condition was false. There was no stop or target overlay. The baseline included a 25 basis point fee and 5 basis points of slippage on each side. The test used vectorbt with all cash of $10,000, long only, and an independent cash and share ledger with a maximum error of less than 1e-6. The training period was from September 2024 to December 2025, and the diagnostic chronological holdout was from January 1 to September 4, 2026. The present-day universe selection means the results are not independent prospective evidence. The training return was -35.39128324423077%. The lookbacks of 15, 20, and 30 were tested only in training, and the fixed 20 was retained. The daily return was 13.0683957825984%, and the drawdown was 11.86325956746298%. The identical daily signal on hourly returns matched with a drawdown of 18.182406721058307%. A separate 20 observed hour rule had a return of -83.3970961790418%, a drawdown of 83.39709617904185%, and a win rate of 18.373493975903614%, with 332 trades. The high turnover and gaps meant the 20 observed bars were not exactly 20 hours, so this rule was rejected. The double cost daily was 4.271874489078182%. The UNI open terminal position contributed to the marked return, with 13 natural exits averaging a negative return. A false signal on September 4 would have exited on September 5 open outside the cutoff, but the exit logic was not missing. There was no current completed September 4 breakout, and mandatory venue exits and +1R management and 15-minute fills were untested. XRP had 6 positive trades, but it was insufficient for size, and others were unstable or negative. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the artifacts and test.py and backtests.json files have all metrics and observations saved. The failed initial truncated hourly outputs were discarded, and the current figures are final.
A news scan found the note: Registered note125 before results. The signal completed when the daily close was greater than the previous 20 closes, and the desired action was to go long at the next open as long as the condition was true, and exit at the next open when the condition was false; there was no stop or target overlay. The baseline included a 25 basis point fee and 5 basis points of slippage on each side. The test used vectorbt with all cash of $10,000, long only, and an independent cash and share ledger with a maximum error of less than 1e-6. The training period was from September 2024 to December 2025, and the diagnostic chronological holdout was from January 1 to September 4, 2026. The present-day universe selection means the results are not independent prospective evidence. The training return was -39.712282707450555%. The daily return was 0.2949758879143155%, and the drawdown was 16.073828141163325%. The identical daily signal on hourly returns matched with a drawdown of 17.771428331495475%. A separate 20 observed hour rule had a return of -82.69256923103462%, a drawdown of 83.21785242055526%, a win rate of 17.159763313609467%, and 338 trades. The high turnover was rejected due to cost, and gaps meant the 20 observed bars were not exactly 20 hours. The double cost daily was -7.78575922941096%. There was no terminal open position. There was no current completed September 4 breakout. Mandatory venue exits and +1R management, and 15-minute fills were untested. XRP had 6 positive trades but was insufficient in size, and others were unstable or negative. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the code and results are saved in artifacts/test.py and backtests.json. The failed initial truncated hourly outputs were discarded, and the current figures are final.
A news scan found the note: Registered note125 before results. A signal was completed when the daily close was greater than the previous 20 closes, which meant a long position was desired at the next open, and the position would be exited at the next open when this was no longer true. There was no stop or target overlay. The baseline included a 25 basis point fee and 5 basis points of slippage on each side. The backtest was done using vectorbt with $10,000 in cash, and only long positions were considered. The cash and share ledger were independent, and the maximum error was less than 1e-6. The training period was from September 2024 to December 2025, and the diagnostic chronological holdout period was from January 1, 2026, to September 4, 2026. The training return was 257.08276701988984%. The daily return was 49.30134045526893%, and the drawdown was 10.42241544981568%. The same daily signal was applied to hourly returns, which resulted in a match with a drawdown of 13.486776844598625%. A separate 20-observed-hour rule was tested, which resulted in a return of -71.61556289760512%, a drawdown of 75.69734866102567%, and a win rate of 16.206896551724135% over 290 observations. This rule was rejected due to costly turnover, and gaps meant that 20 observed bars were not exactly 20 hours. The double cost of daily trades was 44.0220253594203%. There was no terminal open position. There was no current completed breakout on September 4, and mandatory venue exits and +1R management with 15-minute fills were untested. XRP had 6 positive trades, but this was insufficient for size, and other trades were unstable or negative. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, and the code and backtest results are saved in artifacts/test.py and backtests.json. The initial truncated hourly outputs that failed were discarded, and the current figures are final.
A news scan found the analysis: the calendar-defined USDG volume is $2.103M with missing days. This information supersedes the previous USDG volume and continuity statements and observation 273 windows. The initial selection of the last 7 observations, from August 26 to September 4, overstated the seven-calendar-day amount as $4.103M. The explicit data from August 29 to September 4 gives $2.103309M across 4 available days, with three missing days unknown, and the reported volume on September 4 was 0. USDG remains second by observed USD notional, but a complete ranking cannot be certified without missing-day reconciliation. The last 30 calendar window has 10 rows and 3 zero volume rows, not six zero-volume days, which were previously stated as 30 observations. The prior 20 close window also spans more than 20 days, so no valid 20-day signal is claimed. Other 35 names have 7 recent daily rows, and the names XRP, LINK, UNI, and PAXG were tested and found to have complete latest 21 rows. A strong counterargument is that no-trade days may simply be omitted, but this should not be assumed without data semantics. The corrected collector will explicitly label day counts, missing coverage, and eligible 0. The primary source is https://runway.adityaperswal.workers.dev/data/crypto-bars. This audit changes the data-quality assessment, not the zero-order decision. Calendar filtering is needed before rolling window interpretation on sparse assets.
The manager separates premises and hypotheses from facts. Facts include observations 317-319, 331, 336-337, which show changes in issuer fees, sponsorship, USDC issuance, LINK customer integration, and historical UNI burns. For example, 1 USDC and 0.61 billion in 7 days implies a 0.827% increase from the prior 73.72 billion, and 30 days with 2.25 billion implies a 3.122% increase from 72.08 billion. The supply expansion increases settlement capacity, and users may use tokens for payments or collateral instead of major tokens, but without destination data, higher stock does not identify marginal token buyers. The XRP logo raises awareness, and adoption may follow, but token returns still require buyer flow, not advertising spend. Circle uses Chainlink reserves, and service usage may increase oracle fees, but the LINK price benefit depends on paid token demand and retention, which is not present here. UNI's June fee of 5.1 million annualizes to 61.2 million if constant, but burns can reduce units, and LP migration, governance changes, or treasury releases may offset the marginal benefit. There is no September burn flow or current net flow that has been audited, so there is no fair value multiple. PAXG's primary redemption frictions change arbitrage economics, and the weekend secondary premium may persist, but the cash book cannot assume immediate physical delivery or fee-free conversion. The strong bull case is confirmed by infrastructure and burn mechanisms, but it is still not proof that the historical net price edge survives. The beneficiaries include sponsored universities, oracle providers, users, LPs, searchers, and custodians, according to the mechanism, but token holder benefit is conditional. The manager starts flat, with the goal of repeatable net gains, and a binding constraint of independent executable payoffs. Sources include Circle's transparency report, a blog post on the reserve standard for wrapped BTC, a Uniswap governance post, a PAXOS support article, and a Florida Athletics URL observation. The conclusion is to retain cash until tests and current entry qualify.
The audit of primary demand found that the issuance of USDC is positive, with a reported 74.33 billion USDC in circulation, a 7-day net increase of 0.61 billion, and a 30-day increase of 2.25 billion, according to Circle's September 3 report. The issuer's page is the source of this information, but it does not prove buying on exchanges. On September 4, IBIT reported 1,384,960,000 shares, establishing a baseline, but there is no prior count to measure creation delta, and the page mixes September 3 price and volume with September 4 NAV and AUM, so AUM change should not be labeled as flows. Florida Athletics confirmed a multi-year Ripple field logo sponsorship on September 4, but there is no mandatory XRP purchase or token holder cash flow quantified. The PAXG issuer's new redemption bands affect primary conversion, not Alpaca spot trade fees, and the creation waiver ends on September 30. The conclusion is that demand narratives remain uncalibrated. The validity of the audit is limited to issuer reports, rounded USDC values, missing destination flows, and ETF history. A strong bull case could be made if more settlement capital and visibility precede adoption, but actual token demand and independent net payoffs are needed. The source URLs are provided. The starting cash is 235.86, with a goal of repeatable net gain and a binding constraint of positive executable edge, but there is no order.
A news scan found the headline about an expanded 36-pair census. The Alpaca trading platform published a list of 36 USD pairs on May 19, which is the primary confirmation of added assets. Daily bars from September 2024 to September 4 show that the top ten assets by 7-day sum of volume times volume-weighted price are USDC, USDG, BTC, ETH, SOL, XRP, PAXG, LINK, UNI, and USDT. USDG was missing from the prior 33-name sample and replaced AAVE. On September 4, USDG had a traded volume of 0, despite having a 7-day volume of 4.103M and six zero-volume days in the last 30 days, which means that activity concentration is not always available liquidity. Only USDT closed above its prior 20 closes, with prior anomalous pricing, and there is no non-stable signal. The conclusion is that there is no entry based on the screen, and the current top 10 is based on the published list, 7-day, and USD-only, not verified account assets inventory or cross-quote consolidated ranking. The API gateway for data assets is not available, and there are no account credentials available, so it is not claimed that the list is exhaustive or live tradability. The core verdict for BTC, ETH, and SOL remains unchanged at 66. BONK and LDO are current movers in headlines, but they do not satisfy the seven-day top 10 or validated signal. The current universe selection introduces survivorship bias in historical tests. The sources for this information are the Alpaca markets support page and blog, as well as a crypto bars data page. Observations 272-307 retain every screened row, including rejected names. The top candidates for new audits are XRP, PAXG, LINK, and UNI, which are registered in note 125.
A news scan found the headline 'September 5 verdict: reject fee-paying BTC/ETH and hourly20; SOL daily WATCH, retain cash'. The current status of the automated trading fund is that it has rejected the implementation of fee-paying BTC/ETH and hourly20, and it will only watch SOL daily, without adopting it. The fund's diagnostic test from January 1 to September 4 showed returns of -7.44% for BTC, -15.57% for ETH, and +22.23% for SOL. The 20-hour losses were -83.70% for BTC, -80.99% for ETH, and -87.42% for SOL. The daily signal replication matches the returns and payoffs, with hourly mark-to-market losses of 16.52% for BTC, 25.33% for ETH, and 15.01% for SOL. The fund's cash ledger confirms this. The SOLN12 win50 meanW6.123 meanL2.270, IIDmean95%CI[-1.213,6.606]%, but this is not independent proof due to selection bias. The fund completed signals on September 4 for BTC, ETH, SOL, and UNI, but all of them were false. The cash ledger has no open or pending positions, and there are no stop or target levels to reset. The fund has not validated any fee-inclusive measured-range or 15-minute venue exit, and there is no independent untouched holdout. Therefore, the fund will not propose any entry, exit, or adjustment, and it will allocate zero notional value. This decision is valid under the fund's mandate, but its soundness is limited by sparse volume, missing hours, and selected historical sample, as well as assumed slippage. Rejecting these implementations does not prove that all momentum is false. The raw SOL quarter Kelly is 7.866%, but it is not authorized due to conditioning and selection uncertainty, and missing exits. The nominal 1% equity is $8.5108, the position cap is 212.77, and the fund is 235.86. The goal is to achieve repeated net gains, with a binding constraint of independent implementable edge. The monthly shortfall of $348.92 does not change the criteria. A prospective frozen window will start on September 5, and the earliest review will be on October 5, requiring at least 30 closed trades, with no retuning or early adoption. The range and simulation results illustrate risk, but do not guarantee fills. The sources of the data are https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The primary catalyst analysis is 52. The source claims and test metrics are saved observations, and the artifacts are local. Watch11 remains research-only, and phrase validation is pending. A new funding/OI job will be run every 120 minutes, and the first scheduled sandbox output must be checked next run. There will be an ETH monitor on September 28, but there is no exact mainnet date, and no invented SOL date. There will be no action in other funds. The gaps in the research include the complete tradable-volume universe, ETF/exchange/stablecoin flow changes, full unlock inventory, independent edge, and 15-minute fill stress. The research was finished early, and no thread exceeded the allowance, with the final recording phase preserved.
A research watch is in place for low funding rates on the OKX exchange for Bitcoin, Ethereum, and Solana. Data from the exchange's API on September 5 at 00:04 UTC showed funding rates for BTC, ETH, and SOL USDT swaps. The current funding forecasts are 0.01034, 0.58253, and -0.84163 basis points for 08 UTC, compared to the 100 realized 8-hour settlements from August 3 to September 5. The empirical percentiles are 4, 46, and 2. The last settled funding rates were 0.00702, 0.57593, and -0.85247 basis points. The open interest is $2.192 billion, $1.478 billion, and $287.585 million, which is a single-venue baseline. If the forecast stays constant, the simple annualized carry would be +0.113%, +6.379%, and -9.216%, and the 72-hour total would be +0.000931%, +0.052428%, and -0.075747%. This is based on assumptions, not predictions, and the spot market receives none of this funding. Three premises are considered: low funding weakens the assertion of a crowded long position, it does not prove a short squeeze, and daily price breakouts are false for all three. The conclusion is that there is no conjunction producing a tested buy signal. The strongest bear interpretation is that negative SOL funding may reflect persistent adverse expectations, while the strongest bull interpretation is that shorts may later cover. Neither interpretation identifies an expected spot win or loss. A scheduled job will run every 120 minutes to store timestamped forecast and settled funding data, as well as open interest, to build changes. This data is from OKX, an independent source, and does not include derivatives orders within a cash-only mandate.
The goal is to achieve repeatable net returns. The current state has no existing stops or targets to change. The strongest bear case for Bitcoin (BTC) is that it failed to close a breakout and has unverified new Exchange-Traded Fund (ETF) creations, which is a type of investment fund that trades on a stock exchange. There are three reasons to be cautious: chasing yesterday's signal incurs costs from both sides, reversal consumes an uncalibrated stop budget, and replacement risk with Ethereum (ETH) and Solana (SOL) compounds losses. However, some argue that ETF demand and momentum could revive, but this is not a calibrated current edge. The bear case for ETH is that upgrade repricing changes may disrupt some contracts. This could lead to three issues: builders must alter gas assumptions, implementation delays or higher state-operation costs can slow activity, and token returns depend on adoption, issuance, burn, and valuation, not just throughput. However, most contracts may be unaffected, and fixes and capacity expansion may improve adoption. The bear case for SOL is that there is a conflict between the Q3 date and the October date, and the current close is below $109.15. This could lead to three issues: anticipated launch demand may precede actual activation, delay or rollout risk can reverse positioning, and limited price-rule calibration and a high beta amplify sizing error. However, a successful 150ms-finality target could improve applications. The beneficiaries of a successful capacity increase are users, builders, liquidity providers, and exchange perpetual funding paid between leveraged parties. Missing a rally is an opportunity cost, but a funding deadline changes incentives, not conditional expectancy. Current evidence supports a zero-risk approach, not the assertion that major assets must fall. Facts are based on recorded prices, tests, and issuer schedules, while economic chains are hypotheses. Sources include crypto market data and information on Bitcoin, Ethereum, and Solana.
A news scan found the source https://runway.adityaperswal.workers.dev/data/crypto-bars, which is hourly data from January 1 to September 4, 2026. This data was used to run a simulation, which is a computer program that tries to predict what might happen in the future. The simulation used something called NumPy, which is a tool that helps with math, and it was set up on September 5, 2026. The simulation drew 3 complete 24-hour blocks of data, which is called log-return, from 246 days of data, and it did this 10,000 times for each symbol, which is a code for a type of investment, such as BTC, ETH, and SOL. The simulation looked at what might happen over a 72-hour period, which is 3 days. The simulation did not try to predict what would happen based on the day of the week, such as the weekend, and it did not try to predict what would happen based on any specific strategy. Instead, it just looked at what might happen based on the data. The simulation also ignored what happens within an hour and the delay between different places where investments are made. The simulation used something called a stop, which is a rule that says when to sell an investment, and a target, which is a goal for how much the investment should increase. For example, the stop might be 3.392% for BTC and the target might be 5.088%. The simulation found that the probability of hitting the stop first, the target first, or neither was: 32.06% and 15.56% and 52.38% for BTC, 32.51% and 16.08% and 51.41% for ETH, and 17.20% and 6.38% and 76.42% for SOL. The simulation also found that the worst and best cases for each investment were: -6.184% and +7.006% for BTC, -8.939% and +10.032% for ETH, and -9.011% and +9.617% for SOL. The simulation also found that the worst losses for each investment were: 7.508% for BTC, 10.722% for ETH, and 10.492% for SOL. The simulation also looked at what happened in the real world over 35 weekend periods, from Saturday at 00:00 UTC to Monday at 23:00 UTC, and it found that the worst losses were: -7.051% for BTC, -6.750% for ETH, and -9.198% for SOL. The worst losses that actually happened were: -11.414% for BTC, -20.240% for ETH, and -18.188% for SOL. The simulation also found that the average gains were: 2.644% for BTC, 3.741% for ETH, and 3.510% for SOL, which is less than the targets. The simulation also found that the maximum amount of money that could be lost, which is called the hard cap, was $8.5108, and that the maximum amount of money that could be invested, which is called the notional ceiling, was: 212.77 for BTC, 196.25 for ETH, and 121.83 for SOL. However, the actual amount of money that was invested was 0. The simulation found that the stops and targets that were used in the simulation might not work in the real world, and that there is no guarantee that the losses will be limited to 1% or that the targets will be reached. The simulation also found that there is no evidence that the investments will go back up to their original value, which is called mean-reversion, or that it is a good idea to invest on the weekend. The data from the simulation is available in files called simulation.json, weekend.json, risk_analysis.py, and weekend.py.
A news scan found the headline about a closing breakout failure on September 4. The primary venue bars were retrieved on September 5 and completed on September 4, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars, with indicative snapshots at /data/crypto-snapshots. The prices of Bitcoin, Ethereum, Solana, and Uniswap were compared to their prior 20-day close maximums: Bitcoin was at 79676.7475 versus 81269.4915, Ethereum was at 2456.376 versus 2515.933, Solana was at 101.95 versus 109.15, and Uniswap was at 6.18484 versus 6.32576. All of these were false, meaning they did not meet the expected levels. The alerts for Bitcoin and Uniswap from September 3 expired when the September 4 close failed. The daily returns were -1.960% for Bitcoin, -2.067% for Ethereum, -1.941% for Solana, and -2.228% for Uniswap. The core remains above the 20-day Simple Moving Average, which is a calculation of the average price over 20 days, but that alone is not enough to register an entry. The Average True Range, which is a measure of volatility, for 14 days was 2702.40 for Bitcoin, 106.53 for Ethereum, and 7.122 for Solana. The Average Daily Range, which is a measure of the average price range over 20 days, was 2921.85 for Bitcoin, 118.75 for Ethereum, and 6.530 for Solana. The illustrative stop and target percentages were 3.392% and 5.088% for Bitcoin, 4.337% and 6.505% for Ethereum, and 6.986% and 10.479% for Solana, but these are not order levels or validated exit rules. The 1.5 Average True Range is inside the 2 Average Daily Range, but there is no tested conditional win rate. Thirty-three observed USD candidates were ranked by their 7-day sum of volume times value, and the top 10 were USDC, Bitcoin, Ethereum, Solana, XRP, PAXG, LINK, Uniswap, USDT, and AAVE. The full tradable census is unavailable, and the user should not mislabel the global or top 10 Alpaca coverage. The only true signal was USDT, but its anomalous stale-looking prior prices are not a validated stablecoin strategy. All other 32 signals were false. The user book has no holding to reset. There is a strong bear case due to the latest daily close failing to break out and no delayed-entry rule being validated. However, a counterargument is that the 20-day returns remain strong, and trend positivity alone cannot override entry definition or costs. The proposed size is 0.
A news scan found the data from a crypto market analysis. The data, available at https://runway.adityaperswal.workers.dev/data/crypto-bars, covers daily returns from January 1 to September 4, 2026. It shows that the correlation between Bitcoin (BTC) and Ethereum (ETH) is 0.9124, between BTC and Solana (SOL) is 0.8805, and between ETH and SOL is 0.8801. This means that these currencies tend to move together in the market. The beta values, which measure volatility, are 1.254 for ETH and 1.202 for SOL, based on ordinary least squares (OLS) regression, a statistical method. The daily standard deviation, a measure of risk, is 2.452% for BTC, 3.370% for ETH, and 3.347% for SOL. The worst close-to-close return, which is the change in price from one day to the next, was -13.986% for BTC, -14.890% for ETH, and -14.931% for SOL. Two premises are considered: first, that the three major long positions are strongly correlated and represent a risk, rather than independent investments; and second, that the registered net daily closed-trade bootstrap, a statistical method, shows mean 95% intervals of -1.424% to 1.323% for BTC, -2.670% to 0.678% for ETH, and -1.213% to 6.606% for SOL. For SOL, there were 12 trades, with a 50% win rate, a mean win of 6.123%, and a mean loss of 2.270%. The naive quarter-Kelly, a betting strategy, is 0.07866, but the authorized size is 0 because the interval crosses 0, and there is no prospective validation or tested venue exits. The bootstrap-positive fraction of 0.7897 indicates resampling stability, but not the probability that the strategy is profitable. The assumption of independent and identically distributed (IID) data ignores serial dependence, which may lead to understated uncertainty. All hourly net mean intervals were entirely negative, ranging from -0.611% to -0.443% for BTC, -0.609% to -0.346% for ETH, and -0.682% to -0.465% for SOL. Despite this, SOL may have a real positive momentum edge, but there is insufficient confidence to prove a negative return. The goal is to achieve repeatable net gains, with the constraint of independent payoff validation. The methods and results are available in the return-statistics.json, bootstrap.json, and risk_analysis.py files, but no predictive regression is claimed.
A news scan found the results of a command line interface control: Ran mandated CLI node /app/agent/dist/backtest.js --symbols BTC/USD,ETH/USD,SOL/USD --strategy breakout --lookback20 --days730. Results were: for BTC/USD, 731 bars, 26.56% return, 13.54% drawdown, 34% win rate, 50 samples, and 41.82% buy and hold return; for ETH/USD, 730 bars, 28.04% return, 24.94% drawdown, 21.05% win rate, 38 samples, and 3.73% buy and hold return; for SOL/USD, 731 bars, 75.21% return, 18.80% drawdown, 52.94% win rate, 34 samples, and -21.21% buy and hold return. This control includes current-day partial data because the command line interface omits an end cutoff, has varying 731/730 bars, uses same signal-close fills, no fees or slippage, no independent chronological holdout, and no intrabar-mark drawdown. Hence, it is not directly the same window as the registered experiment and not positive sizing evidence. The full command line interface JSON is saved as artifacts/cli.json. The source is /app/agent/dist/backtest/{bars,strategies,engine}.js and https://runway.adityaperswal.workers.dev/data/crypto-bars. This fresh control demonstrates why published positive exploratory returns can coexist with rejected next-open net holdouts. The registered vectorbt tests, not this control, govern the current verdict.
A trading signal that was previously made available at the start of each day, known as a DAILY prior20-close signal, was tested to see how it would perform if it were used at the start of every hour instead. This is not a 20-hour strategy, but rather the same daily signal used every hour. The test used a fee of 25 basis points, which is a cost of 0.25%, and 5 basis points of slippage, which is the difference between the expected price and the actual price, on both the buy and sell sides. The test was run from January 1, 2026, to September 4, 2026, and used data from September 2024 as a warm-up period. The results of the test showed that the returns were the same as the daily implementation, but the drawdown, which is the largest decrease in value, was greater. The results were as follows: for a fee of 25 basis points and slippage of 5 basis points, the return was 22.23% and the drawdown was 15.01%, and for a fee of 50 basis points and slippage of 10 basis points, the return was 13.74% and the drawdown was 18.99%. A news scan found the source of the data to be https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The test was run using the files extended_audit.py, artifacts/daily-on-hourly.json, and ledger-verification.json, which are located in the /tmp/codex-run-fYGHo5 directory. The costs were estimated using the asset-denominated fee as cash, and there was no executable stop or target, or 15-minute exit delay model. One shared hourly holdout bar was missing and was retained as a gap. The positive result for SOL remains selected, but it is uncertain and there is no entry authorization.
A news scan found the results of an automated trading fund manager. The manager used a daily signal, made available at the start of each day, and filled at the open of each hour. This is not a 20-hour strategy. The manager used a fee of 25 basis points and 5 basis points of slippage on each side, and tested the strategy from January 1, 2026, to September 4, 2026. The results show that the returns are the same as the daily implementation, but the drawdowns are greater. A drawdown is the difference between the highest and lowest values of an investment. The results are as follows: the symbol is ETH/USD, the variant is daily20 executed hourly, the fee is 25 basis points, the slippage is 5 basis points, the return is -15.569340386315634, the drawdown is 25.32511446164405, the sample size is 14, the win rate is 21.428571428571427, the average win is 4.470243968974899, and the average loss is 2.6860993280793997. Another test with a fee of 50 basis points and 10 basis points of slippage showed a return of -22.371892459130414 and a drawdown of 30.928453319792336. The cash and quantity ledger matches the baseline to numerical precision. The manager used data from https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The files extended_audit.py, artifacts/daily-on-hourly.json, and ledger-verification.json are under /tmp/codex-run-fYGHo5. The costs are approximate and asset-denominated, and there is no executable stop or target, or 15-minute exit delay model. One shared hourly holdout bar is missing and retained as a gap. A positive result from SOL remains selected, but it is uncertain and there is no entry authorization.
A news scan found the report that the daily20 signal for BTC/USD, which is a signal that is made available at the start of each day and held constant through that day, was replicated on an hourly basis, meaning it was filled at the open of each hour. This is not a 20-hour strategy, but rather the same daily signal being executed at each hour. The results were calculated using a fee of 25 basis points, which is a measure of the cost of a transaction, and 5 basis points of slippage, which is the difference between the expected price of a trade and the actual price. The diagnostic holdout period, which is a period of time used to test the strategy, was from January 1, 2026, to September 4, 2026, and the warmup period, which is a period of time used to train the strategy, was from September 2024. The results show that the returns and payoffs from this hourly strategy match the returns and payoffs from the daily strategy, but the hourly strategy has deeper drawdowns, which is a measure of the largest peak to trough decline in value. The results are as follows: for a fee of 25 basis points and slippage of 5 basis points, the return was -7.44%, the drawdown was 16.52%, the win rate was 27.78%, the average win was 3.12, and the average loss was 1.56. For a fee of 50 basis points and slippage of 10 basis points, the return was -17.17%, the drawdown was 22.85%, the win rate was 27.78%, the average win was 2.51, and the average loss was 2.16. The cash and quantity ledger, which is a record of all transactions, matches the baseline to numerical precision. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees. The files extended_audit.py, artifacts/daily-on-hourly.json, and ledger-verification.json are located under /tmp/codex-run-fYGHo5. The costs are approximate and are denominated in the asset, meaning they are based on the value of the asset being traded. There is no executable stop or target, which is a price level at which a trade is automatically stopped or reversed, and there is no 15-minute exit delay model. One shared hourly holdout bar is missing and is retained as a gap. The positive SOL result remains selected, but it is uncertain and there is no entry authorization.
The automated trading fund manager has conducted a primary catalyst audit. The audit is based on four premises. Premise 1 looks at the iShares Bitcoin Trust ETF, which reported 1,375,120,000 shares and a net asset value of $46.13 on September 3. However, the number of creations on September 4 was not available. Premise 2 examines the Ethereum roadmap, which indicates that the Sepolia upgrade will occur on September 28, while the mainnet upgrade is expected in Q4, but the exact date is not confirmed. The Ethereum blog also mentions contract compatibility risks under EIP8037/8038, but notes that most contracts are unaffected and many issues can be fixed with gas-limit changes. Premise 3 looks at the Solana upgrades, which include Agave 4.3 and Alpenglow in October, although the exact mainnet date is not confirmed. Premise 4 notes that the Circle website did not render numerical fields, making it impossible to determine stablecoin supply growth. The conclusion is that there is no issuer-supported dated-price catalyst to justify a new position. The validity of the audit is based on the fact that network capacity improvements do not necessarily ensure token appreciation. The soundness of the audit is limited by the fact that sources closest to implementation have a commercial or ecosystem interest, and missing data on token fees, burn, issuance, and valuation prevent a fair-value calculation. A strong bear case is that fees per unit can fall faster than usage grows, and technical rollout can be delayed. A strong counterargument is that successful upgrades can improve demand and reduce costs, but expected benefits may already be priced in. The audit is a qualitative fundamental analysis, not a token valuation model. The ETH monitor will review the situation on September 28 at 00UTC, and success will require a subsequent tested response.
A preregistered note, called note92, was audited. It checked if the closing price was higher than the highest closing price of the previous 20 hours. The rule was to hold the position only while this signal was true, and the next observed bar's open price would fill the order. The 20 hours is a separate time frame, not a repetition of 20 days. The data used was from September 1, 2024, to September 4, 2026, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars. The development period was before January 1, 2026, and the diagnostic test period was from January 1 to September 4, 2026, which is 247 days. However, the test period overlaps with the history that was previously inspected, so it is not a genuinely independent test. The prospective window in note92 remains untouched. The test used vectorbt version 1.1, with all cash per symbol, an initial amount of 10000, and no leverage. The fees were 25 basis points plus slippage of 5 basis points per side, with a stress test of 50 plus 10. The official fee can be found at https://docs.alpaca.markets/us/docs/crypto-fees. The slippage is assumed, not measured. The training-only sweep was done with L15, L20, and L30, and L20 was fixed before the results. There was no stop or target overlay, and it was not an executable venue exit rule. The baseline was independently reproduced by a cash and quantity ledger to less than 4e-13 percentage points. The return includes mark-to-market open trades, and payoff metrics were only closed. There was one missing hourly data point on March 12, and zero-volume bars mean that liquidity was not demonstrated. The raw scripts and hashes can be found in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results in percentages are as follows: development with L15 had a return of -99.65586598112374, development with L20 had a return of -99.32688214549539, development with L30 had a return of -98.55344598044546, holdout with L20 and zero fees had a return of 5.821752539654608, holdout with L20 and 25 basis points fees had a return of -87.42447617299416, and holdout with L20 and 50 basis points fees had a return of -98.50559777346359. The conclusion is to reject this implementation because a positive return without fees does not mean there is an edge when paying fees. This does not make any inference about all momentum strategies.
A news scan found the headline 'SOL/USD daily close-breakout20: next-open cost-adjusted diagnostic holdout'. This is a note from an automated trading fund manager. The manager preregistered a note, which is like a plan, called note92. The plan is to hold a position only while a certain signal is true. The signal is when the closing price is greater than the maximum of the previous 20 closing prices. The manager uses a 20-day period to decide when to enter or exit a trade. The data used is from September 1, 2024, to September 4, 2026, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars. The manager developed the plan before January 1, 2026, and tested it from January 1 to September 4, 2026, which is 247 days. This test period overlaps with the history that was previously inspected, so it is not a genuinely independent test. The manager used a tool called Vectorbt, version 1.1, to simulate trades with an initial amount of $10,000 and no leverage, which means not using borrowed money. The manager also assumed fees of 25 basis points, or 0.25%, and slippage of 5 basis points, or 0.05%, per trade, which means the difference between the expected price and the actual price. The official fee can be found at https://docs.alpaca.markets/us/docs/crypto-fees. The manager did not measure the slippage, but assumed it. The manager tested different parameters, called L15, L20, and L30, and fixed L20 before looking at the results. The manager did not use a stop or target overlay, which means not setting a specific price to stop trading or to take profit. The results were compared to a baseline, which was reproduced independently by a cash and quantity ledger to a very small percentage point, less than 4e-13. The return includes the value of open trades, and the payoff metrics only include closed trades. There was some missing hourly data, and zero-volume bars mean that liquidity, or the ability to buy or sell, was not demonstrated. The raw scripts and hashes can be found in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results are as follows: during the development phase, the manager tested L15, L20, and L30, and got returns of -8.699%, 8.193%, and 22.058%, respectively. During the holdout phase, the manager tested L20 with different fees and slippage, and got returns of 31.351%, 22.226%, and 13.735%. The conclusion is that the results are positive, but the small number of trades, selection, and uncertain exits mean that the manager will keep watching and not trade with a large size. This does not mean anything about all momentum strategies, which are a type of trading strategy that involves buying assets that are going up in price and selling assets that are going down in price.
A preregistered note, called note92, was audited. It checked if the closing price was higher than the highest closing price of the previous 20 hours. The system would hold a position only while this condition was true, and the next bar's opening price would be used to fill the position. The 20 hours referred to here is a separate time frame, not the same as 20 days. The data used was from September 1, 2024, to September 4, 2026, and can be found at https://runway.adityaperswal.workers.dev/data/crypto-bars. The development phase was before January 1, 2026, and the diagnostic phase, which is used to test the system, was from January 1 to September 4, 2026, lasting 247 days. However, the diagnostic phase is not completely independent because it overlaps with data that was previously inspected. The prospective window, which is the period of time used to predict future outcomes, remains untouched in note92. The system was tested using vectorbt version 1.1, with all cash per symbol, an initial amount of $10,000, and no leverage. Fees were set at 25 basis points, or 0.25%, plus slippage of 5 basis points, or 0.05%, per side. The official fee can be found at https://docs.alpaca.markets/us/docs/crypto-fees. Slippage, which is the difference between the expected price and the actual price, is assumed, not measured. The system was trained using a sweep of parameters, specifically L15, L20, and L30, and L20 was fixed before the results were obtained. There was no stop or target overlay, meaning there were no specific rules to exit a position. The baseline, which is the base case used for comparison, was independently reproduced using a cash and quantity ledger to a precision of less than 4e-13 percentage points. The return includes the mark-to-market value of open trades, and payoff metrics are only calculated for closed trades. There was one missing hourly data point, on March 12, which was not filled. Zero-volume bars mean that liquidity, or the ability to buy or sell, was not demonstrated. The raw scripts and hashes can be found in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results, in percentage, are as follows: for the development phase with L15, the return was -98.98%, the drawdown, or the maximum loss, was 98.98%, and the win rate was 17.11%. For L20, the return was -97.33%, the drawdown was 97.34%, and the win rate was 18.10%. For L30, the return was -92.93%, the drawdown was 93.03%, and the win rate was 19.44%. For the holdout phase with L20 and zero fees, the return was 43.62%, the drawdown was 8.98%, and the win rate was 33.83%. With fees of 25 basis points and slippage of 5 basis points, the return was -80.99%, the drawdown was 81.06%, and the win rate was 13.95%. With fees of 50 basis points and slippage of 10 basis points, the return was -97.48%, the drawdown was 97.48%, and the win rate was 7.72%. The conclusion is to reject this implementation because a positive return without fees does not mean there is an edge, or an advantage, when fees are included. This does not imply anything about all momentum strategies.
A preregistered note, called note92, was audited. It checked if the closing price was higher than the highest closing price of the previous 20 days. The rule was to hold the position only while this signal was true, and the next bar's open would determine the entry or exit. This was tested over 20 days, from September 1, 2024, to September 4, 2026, using data from https://runway.adityaperswal.workers.dev/data/crypto-bars. The data was split into two periods: development, which was before January 1, 2026, and a diagnostic holdout period from January 1 to September 4, 2026, which was 247 days long. However, the holdout period overlaps with previously inspected history, so it is not genuinely independent. The prospective window in note92 was not touched. The test used vectorbt version 1.1, with all cash per symbol, an initial amount of $10,000, and no leverage. Fees were 25 basis points plus slippage of 5 basis points per side, with a stress test of 50 plus 10. The official fee is listed at https://docs.alpaca.markets/us/docs/crypto-fees, and slippage was assumed, not measured. The test used a training-only sweep of L15, L20, and L30, with L20 fixed before results. There was no stop or target overlay, and this was not an executable venue exit rule. The baseline was independently reproduced by a cash and quantity ledger to less than 4e-13 percentage points. The return included mark-to-market open trades, and payoff metrics were only for closed trades. There was missing hourly data that was not filled, and zero-volume bars meant liquidity was not demonstrated. The results, in percentage, were: development with L15 had a return of 5.0745%, development with L20 had a return of 13.4134%, and development with L30 had a return of 24.2924%. The holdout period with L20 and no fees had a return of -8.1708%, with fees of 25 basis points it had a return of -15.5693%, and with fees of 50 basis points it had a return of -22.3719%. The conclusion is to reject this implementation because a positive return without fees does not mean there is an edge when paying fees. This does not make any inference about all momentum strategies.