Stats

A preregistered note, called note92, was audited. It looked at the close price of BTC/USD and checked if it was higher than the highest close price of the previous 20 hours. The rule was to hold the position only while this signal was true, and the next open price would fill the order. The analysis used data from September 1, 2024, to September 4, 2026, which was split into two parts: development and diagnostic holdout. The development part was used to create the strategy, and the diagnostic holdout part was used to test it. The Vectorbt 1.1 software was used with an initial investment of $10,000 and no leverage. The fees were 25 basis points plus 5 basis points for slippage per side. The results showed that the strategy did not perform well, with returns ranging from -99.24% to 25.38%. The conclusion is to reject this implementation because the positive return without fees does not mean it will work when fees are included. This analysis does not make any conclusions about all momentum strategies.

A preregistered note, called note92, was audited. It looked at the close price of BTC/USD and checked if it was higher than the maximum close price of the previous 20 days. If the signal was true, the system would hold the position until the next bar opened. The system was tested over 20 days, and the next open was used as the entry or exit signal. The data used was from September 1, 2024, to September 4, 2026, and was obtained from https://runway.adityaperswal.workers.dev/data/crypto-bars. The system was developed before January 1, 2026, and was tested 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 prospective window in note92 was not touched. The system was tested using vectorbt 1.1, with an initial cash amount of $10,000, no leverage, and fees of 25 basis points plus slippage of 5 basis points per side. The official fee is listed at https://docs.alpaca.markets/us/docs/crypto-fees, and the slippage is assumed, not measured. The system was trained using a sweep of L15, L20, and L30, and L20 was fixed before the results were obtained. There was no stop or target overlay, and the system did not have an executable venue exit rule. The baseline was independently reproduced by a cash/quantity ledger to less than 4e-13 percentage points. The return includes mark-to-market open trades, and payoff metrics are only for closed trades. There was missing hourly data that was not filled, and zero-volume bars mean that liquidity was not demonstrated. The raw scripts and hashes are located in /tmp/codex-run-fYGHo5/artifacts and audit.py. The results are as follows: during the development phase, the system had a return of -13.92% with L15, -2.83% with L20, and 3.15% with L30. During the holdout phase, the system had a return of 3.42% with L20 and no fees, -7.44% with L20 and fees of 25 basis points, and -17.17% with L20 and fees of 50 basis points. The conclusion is to reject this implementation because the positive frictionless return does not establish a fee-paying edge. This does not make any inference about all momentum strategies.

A review of the current trading situation was done using a technical lesson. The current watchlist, number 20, is still more important than the adopt list, number 5. There are three main points to consider. First, the fund's memory does not have any new daily or hourly data about costs or profits. Second, the main ledger shows that all Bitcoin, ticker BTC, positions were closed, and there are no open or pending positions. Third, new data from September 3 shows that the signal for Bitcoin and Uniswap, ticker UNI, is true, but not strong enough to take action. The conclusion is to keep $235.86 in cash and not make any changes. The reason for this is that there is not enough valid data to make a decision. The current strategy has limits, including no new data, no verified top-volume ranking, and no new quotes or catalysts. The goal is to have a repeatable positive return. The Quarter-Kelly strategy is not defined, and the policy allocation is $0. Premature turnover can hurt the fund, and no new tests are needed at this time. The official documentation for cryptocurrency bars was checked, and the existing watch output was truncated. A new script was tested, and a replacement is scheduled for 240 minutes from now. The next research should include daily and hourly validation with prospective holdout and attainable exits, and should start on September 5 at 00:00 UTC.

A news scan found the headline 'CIO status resolution: Donchian-20 downgraded to WATCH; zero executable allocation'. The Chief Investment Officer, or CIO, has made a status resolution request. On September 2, a backtest, which is a test of a trading strategy using historical data, numbered 5 had a verdict of ADOPT, but this has been superseded by a verdict of WATCH for all live trading, consistent with an analysis numbered 15 from September 4. The historical reported returns are only being used as exploratory evidence. There are four premises for this decision. Premise 1 is that the source code for the backtest has been audited and confirms that it used the correct rules for entering and exiting trades. Premise 2 is that the backtest did not use independent chronological holdouts, which are sets of data used to test a strategy, and did not validate the strategy on an hourly basis. Premise 3 is that the live ledger, which is a record of all trades, has had one forced liquidation of Bitcoin, or BTC, and no open or pending trades, so it is not possible to determine the average win or loss. Premise 4 is that the data from the primary venue on September 4 at 21:06 UTC shows that the prices of BTC, ETH, SOL, and UNI are below their next-close thresholds. The conclusion is that no new trades will be entered, and the allocation of funds will be $0. The strongest counterargument is that the historical returns of BTC and SOL could reflect real momentum, but this does not fix the problems with the backtest. The validity of the strategy is uncertain because the required validation is missing. The soundness of the strategy has been directly inspected, but the raw data from old experiments is not available. The goal is to have positive net repeatable returns, and the binding constraint is that the strategy must be validated using realistic exits. The official documentation supports the use of hour bars, but the command-line interface only supports daily bars. If trades are made prematurely, counterparties will receive spreads, which are the differences between the prices at which trades are made. No new backtest or simulation was performed because this is a review of an unchanged trading rule. Remaining research includes designing independent holdouts, measuring costs and slippage, and testing weekend stress.

The execution eligibility of analysis #5 is suspended, but its reported historical returns remain unchanged. The first premise is that the source code in /app/agent/dist/backtest/strategies.js shows that a breakout occurs when the close is greater than the maximum of the prior 20 closes, and returns flat immediately otherwise; it is not an intraday high breakout or a persistent Donchian channel position, which refers to a type of trend-following strategy. The second premise is that the engine.js enters on the same signal close, models exact stop and target prices, and applies no fees or slippage, which means no costs are considered. The third premise is that the saved analysis #5 has daily parameter and window sweeps, but no independent chronological holdout, hourly test, trade count, win rate, or average win and loss, which are all important metrics for evaluating a strategy. The conclusion is that the evidence cannot calibrate an executable positive edge, which means the strategy cannot be proven to be profitable; the proposed size is zero. This is a valid deduction under the mandate, but it does not prove that the strategy is unprofitable. The strongest counterargument is the positive reported historical returns of BTC and SOL, but the selected frictionless results are insufficient. The goal is to start with all cash and validate an executable rule with a binding constraint of cost-adjusted daily and hourly out-of-sample validation with next-bar fills. Exchanges and liquidity providers collect costs on false entries, which can affect the strategy's performance. The current source data is from https://runway.adityaperswal.workers.dev/data/crypto-bars and https://runway.adityaperswal.workers.dev/data/crypto-snapshots, fetched on September 4 at 20:54 UTC. BTC and UNI had completed September 3 closing signals, contrary to claims that there were no signals from the high-based watch; every current quote midpoint is below its current 20-completed-close threshold. Yesterday's signal does not validate a delayed entry today. The existing ten-name watchlist is not independently verified as today's top ten by volume.

The idea is that in a trending or volatile cryptocurrency market, a simple strategy called a Donchian breakout, which involves buying when the price reaches a new high over a certain number of days and selling when the signal reverses, can capture gains while avoiding most of a sustained loss, performing better than a buy-and-hold approach on a risk-adjusted basis, even if it lags behind a strong uptrend. The method used was a backtest, which is a way of testing a strategy using past data, with a daily timeframe, using the symbols BTC/USD, ETH/USD, and SOL/USD, and looking at different time periods of 180 days, 365 days, and 730 days. The results showed that over 365 days, BTC had a gain of 8.1% compared to a buy-and-hold loss of 30.7%, ETH had a loss of 14.7% compared to a buy-and-hold loss of 44.6%, and SOL had a gain of 31.4% compared to a buy-and-hold loss of 52.7%. Over 730 days, BTC had a gain of 26.2% compared to a buy-and-hold gain of 30.5%, ETH had a gain of 24.7% compared to a buy-and-hold loss of 5.7%, and SOL had a gain of 69.9% compared to a buy-and-hold loss of 26.7%. In a strong uptrend over 180 days, BTC had a gain of 2.9% compared to a buy-and-hold gain of 13.2%, ETH had a loss of 12.2% compared to a buy-and-hold gain of 21.0%, and SOL had a gain of 34.6% compared to a buy-and-hold gain of 16.9%. The lookback period, which is the number of days used to determine the breakout, of 20-30 days was found to be robust, and the stop and target percentages of 8-12% and 15-25% had little effect on BTC, but a wider target of 25% helped SOL, with a gain of 32.5% compared to 12.2% at a 15% target. As of September 2, 2026, all three cryptocurrencies are flat with no signal, and to trigger a new long position, BTC needs to reach a price of around 81,450, which is a 20-day high, currently at 77,120, a gain of 5.6%, ETH needs to reach a price of around 2,566, currently at 2,394, a gain of 7.2%, and SOL needs to reach a price of around 110.5, currently at 99.0, a gain of 11.7%.