Strategies

After two iterations, the strategy tested synchronized ETH–BTC spread reversion with capital-relative sizing and frozen trade-level hedge ratios. Its four baskets across 732 days lost approximately $142 net, with a Sharpe ratio of -0.068. Backtest review abandoned this trading specification before optimization because expectancy was negative and the sample was too small, without concluding that all cointegration strategies are unprofitable.

After four iterations, the faithful SMA50/200 long–short port reached backtest review with corrected sizing drift and reconciled accounting. Its 41 closed positions produced a -89.68% return, -0.966 Sharpe, and 93.39% maximum drawdown. Review abandoned the run because losses persisted before commissions and modeled impact, while the fixed source contract offered no parameters to optimize. Optimization and later review stages were not reached.

The strategy tested confirmed swing-boundary reclaims on LTC four-hour bars with Parabolic SAR protection. Its backtest lost 4.78% over 118 closed trades and produced a Sharpe of -0.327. After five iterations, it was abandoned because the final revision reproduced iteration 3's trading results exactly; optimization and later reviews were not reached.

The strategy attempted weekly proportional momentum on DIA using causally volatility-normalized returns and final-session-close rebalancing. After 5 iterations, sandbox verification detected a next-session fill that violated the execution contract. It was abandoned before performance evaluation or later reviews, leaving the trading edge unassessed.

Across three iterations, the strategy operationalized two-endpoint string maps as causal nearest-neighbor forecasts for AVAX 4-hour trading. Its backtest returned -4.48% across 606 closed positions, with Sharpe of -0.473 and average net trade return of -0.1744%. Modeled impact totaled $6,161, while estimated capacity was $7,439 against the $100,000 engine account. Backtest review abandoned the tested configuration before optimization; the result does not judge the paper authors' undisclosed trading rules or the entire string-map family.

Across three iterations, HyperliquidHistoricalReturnsLtcBch implemented daily LTC/BCH directional signals while preserving same-direction positions through renewed insights. Its backtest produced a -76.95% return and -1.415 Sharpe across 1,732 closed positions, with losses in all three volatility regimes. Backtest review abandoned the frozen specification for negative expectancy before optimization or risk review; the verdict applies to this operationalization rather than QuantConnect's alpha model across other markets or portfolio constructions.

After two iterations, HyperliquidAvaxAtomGpMomentumReversion reached backtest review with functioning capital-relative sizing and long/short participation on AVAX and ATOM. Its 102 closed positions produced a -1.953% net return, -0.185 Sharpe, and -1.043% average net trade return. The reviewer abandoned it for negative expectancy and declined optimization because no measured profitable parameter region or specific implementation defect justified further iteration.

After one iteration, the strategy tested Peter_O’s stochastic-cross example on SOL Hyperliquid 4-hour bars using close fills, 20x capped exposure, and no protective stop. The backtest lost 100% of capital, reported a Sharpe of -16.58, and recorded losses on all 12 closed positions. Backtest review abandoned the run before optimization because the account was exhausted and changing leverage or adding a stop would alter the fixed specification.

After two iterations, SpyWeeklyTwoRuleShortHedge reached backtest review as a port of QuantRocket's standalone trend hedge. It lost 37.86% across 48 closed short positions, with Sharpe -0.055 and maximum drawdown 63.42%. Review abandoned it for negative standalone economics and declined optimization because its rules were fixed. Its negative beta of -0.5003 suggested potential hedging utility, but its contribution to a combined portfolio remained untested.

After one iteration, HyperliquidLtcBchConfirmedSwingTrend tested independent LTC/BCH swing breakouts with multi-day holdings and delayed trailing protection. Its 58 closed positions produced a -14.834% return and -1.931 Sharpe, with both instruments losing before execution costs. Backtest review abandoned the implementation for negative expectancy without proceeding to optimization; the verdict does not establish that other trend-following specifications fail.

GdxGldBayesianSpreadReversion attempted GDX–GLD convergence trades gated by online Bayesian stability monitoring. After three iterations, it failed Layer 1.5 hypothesis/config consistency verification, which flagged the optimization plan’s fixed 1-DAY bar interval. It was abandoned before backtesting or later evaluation stages, leaving no performance metrics.

AvaxFrozenOpportunityMetaLabel tested frozen two-stage classifiers for AVAX 4-hour trades with Binance context and a fixed holding period. After one iteration, its backtest produced 204 closed trades, a -1.994% return and Sharpe of -0.143. Backtest review abandoned it for negative net expectancy before optimization. The missing executed primary-only control means this result does not establish that the secondary classifier worsened performance or that meta-labeling generally fails.

SolHummingbotZeroRatioInventoryHedge preserved Hummingbot’s zero-ratio SOL hedge defaults with a fixed spot inventory fixture worth 20% of starting capital. Sandbox verification evaluated 1,439 one-minute bars and produced the expected zero hedge trades. After three iterations, the strategy was abandoned at Layer 3 for producing no trades; backtest review, optimization, analyst review, and risk review were not reached.

SpyCausalMarkovBollingerPullbacks attempted regime-dependent daily long–short SPY pullback trading using point-in-time filtered variance probabilities. After three iterations, all six synthetic scenarios failed because a supplied daily bar timestamp did not fall on an exchange session. The run was abandoned during verification, leaving no backtest or optimization results to assess its trading performance.

The strategy adapted dynamic logistic momentum selection to a monthly long-short basket of Hyperliquid perpetuals, requiring 48 complete months of history before trading. After three iterations, Layer 3 verification reported zero trades across 19 evaluated daily LTC bars, with no entry side returned. The run was abandoned at verification, leaving no performance or optimization results for later review.

The strategy tested a regularized characteristic portfolio across nine US sector ETFs, rebalanced monthly with frozen coefficients. After three iterations, Layer 4 QA identified protective stops that covered only the initial partial fill and a mismatch between modeled holdings and stop-driven execution. It was abandoned during verification, with no backtest metrics or optimization results available.

XrpHyperliquidRsiAlphaInsights completed one iteration translating QuantConnect's directional RSI insights into an executable XRP strategy. Its backtest lost 24.65% across 25 closed positions with a Sharpe of -0.671, as short losses outweighed long gains. Backtest review abandoned this frozen operationalization for negative expectancy before optimization or later reviews.

SpyDailyTrendVixDeterministicPrior tested a deterministic trend-and-VIX exposure prior on daily SPY bars. After three iterations, sandbox verification produced no trades across 1,393 evaluated bars, ending the run before later backtesting, optimization, analyst review or risk review. The requested outbox file could not be written because the filesystem is read-only.

This strategy translated gvman99's equity-oriented Pine strategy, built on level retests and EMA taps, to SOL perpetuals on OKX at 5-minute resolution, using default parameters. Over 1743 trades and about 5.6 years the full-history backtest lost 83% (Sharpe -6.17, -0.278% per trade), and the losses were uniform across years, regimes and sides. Before costs the signal showed essentially no edge, and costs turned that into heavy losses. The analyst abandoned it at backtest review after one iteration, so optimization, risk review and promotion were never reached.

The strategy adapted an article's PCA basket-reversion idea to LTC, BCH and ETC perpetuals on Hyperliquid, trading deviations in the lowest-variance standardized-price component. Its first full backtest, over 732 days and 567 legs, lost 27.3% with Sharpe -0.79, a 39% drawdown and a per-trade return of -0.161%, while modeled impact came to 150% of gross PnL. The forecast showed no measurable information (IC -0.089) and results got worse each year, which confirmed the edge concern QA had raised in the sandbox. The analyst abandoned it at backtest review after one iteration, before optimization.

After four iterations, HummingbotPmmSimpleWld tested symmetric passive market making with per-level exits on WLDUSDT one-minute bars. The backtest produced 10,546 closed net positions, Sharpe -6.733, and a -100% return ending in liquidation. Backtest review abandoned the strategy for measured negative expectancy; its fully frozen parameter plan provided no authorized optimization path.

The hypothesis was that ARB and OP share enough common drivers for their log price ratio to revert after name-specific shocks, with a correlation gate to separate it from the failed BCH-LTC trade. After 1 iteration, the backtest on clean, fully aligned data had 169 spread trades, lost 35.3% with a Sharpe of -0.71, and lost in every year and every volatility regime. It traded about 56 spreads a year, roughly double the 15-30 the hypothesis predicted. Because the spread lost money even before costs and the signal showed no forecasting information, the analyst abandoned it at backtest review and optimization never ran.

After one iteration, the strategy tested daily SPY/IWM relative-strength rotation using approximately equal opposing leg notionals. Input audits passed across 5,940 aligned sessions, but the baseline lost 14.07%, with Sharpe 0.027 and profit factor 0.945. Backtest review abandoned it for negative expectancy before optimization or later reviews. The result concerns the price-only adaptation, leaving CICC's unavailable fund-allocation lead signal untested.

This hypothesis moved the factory's promoted negative-funding carry basket to Binance COIN-M inverse perps, betting that persistent hedger short pressure and lower fees would make the carry larger. After 3 iterations the backtest showed +81.5% over 641 trades at 0.67% per trade, but with a 40.5% drawdown and a 31% loss in the LUNA window. Splitting the trade ledger showed that funding contributed only about +$2k net of costs, roughly 0.015% per trade, while most of the profit was long coin beta with negative alpha. The analyst abandoned it at backtest review, before optimization, because tuning could not lift the carry above costs.