Strategies

This strategy turned a SuperMind weekly-MA golden cross plus daily KDJ recipe into fixed rules and tested them on six sector ETFs over 24.5 years of daily data. It traded 193 times with low risk (5.66% max drawdown, beta 0.009) and its average trade return cleared the USEQ cost floor, but it earned only +2.32% in total, with Sharpe 0.099 and a signal IC that could not be told apart from noise. The analyst found that 2020 alone produced more than all of the profit and that only 44% of years were positive, and abandoned it after one iteration as no_edge. Risk review, promotion and paper trading were never reached.

PerpDiscountSpotAccumulationRotation used the perp-vs-spot basis on 10 alts as a signal-only input to a weekly long-only spot rotation, gated by a 50-day SMA trend filter. Its initial backtest looked strong (Sharpe 1.30, 7.82% average trade, 191 trades), so the backtest review sent it to optimization despite a 44.5% drawdown. Walk-forward results were erratic (OOS Sharpe mean about 0.52, 4 of 10 windows negative, IS/OOS about 7.8x), and the strategy underperformed the equal-weight alt basket (IR -0.27). After 2 iterations, the analyst abandoned it for breaching its own 35% drawdown limit, uneven yearly results, and no evidence that basis selection added value.

HyperliquidBchLtcShockExhaustion tested whether large, volume-backed relative moves between BCH and LTC partly reverse once they stop accelerating, using equal-dollar long/short pairs on Hyperliquid 4h bars. Its first backtest returned +4.58% with Sharpe 0.37 and an average trade return of 0.164%, just above the 0.15% Hyperliquid floor, so the backtest review sent it to optimization as a test, not as evidence of robustness. Walk-forward optimization found no real edge: out-of-sample Sharpe averaged about 0.06 against about 2.6 in-sample, parameters were unstable with cliffs, and all the profit came from 2024. The analyst abandoned it after one iteration, describing it as overfit with no edge, the fifth BCH/LTC relative-shock attempt.

This run translated Jesse's KDJstrategy onto LTC spot daily bars and kept the source's default parameters. It took 3 iterations to fix configuration and order-contingency issues, after which the port ran cleanly. The full-history backtest lost 39.1% over about 6 years on 336 trades, losing money in six of seven years and in all three volatility regimes, and it lost money even before commissions. The analyst abandoned it at backtest review for negative expectancy, which confirmed QA's earlier concern about the edge, and optimization, risk review and promotion were not reached.

JesseMAGenLinkHyperliquid reproduced the Jesse MAGen linear-regression cross with ADX and fixed ATR brackets on LINK Hyperliquid 4H bars, keeping all source defaults unchanged. After two iterations, the full 777-day backtest was mechanically clean (no missing bars, accounting reconciled, funding included) but lost 60.4% over 166 trades, with Sharpe -0.57 and a 70.5% drawdown. The loss came from a negative gross edge rather than fees, and the signal showed no information (IC -0.012), so the analyst abandoned it at backtest review. Optimization, risk review and promotion were never reached.

HyperliquidThreeSleeveDualThrust tested a perpetual-futures, completed-bar, 1x-leverage adaptation of the Dual Thrust breakout on LINK, LTC and BCH 4-hour bars. Its first backtest covered 721 trades on clean data and lost 71.7%, with a Sharpe of -1.98, profit factor 0.658, a 75.4% max drawdown and losses in every year and every volatility regime. The analyst abandoned it at backtest review, before optimization, because the edge was negative even before commission and impact (about -0.31% per trade). Optimization, risk review and promotion were never reached, and the verdict applies only to this adaptation, not to the original intrabar quarterly-futures version.

After two iterations, HlFundingWillingnessMomentumBasket tested funding-confirmed relative momentum across six Hyperliquid perpetuals. Its 89 completed baskets produced -8.71% return, Sharpe -0.496, and average net trade return of -0.613%, while drawdown exceeded the registered limit. Backtest review abandoned the tested configuration before optimization; the verdict did not disprove the broader mechanism or every parameter variation.

The strategy attempted a faithful NumPy port of Quantiacs’ LSTM model for daily AAPL target weights, with a declared initialization RNG deviation. Across five iterations, execution fidelity remained unresolved because the platform lacked the required next-session-open hook. The final sandbox verification processed 0 bars and failed, leaving performance unevaluated and the run abandoned before optimization or later reviews.

EthOkxFrozenAtrRsiStrategy attempted a frozen translation of vn.py’s ATR/RSI strategy with capital-relative sizing. Its five-day verification smoke test produced -0.139% per trade over 38 closed positions and a profit factor of 0.626. After three iterations, QA still found entry and trailing-stop execution inconsistent with the source, leading to abandonment before full backtesting or optimization.

The strategy translated QuantRocket’s fixed skipped-month momentum rules to a declared ten-stock USEQ universe with monthly equal-weight long/short allocations. Its sandbox produced negative average trade returns and a Sharpe of -0.16 across 104 closed legs. After three iterations, QA identified a blocking warm-up defect caused by differing instrument history start dates, and the strategy was abandoned before full-history backtesting or optimization.

Across two iterations, the strategy implemented percentage-rebound reversal entries and rolling-midpoint exits on ATOM Hyperliquid 4-hour bars. Its backtest lost 35.14% over 260 closed trades, with negative returns in every reported year and volatility regime. Backtest review abandoned it because the estimated gross edge was below fees and drawdown exceeded the pre-registered limit, so optimization and later reviews were not reached.

AvaxFrozenChannelTurtleSignal reproduced the supplied Turtle rules with four-hour bars and capital-relative sizing while keeping parameters fixed. Its backtest produced a 10.77% return, Sharpe of 0.426 and 212 closed trades, with 15.26% maximum drawdown. After four iterations, the run was abandoned because the final comment-only revision reproduced the previous results exactly, leaving nothing new for the analyst to judge.

The strategy sought to capture failed early-week breakdowns below altcoin weekend lows using long altcoin positions hedged with BTC shorts and a fixed 96-hour holding period. Its supporting prestudy reported a +1.04% mean hedged spread return across 1,316 events, gross of costs, but the pipeline produced no backtest results. After three iterations, a Layer 1.5 consistency failure concerning fixed leverage of 1.5 ended the run before backtesting, optimization, or later reviews.

LtcHowtraderTurtleSignal ported the Howtrader TurtleSignalStrategy to LTC perpetuals on Hyperliquid at 4h, keeping the source defaults and changing only the bar timeframe and the position sizing. After 2 iterations, the backtest produced 236 trades over 799 days with a -36.2% return, Sharpe -1.59, profit factor 0.545 and a 36.5% drawdown, more than twice the pre-registered 15% limit. QA had pointed to execution costs, but the analyst found that the signal also loses before costs, and the signal direction was confirmed correct, so the loss is not a code defect. The analyst abandoned it at backtest review, and optimization, analysis and risk review were never reached.

After two iterations, DotHyperliquidDailyRSI2 tested the source strategy’s frozen rules with entire-balance, fee-adjusted sizing at 1.0 leverage. Its 55 closed trades returned -25.68% net, with Sharpe 0.024 and maximum drawdown of 58.96%, exceeding the declared 40% constraint. Backtest review abandoned the run because both trading directions lost after costs and funding, and the frozen specification offered no authorized optimization path.

LiminLtcTickTurtle tried to replicate limin's Apache-2.0 TqSdk Turtle example on Hyperliquid LTC, executing on trade ticks and keeping N and unit size fixed for the length of each campaign. Over 3 iterations the code never got past verification. Every Layer-2 synthetic scenario crashed at setup on a self-imposed check that parameters.leverage must equal config.leverage. The pipeline abandoned the run in the verification loop, so backtesting, optimization, analyst review and risk review never ran.

CryptoEquityResidualShockReversal paired COIN and HOOD daily bars with an hourly Bybit BTC perp hedge. It planned to fade 2-sigma residual shocks, using a rolling OLS beta and bisect-based BTC price lookup. Over 3 iterations the code was reworked, including a change to bounded extra-bar deques in iteration 3 to address an earlier timeout, but it never passed verification. The final Layer 3 sandbox backtest failed because a COIN.USEQ position executed after the evaluation boundary, with 0 primary bars reported as processed, so the run was abandoned before backtest review, optimization, or analysis.

IanzengRwaRiskParityV25 tried to port ianzeng123's four-asset Bybit risk-parity allocator (BTC, gold, SPY, crude oil) to the platform, keeping the source's paper-branch allocation logic. Over 3 iterations the code never passed Layer 3 verification, because the sandbox smoke test went over its 300s limit on a window of about 20,000 hourly bars. The likely cause is that every bar rebuilt timestamp maps and intersected them across all four legs' full histories. The run was abandoned before any backtest, optimization or analyst review, so no performance numbers exist.

LtcBchFiveLevelCappedGrid tested an FMZ-inspired five-level range grid on LTC and BCH perpetuals, swapping martingale sizing for capped, capital-relative tranches. It went through 3 iterations but never passed code verification: the Layer 3 sandbox smoke test kept exceeding the 300s limit on a 2,400-day 4-hour LTC Hyperliquid window. No backtest, optimization, analyst or risk review was produced, so the grid's economics were never measured before the run was abandoned.

The hypothesis tested whether short-lived deviations in the LTC/BCH log-price spread on Hyperliquid revert profitably over 1–5 days, using a Kalman/AR(1) state model, explicitly defined trade direction and a cost-aware forecast filter for entries. The code never cleared Layer 3 verification because the sandbox backtest kept going over the 300s limit. In iterations 2 and 3 the developer left the trading rules unchanged and noted that timed replays showed the time going to catalog directory scans under heavy host I/O pressure, with a no-logic baseline strategy no faster. After 3 iterations the run was abandoned as a verification loop, before any backtest metrics, optimization or analyst review existed.

GdxGldJohansenBollingerPairs set out to port QuantRocket's GDX/GLD Johansen/Bollinger pairs strategy to the USEQ venue as faithfully as possible, re-implementing statsmodels' coint_johansen in numpy because statsmodels is not available in the sandbox. Across 3 iterations, the Layer 3 sandbox smoke test kept going over its 300-second limit on the ~20,000-bar GDX daily window. The run was abandoned in that verification loop. No backtest, optimization or analyst verdict was ever produced, so it is still unknown whether the strategy has an edge.

FSupertrendStrategyLinkSpot was a line-for-line port of an open-source Freqtrade triple-Supertrend strategy, run long-only at leverage 1 on LINK spot 1h bars because the source never enables shorting. It went through 3 iterations and never got past code verification. Each time it failed the Layer 1.5 hypothesis/config consistency check, which flagged that the optimization_plan fixes leverage at 1.0, and it was then abandoned as a verification loop. No backtest, optimization or analyst review took place, so the strategy's edge was never measured.

HyperliquidDownsideBetaPremiumPair tested whether a published crypto downside-beta premium carries over to four Hyperliquid perpetuals, using beta-matched long/short pairs on synchronized 4-hour bars. Over 4 iterations the code was revised: leverage was set explicitly to 1.0 in iteration 3, and iteration 4 replaced the per-bar array rebuilds with a numpy ring buffer. Every attempt still timed out in the Layer 3 sandbox backtest at the 300s limit, so the run was abandoned in verification. It never reached backtesting, optimization, analyst review or risk review, and the hypothesis is still untested.