BCH/ETC Hyperliquid 4H Crossover with Profitable Pyramiding

Hypothesis

Derived by the Research Lead from MQL5 Code Base's Exp_2pbIdealMA_ReOpen by GODZILLA: test bar-close moving-average crossings and additions after favorable movement from the latest fill. The page describes these rules and reports tests without stop-loss or take-profit. The downloadable EA and custom-indicator files could not be retrieved, so their formulas, defaults and exact exit implementation remain unverified. This is a disclosed adaptation using standard EMAs, not a reproduction of 2pbIdeal1MA or 2pbIdeal3MA. Research Lead choices are BCH and ETC Hyperliquid perpetuals, 4-HOUR bars, EMA periods 20/60, percentage-based addition spacing, equal equity-relative tranches, bounded additions and opposite-cross exits. Orders execute at the signal bar's close. Source performance claims are unverified.

Hyperliquid is under its 20% venue target, while long-only and single-instrument experiments are overrepresented. Both instruments trade independently long and short; BCH is primary and ETC is an additional traded instrument, not a hedge. Corpus concentration was empty, so neither instrument's recent share is known. The source's profitable pyramiding distinguishes this experiment from the failed frozen LTC SMA baseline, but does not establish an edge. Options or cross-venue hedges would change the submitted mechanism. Four-hour bars permit substantially more Hyperliquid history than sub-hourly bars.

negative_expectancy: decline optimization of this BCH/ETC 4H EMA-pyramiding operationalization because both legs lose substantially after costs across 150 closed positions. This rejects the disclosed standard-EMA adaptation, not the unavailable original MQL5 indicator or trend following generally.

Implementation

Independent BCH and ETC 4-hour EMA crossover longs and shorts; 10% equity tranches with up to two additions after 2% favorable movement from the latest fill. Opposite crossings close and reverse after the close fills. No stop-loss or take-profit.

Implements the disclosed standard-EMA adaptation rather than reproducing unavailable custom MQL5 indicators. Uses scalar restart-persisted EMA and fill state, continuous percentage EMA spreads, aligned extra bars at or before the primary close, and full-tranche checks against 30% instrument and 60% portfolio mark-to-market exposure caps. EMA periods, spacing and addition count retain the research permissions; other settings remain fixed. Exposure caps constrain new orders rather than forcing trims of appreciating positions. Unverified source claims and uncapped adverse price movement remain research risks.

Verification Results

The sandbox reports -61.78% net return, 64.44% drawdown and -4.09% average closed-position return. Full-history evaluation should examine crossover whipsaws and adverse moves without protective stops. These results do not establish a correctness defect.

Backtest Review

Sharpe

-1.66

Total return

-64.46%

Max drawdown

67.05%

Trades

150

Win rate

15.3%

Profit factor

0.41

150 closed positions across 735 days provide a functioning sample; trade-detail PnL reconciles to the reported $64,459.06 loss.

Both instruments trade long and short, with sampled same-time reversals consistent with opposite-cross exits. Code implements bounded additions after favorable movement from the latest fill.

Net expectancy is negative: average trade return -4.0654% versus the +0.15% Hyperliquid review floor, profit factor 0.414 and Sharpe -1.664.

Both legs lose: BCH -$25,966.78 across 69 positions; ETC -$38,492.28 across 81. Adding back all reported commissions and impact still leaves approximately $53,962 of losses.

QA's economic concern is borne out by full-history results. Drawdown of 67.05% compounds the losses but is not the sole rejection reason.

Aggregated position records cannot independently verify each addition's fill spacing. No verified code defect or measured profitable parameter region supports iteration; no optimization sweep has been evaluated.

Outcome Summary

This standard-EMA adaptation needs evidence of a viable edge before parameter optimization; its rejection does not establish anything about the unavailable original MQL5 indicators.

The backtest-review verdict was abandon because both instruments lost substantially and expectancy remained negative even after adding back reported commissions and impact. Optimization and subsequent analyst and risk reviews were not reached.

Trade BCH and ETC Hyperliquid perpetuals using 4-hour EMA crossovers, bounded additions after favorable price movement, and opposite-cross exits.

The backtest returned -64.46% with Sharpe -1.664 across 150 closed positions over 735 days. Maximum drawdown was 67.05%, profit factor was 0.414, and average trade return was -4.0654%.

Analysis

Exp_2pbIdealMA_ReOpen