SolHyperliquidMomentumBtcRegimeGatedLS

SOL Hyperliquid Momentum Gated by BTC Market-Regime, Long-Short (Execute SOLUSD.HYPERLIQUID Perp on 4H Momentum ONLY When the Market Leader BTCUSD.HYPERLIQUID Trend Agrees — Never Fight Bitcoin's Regime — Flat on Disagreement, ATR-Trailing, 3-Parameter)

A LONG-SHORT, two-instrument (same-venue) pure-OHLCV trend strategy that EXECUTES only on SOLUSD.HYPERLIQUID but GATES every directional trade by the Bitcoin market regime (BTCUSD.HYPERLIQUID). It stays in the only family that has survived optimization here (directional momentum, long-short, pure OHLCV — the BTC dual-timeframe confluence at Sharpe ~1.99) but its distinguishing idea is a CROSS-ASSET MARKET-BETA GATE rather than a same-asset confirmation: high-beta alts like SOL follow Bitcoin, so SOL momentum only continues reliably when BTC — the market leader — is trending the SAME way; SOL momentum that fights BTC's prevailing regime is exactly where alt breakouts whipsaw and bleed. So the strategy goes long SOL only when SOL is moving up AND BTC is in an uptrend, shorts SOL only when SOL is moving down AND BTC is in a downtrend, and sits FLAT whenever SOL and BTC disagree. BTC is a GATE ONLY — it is never traded — so this is NOT a relative-strength / pairs-rotation spread trade (the empirically-exhausted family) and NOT the same-asset dual-timeframe confluence; the second instrument is a market-context filter. It is materially different from the four HL momentum strategies already in this factory's pipeline: SOL Donchian breakout (single-instrument channel), ETH RSI pullback, AVAX efficiency-ratio gate, BTC dual-TF confluence — none uses a cross-asset market-leader regime gate. Fills the most under-represented data-safe buckets: HYPERLIQUID venue (5.9% vs 20% quota), long_short direction (13.7% vs target), and multi-instrument scope. Both legs are HYPERLIQUID 4H bars (~2.3 years usable history, above the HL sub-hourly wall) — same venue, so NO cross-venue price_precision schema-merge landmine. Pure OHLCV, no supplementary feed (best-surviving tag family at 4%). Avoids every recent death mode: no options-expiry cadence, no cross-venue spot-lookup/schema bug, no funding fee-floor, no 90-day liquidation-feed wall, no COIN-M booking defect.

Implements the hypothesis directly: SOL executes, BTC gates, never fight Bitcoin, flat on disagreement, ATR trailing, long-short, same-venue Hyperliquid 4H, pure OHLCV, three tunable parameters. Four notes. (1) The structural risk in a two-leg SAME-timeframe strategy is the base class's cross-leg alignment barrier, which defers the primary signal until every same-timeframe extra leg has a bar at the identical timestamp — if coverage were patchy the signal would starve. I measured it rather than assuming: SOL and BTC Hyperliquid 4H share 100% of timestamps (2,191/2,191 across the 365-day verification window and 5,494/5,494 across the full history back to 2024-01-25), so the barrier is pure upside here — it guarantees a contemporaneous regime read instead of a stale T-1 one, which is precisely the phantom-signal bug that barrier exists to prevent. (2) I also dry-ran the FULL compound gate (momentum threshold AND BTC agreement) before submitting, since stacked conditions are the top zero-trade failure mode: at the shipped defaults it is satisfied on 280 long and 416 short bars inside the verification window, and SOL's |momentum| distribution has median ~1.4 / 75th pct ~2.3 ATR so the 1.5 threshold sits just above the median rather than in the tail. (3) BTC is genuinely gate-only — there is no code path that submits an order on the extra instrument — so this is not a pairs/relative-strength spread and carries no second-leg fees; and if the BTC feed were ever absent the regime stays 0 and the strategy trades nothing rather than degrading into ungated SOL momentum. Both legs are the same venue, so there is no cross-venue precision/schema merge. (4) Risk controls follow the compounding-blowup lesson: 1% of equity risked per trade across the initial 3-ATR stop, gross notional hard-capped at 1x equity, no leverage set or referenced, and the cooldown is anchored to bar timestamps rather than a bar counter so it is restart-safe and identical in backtest and paper. One honest caveat for the analyst: the BTC regime gate is highly correlated with SOL's own trend, so it will remove fewer trades than an independent filter would — its value is concentrated in the alt-specific whipsaws where SOL moves against the market, and if the backtest shows the gate barely changes the trade set, the right conclusion is that the market-beta filter adds nothing here rather than that the parameters need tuning.

Edge too thin to optimize. Across a decisive 187-trade sample the BTC-regime-gated SOL momentum nets Sharpe 0.205 with CI [-0.797, 1.17] straddling zero, profit_factor 1.086 (< 1.2, fees eating nearly all gross edge), information_ratio -0.29, and only 9.3% total return over ~2.5 years (annual returns ~1-3% each). The per-trade edge (avg_trade_return_pct 0.388%) clears the fee floor, so this is not a hard fee-edge death, but the aggregate profitability is statistically indistinguishable from zero — the cross-asset 'never fight Bitcoin' gate, the entire distinguishing mechanism, did not create a meaningful edge. This is the pure-OHLCV momentum/trend family (PF < 1.2, zero survivors historically). Optimization requires OOS Sharpe > 0.5; lifting a 0.20 Sharpe with a zero-straddling CI to that bar would require a 2.5x out-of-sample improvement, i.e. overfitting the search, and no robust parameter region is evident. Market impact (impact_cost_pct 24.2%, capacity ~$1.7M) further degrades the thin edge. Abandon rather than spend 2 hours optimizing a near-breakeven signal.

Long-short 4H momentum on the SOLUSD Hyperliquid perpetual, permitted only in the direction of the Bitcoin market regime. Each bar it computes SOL's ATR-normalized 10-bar momentum (sign = direction, magnitude = ATRs travelled) and reads BTCUSD.HYPERLIQUID's regime as its close versus a 100-bar SMA. It goes long SOL when momentum is at or above +1.5 ATR AND Bitcoin is in an uptrend, shorts SOL on the mirror condition, and sits flat whenever SOL and BTC disagree or Bitcoin's regime is unknown. BTC is a gate only and is never traded — no orders, no notional, no fees on that leg. Exits use a 3x-ATR chandelier trail from the extreme since entry or a Bitcoin regime flip against the position, with a 12-hour calendar-anchored cooldown after each exit. Sizing risks 1% of equity across the initial stop with gross notional capped at 1x equity; leverage 1.0. Exactly three tunable parameters (momentum_window, momentum_threshold, atr_trail_mult); the BTC regime window, ATR window, risk fraction, notional cap and cooldown are locked constants.

Judge on the full ~2.3yr HL history (~150 trades) plus walk-forward; abandon at backtest_review only if full-history avg_trade_return_pct stays below the floor.

Marginal-negative, NON-DECISIVE sandbox. total_return -1.87%, profit_factor 0.93, win_rate 0.31, avg_trade_return_pct -0.248% (below the fee floor), Sharpe -0.21 with CI [-1.86, +1.40] straddling zero, over 77 trades. Essentially break-even with FAVORABLE reward:risk (avg_win $1076 vs avg_loss $523 = 2.06x; break-even win rate ~32.7% vs realized 31.2% — razor-thin miss), the same profile as the SOL Donchian sibling. Not a decisive no-edge verdict on a single 365-day / 77-trade window.

At backtest_review, compare the gated trade set to an ungated SOL-momentum baseline: if the BTC gate removes only a handful of trades and does not improve expectancy/Sharpe, the market-beta filter is not doing work.

GATE MAY BE NEAR-INERT (developer's honest caveat, seconded). The BTC regime is highly correlated with SOL's own trend, so the 'never fight Bitcoin' gate likely removes few trades and may add little over naked SOL momentum. If the gate barely changes the trade set, the strategy reduces to the OHLCV single-asset momentum class (zero survivors, L7) and the cross-asset thesis adds nothing. Edge/attribution question, not a code defect.

Require robust walk-forward with OOS Sharpe clearly above zero before promoting; use the engine trade count (77), not the dry-run's 280/416 eligible-BAR tally.

Pure-OHLCV momentum/trend class prior (L7) is unfavorable (zero OHLCV-only survivors). The cross-asset market-leader gate is a fresh idea within the surviving momentum family and the code is clean, but the class prior plus the below-floor sandbox warrant a high OOS bar.

Clean, adequately-sized sample (187 trades, 916 days) with balanced two-sided book (90 long / 97 short)

Positive in all three calendar years and modest max drawdown (9.9%) — not a single-regime blow-up

avg_trade_return_pct 0.388% clears the HYPERLIQUID fee floor — the per-trade edge is not negative

Sharpe 0.205 with CI [-0.797, 1.17] straddling zero — edge indistinguishable from no-skill

profit_factor 1.086 (< 1.2) — fees consume nearly all the gross edge

Total return only 9.3% over ~2.5 years; annual returns ~1-3% each are trivially small

information_ratio -0.29, probabilistic_sharpe 0.65 — no risk-adjusted edge

Pure-OHLCV momentum/trend family (PF < 1.2, zero survivors); impact_cost_pct 24.2% and capacity only ~$1.7M further erode the thin edge

This strategy proposed a fresh angle on an over-mined family: rather than same-asset confirmation, it gated SOL 4H momentum on Hyperliquid by Bitcoin's market regime, going long or short only when SOL and BTC agreed and staying flat otherwise. The initial backtest was clean and balanced — 187 two-sided trades over 916 days, positive in all three calendar years, and only 9.9% max drawdown — and its per-trade return of 0.388% cleared the venue fee floor. But the aggregate edge was statistically indistinguishable from zero: Sharpe 0.205 with a CI spanning [-0.797, 1.17], profit factor 1.086, and just 9.3% total return over ~2.5 years. At the pre-optimization backtest-review gate the analyst abandoned it after one iteration, concluding the cross-asset 'never fight Bitcoin' gate — the entire distinguishing mechanism — created no meaningful edge and that lifting it to the optimizer's bar would require overfitting; later pipeline stages were never reached.

A cross-asset market-leader regime gate did not lift SOL momentum out of the pure-OHLCV trend family's near-breakeven profile (PF < 1.2), so novel filters layered on an exhausted signal family should be screened for a distinguishable edge before committing optimization time.

The analyst issued an 'abandon' verdict at the pre-optimization backtest-review gate, judging the edge too thin to optimize: Sharpe 0.205 with a zero-straddling CI and PF below the 1.2 floor would need a ~2.5x OOS improvement to clear the optimizer's 0.5 Sharpe bar, which would amount to overfitting. No optimization, analyst-decision, or risk-review stage was reached.

A long-short, pure-OHLCV 4H momentum strategy on SOLUSD.HYPERLIQUID whose distinguishing edge was a cross-asset 'never fight Bitcoin' gate — taking SOL trades only when BTCUSD.HYPERLIQUID's regime agreed with SOL's direction, flat on disagreement, with an ATR chandelier trail.

Over a 187-trade, 916-day backtest (90 long / 97 short) it returned 9.32% total with a Sharpe of 0.205 (CI [-0.797, 1.17]), profit factor 1.086, and information ratio -0.29. Per-trade edge (avg_trade_return_pct 0.388%) cleared the HYPERLIQUID fee floor and max drawdown was a modest 9.9%, but aggregate profitability was statistically indistinguishable from zero.