SolExtremeFundingContrarianShort1H

SOL Extreme Funding Rate Contrarian Short on 1H

A simple single-instrument short-only strategy on SOLUSDT perpetual futures that fades over-leveraged long positioning when 8-hour funding rates spike to extreme levels. The strategy uses 1-hour bars and one supplementary input (funding rate). When funding > +0.05% per 8h (annualized ~55%), longs are crowded and paying shorts heavily; historically these regimes coincide with local tops as cascading liquidations punish leveraged longs. The short position captures (a) the funding payment received per 8h funding period, AND (b) mean reversion in price when crowded longs unwind. Holds are intraday (typically 6-24h), so accumulated fee drag is minimal relative to the 1.5%+ per-trade target. Risk profile: 1.5% capital risk per trade with strong-signal scaling (2% risk when funding > +0.08%), fixed 1.5% adverse-move stop, max 20% equity in any single position. The hypothesis is intentionally minimalist (one instrument, one indicator, one timeframe) to avoid the verification timeouts that plagued recent multi-instrument and multi-indicator hypotheses.

Implements the hypothesis faithfully as a clean single-position strategy that fits the base template natively: should_enter returns OrderSide.SELL to open the short when funding is extreme, and the base covers it on exit — no multi-leg machinery needed. Venue is BINANCE USD-M futures (MARGIN) because the strategy must SHORT and because the edge is explicitly funding-based: only by holding the perp does the short receive the funding payments, which the backtest engine accrues automatically (supp_collector always loads the venue funding series for accounting), so the carry leg of the edge is captured without any PnL bookkeeping in strategy code — the strategy only READS funding for the signal. The funding series is parsed once from the dual-key {ns: rate} dict into sorted arrays and looked up with searchsorted at-or-before, with a 9h staleness guard so a gap in funding yields no signal rather than a stale trade. Sizing is risk-based with a hard 20%-equity notional cap (the cap binds, since 1.5%/1.5% risk/stop would otherwise imply 1x notional), giving controlled exposure at leverage 1.0. Holds are intraday (6-24h) so fee drag is small relative to the 1.5-2% per-trade target. Verified locally: Layer 1 static (clean), Layer 2 synthetic (all 6 scenarios pass), and a real-data check on the actual SOLUSDT funding parquet (6,241 stamps over 5.6 years, 100% coverage, no gaps) confirming ~4.9% of hourly bars exceed the +0.05% entry threshold (~2,400 entry-eligible bars -> hundreds of round-trip shorts, far above the 5-trade floor) and that strong-funding bars correctly trigger the 2% risk tier capped at exactly 20% equity notional.

The SOL extreme-funding contrarian short is catastrophically unprofitable over a thorough 557-trade sample: Sharpe -5.48 (CI [-7.52, -3.16], entirely negative), profit factor 0.70, negative expectancy (-$81/trade), total return -41%, Sortino -6.52, PSR 0.0, max drawdown 42.9%. Funding is credited by the engine, so this is a fair test, and the mechanism is falsified: shorting SOL when 8h funding spikes positive means shorting during strong uptrends — and SOL's 2021 parabolic bull run (where extreme funding clusters) ran the shorts over (2021 -41.1%). The funding received is a tiny fraction of the directional loss. The strategy is structurally a short-only momentum-fade concentrated in frothy bull periods (exposure 5.7%, trades cluster 2020-2021), i.e. the worst possible regime for shorts. This is the known 'fade the funding spike' failure pattern (the prior BTC funding-contrarian short failed the same way). Optimizing funding thresholds over 225 trials cannot flip a -5.48 Sharpe / 0.70 PF / negative-expectancy mechanism — best-of-N over a losing signal would fail holdout. This is the 'results so poor no optimization can help' case. Not revise_hypothesis: contrarian-short-on-extreme-funding has no proven sibling and the failure is the mechanism itself (fading crypto's structural up-momentum), not a salvageable target. FAILURE PATTERN: short-only contrarian fade of extreme positive funding on a crypto perp has strongly negative edge — high positive funding marks strong uptrends, so shorting it is dominated by continuation, producing a deeply negative Sharpe and a bull-run-concentrated drawdown; the small funding credit never compensates the directional loss.

Short-only contrarian funding-fade on SOLUSDT.BINANCE perpetual, 1H bars, using the 8h funding rate as its single supplementary input. Each bar it looks up the most recent funding rate at or before the bar timestamp (8h staleness guard) and returns it as a continuous signal. When funding exceeds entry_funding (+0.05%/8h), longs are crowded and paying shorts, so it opens a SHORT; the position then earns the funding it receives each 8h period (accrued by the engine) plus any mean-reversion as crowded longs unwind. It exits on a 1.5% adverse stop, a 2% take-profit, funding normalizing below exit_funding, or a 24-bar (24h) time stop. Position sizing is risk-based (risk_pct/stop_pct of equity, scaled up to strong_risk_pct when funding > +0.08%) and hard-capped at max_position_pct (20%) of equity notional, so no single position exceeds 20% of capital; leverage stays 1.0. Single instrument, single indicator, single timeframe — intentionally minimal to avoid verification timeouts.

Clean single-instrument design with correct funding-data plumbing (funding is credited by the engine); 557 trades is a thorough test

Catastrophically unprofitable: Sharpe -5.48 (CI fully negative), profit factor 0.70, negative expectancy (-$81/trade), total return -41%, Sortino -6.52, PSR 0.0

Mechanism falsified: shorting high-positive-funding SOL = shorting strong uptrends; SOL's 2021 bull run ran the shorts over (2021 -41.1%)

Short-only momentum-fade concentrated in frothy bull periods (exposure 5.7%, trades cluster 2020-2021) — the worst regime for shorts; funding received nowhere near offsets directional loss

16 max consecutive losses; this is the known 'fade the funding spike' failure pattern (prior BTC funding-contrarian also failed)

SolExtremeFundingContrarianShort1H tried to fade over-leveraged longs by shorting SOL when 8h funding spiked extreme-positive, expecting funding income plus mean reversion. The mechanism was cleanly implemented and thoroughly tested (557 trades, funding credited by the engine), but it was a structural short-only momentum-fade concentrated in SOL's 2021 parabolic bull — and the shorts were run over, producing a Sharpe of -5.48, profit factor 0.70, and -41% return. The analyst judged this an absence of edge in the mechanism itself (shorting crypto's structural up-momentum), matching the known funding-fade failure pattern, and abandoned it at the backtest-review gate. It ended after a single iteration as abandoned, never reaching optimization, analysis, or risk review.

Short-only contrarian fade of extreme positive funding on a crypto perp has strongly negative edge: high funding marks strong up-momentum, so fading it is dominated by continuation and concentrated in bull periods (the worst regime for shorts), and the funding credit never compensates the directional loss — the same failure as the prior BTC funding-contrarian short.

It was abandoned at the pre-optimization backtest-review gate (verdict: abandon) on a falsified mechanism — high positive funding marks strong uptrends, so shorting it is shorting momentum and gets run over, with the small funding credit nowhere near offsetting the directional loss — and with funding correctly credited by the engine (a fair test), optimization and all later stages were never reached.

A short-only contrarian funding-fade strategy on SOLUSDT 1-hour perpetual futures (OHLCV + funding-rate data) that shorted when the 8-hour funding rate spiked extremely positive (≥+0.05%/8h, scaling risk up above +0.08%) — betting crowded longs mark local tops — to capture both the funding received and mean-reversion as longs unwound, exiting on a 1.5% stop, 2% target, funding normalization, or a 24-bar time stop.

Over ~5.3 years (2020-09-14 to 2026-05-17) it took 557 short trades with a 43% win rate and was catastrophically unprofitable: total return -41%, CAGR -13.4%, Sharpe -5.48 (CI [-7.52, -3.16]), profit factor 0.70, expectancy -$81/trade, PSR 0.0, a 42.9% max drawdown, 16 max consecutive losses, with trades clustered in the frothy 2020-2021 bull (2021 -41.1%).