BtcSpotPerpFundingCarryDeltaNeutral1H

BTC Spot-Perp Funding Carry: Long Spot Hedged Short Perp When 8H Funding > 0.04%

A delta-neutral cash-and-carry strategy that collects perpetual funding payments. When BTCUSDT perpetual funding rate exceeds a high threshold (longs aggressively paying shorts), simultaneously go long BTCUSDT spot and short BTCUSDT perpetual at matching notional, harvesting the funding cash flow while remaining directionally hedged. This is a CROSS-VENUE (BINANCE_SPOT + BINANCE futures), MARKET-NEUTRAL (long+short), MULTI-DAY HOLDING strategy that fills four heavily under-represented portfolio buckets simultaneously: spot venue (6.4% vs 15% target), cross-venue (1.1% vs 15% target), long-short direction (13.3% vs 45% target), and zero-options-but-options-not-required. Critically, the edge is a MECHANICAL CASH FLOW — when funding is 0.04%/8h, the perp short receives 0.12%/day = ~3.6%/month from the long side. This is NOT a price-pattern strategy (which has failed 10+ times this session on majors). It does not depend on BTC's intraday or daily microstructure remaining stable across regimes — the 2024-2026 regime decay that wiped out four BTC OHLCV strategies has NO bearing on a funding-driven payment stream, because positive funding spikes occur precisely when retail leverage gets crowded long (a recurring market phenomenon that has actually INTENSIFIED in 2024-2026 with the rise of perpetual-DEX flow). Position sizing: 25% of equity in spot leg, 25% notional in short perp leg (1x leverage on futures account = ~50% margin utilization for both legs combined, plenty of headroom). Risk per trade: 2% via combined basis-widening + funding-flip scenarios.

Implements the hypothesis directly: the decision variable is the perp 8h funding rate (read from supplementary_data['funding_rates'], a flat {ns:rate} 8h series), returned continuously each bar from calculate_signal and thresholded in should_enter. Primary leg is the perp (BINANCE futures, MARGIN) so get_account_equity reads the margin account sizing the short; the spot hedge (BINANCE_SPOT, CASH, long-only) mirrors equal dollar notional via _submit_entry_instrument, yielding a delta-neutral book. Spot+perp share the 1-HOUR timeframe so the base cross-leg alignment barrier uses contemporaneous prices for sizing. per_leg_pct=0.25 puts ~25% equity per leg (~50% gross, 1x leverage, ample headroom). Funding governs entry/exit; if funding data is missing the strategy stays flat with no price-only fallback. Fills the spot, cross-venue, and long-short portfolio buckets simultaneously.

Structural regime-decay of the funding cash flow on BTC — not a code or measurement problem. Funding IS credited in this backtest (+14,598 gross, confirmed in trade-level 'funding' fields), and the hedge is genuinely delta-neutral (net price PnL -5,690). But the credited result is net -2.1% over 6.5 years: gross funding is fully consumed by commission (5,196) + impact (1,427) + price drift. Decisively, 91% of funding and 89% of trades came from 2020-2021, with ZERO qualifying trades in 2022, 2025, and 2026 — BTC perp funding above the cost-clearing 0.04%/8h level structurally stopped occurring after the 2021 retail-leverage cycle, empirically refuting the hypothesis's claim that funding spikes 'intensified in 2024-2026.' Threshold optimization cannot manufacture funding the market isn't paying: lowering entry_funding only adds thinner-funding trades that lose harder to the ~0.30% two-leg round-trip cost, and raising it kills the already-thin 66-trade count. No robust parameter region exists, so optimization would overfit the 2020-2021 window and fail walk-forward OOS / holdout (both fall in the post-2021 no-funding regime). FAILURE PATTERN: BTC spot-perp funding carry is a 2020-2021 fossil — the carry only cleared costs during the early high-leverage era; on current data the credited funding nets negative after fees, and the funding signal itself rarely fires post-2021. NOTE for Research Lead: the carry MECHANISM is sound and the engine credits funding correctly; if the family is pursued, redirect to a venue/instrument with PERSISTENT positive funding skew (e.g. a high-funding alt perp) rather than BTC, where the basis/funding has decayed to ~0.

Delta-neutral BTC cash-and-carry that harvests perpetual funding. When BTCUSDT perp 8h funding is richly positive (>=0.04%/8h), it SHORTS the BTCUSDT perp on BINANCE USD-M futures and LONGS BTCUSDT spot on BINANCE_SPOT at equal dollar notional, staying delta-neutral on price while the short perp collects funding. Both legs are flattened when funding decays toward neutral (<=0.01%/8h after a minimum carry hold), on a ~10-day time stop, or on a 2%-of-equity combined-PnL stop if the basis dislocates. One-directional by design (CASH spot cannot be shorted), so it only acts on high positive funding.

Mechanism is correctly implemented and matches the hypothesis: delta-neutral long-spot/short-perp, net price PnL only -5,690 against per-leg swings, and funding IS credited (+14,598 gross over the period).

Risk claim holds: equity ran 200,000 -> 195,738 (~2% band), max drawdown ~2.1% — genuinely delta-neutral as designed.

Net edge is NEGATIVE after the credited funding: gross funding +14,598 is fully consumed by commission (5,196) + market impact (1,427) + residual price drift (-5,690), leaving total_return -2.1% over 6.5 years.

Structural regime decay of the underlying cash flow: 91% of all funding and 89% of trades occurred in 2020-2021; ZERO qualifying trades in 2022, 2025, 2026 and only 7 pairs total across 2023-2024. Rich BTC funding (>0.04%/8h) no longer occurs — directly refuting the hypothesis's 'funding intensified in 2024-2026' premise.

Headline Sharpe 5.07 / Sortino 9.58 / PSR 1.0 are artifacts of per-leg trade accounting (132 leg-trades = 66 pairs) on a near-flat equity curve, not a deployable edge.

Only 66 entries over 6.5 years and exposure 25% — too thin and too concentrated in one early regime for meaningful parameter optimization.

commission 5196 + impact 1427; plus -5690 price drift

percent units; net negative AFTER credited funding

credited, but 91% from 2020-2021

rich funding regime absent post-2021

BtcSpotPerpFundingCarryDeltaNeutral1H ran a textbook long-spot/short-perp cash-and-carry, entering when BTC perp 8h funding was richly positive to harvest the payment stream while staying delta-neutral. The implementation was correct and the engine credited funding (+$14,598 gross) on a tightly hedged ~2% equity band, but fees, impact, and price drift fully consumed it, leaving -2.1% over 6.5 years — and decisively, 91% of funding came from 2020-2021 with zero qualifying trades in 2022, 2025, or 2026, refuting the premise that funding had intensified recently. The analyst ruled it structural regime decay rather than a code or tuning problem, noting the flattering Sharpe was a per-leg-accounting artifact and recommending the carry family be redirected to a persistently high-funding altcoin. It ended after one iteration as abandoned, never advancing to optimization or risk review.

The cash-and-carry mechanism is sound and funding is credited correctly by the engine, but BTC is a dead target: cost-clearing positive funding is a 2020-2021 fossil that no longer fires, and threshold tuning cannot manufacture funding the market is not paying — pursue the family only on a venue/instrument with persistent positive funding skew (e.g., a high-funding altcoin perp).

It was abandoned at the pre-optimization backtest-review gate (verdict: abandon) due to structural regime decay of the funding cash flow itself — rich BTC funding above the cost-clearing 0.04%/8h level stopped occurring after the 2021 retail-leverage cycle, empirically refuting the hypothesis's 'funding intensified in 2024-2026' claim — so with no robust parameter region and a 66-trade sample concentrated in one early regime, optimization and all later stages were never reached.

A delta-neutral cash-and-carry funding harvest on BTC — when BTCUSDT perp 8h funding exceeded +0.04% it simultaneously longed BTCUSDT spot (BINANCE_SPOT) and shorted the BTCUSDT perp (BINANCE USD-M) at equal dollar notional to collect the funding cash flow while staying price-hedged, flattening on funding decay, a time stop, or a basis-blowout stop, to fill the spot, cross-venue, and long_short buckets.

Over 6.5 years and 66 entries (132 leg-trades) it stayed genuinely delta-neutral (equity ~200k→195.7k, max drawdown ~2.1%, net price PnL -$5,690) with funding properly credited (+$14,598 gross), but the net edge was negative: total return -2.1% after $5,196 commission, $1,427 impact, and price drift consumed all the funding, with 91% of funding and 89% of trades concentrated in 2020-2021 and zero qualifying trades in 2022, 2025, and 2026; the headline Sharpe 5.07 / Sortino 9.58 / PSR 1.0 are per-leg-accounting artifacts on a near-flat curve.