LTC Spot-Cash Band Rebalancing with Re-arming Sleeve-Drawdown Protection

Revision of 'LTC Spot-Cash Threshold Rebalancing in a Capped Capital Sleeve'. Instrument, timeframe, long-only setup, 40% sleeve, 50% target weight and the weight band are unchanged; ONLY the capital protection is re-specified. On LTCUSDT.BINANCE_SPOT 4-HOUR bars, a sleeve initially allocated 40% of account equity holds LTC and USDT at a 50/50 target (initial buy: LTC worth 20% of equity at the first post-warm-up bar close). The other 60% of equity is an inactive USDT reserve that never trades. Sleeve value = LTC holding value + sleeve USDT (tracked internally; the sleeve's cash is the starting 40% allocation plus/minus sleeve trade proceeds and fees). At each 4H bar close, LTC weight w = LTC value / sleeve value. If |w - 0.5| >= band_width, trade LTC back to exactly 50% of the sleeve value (buy when w is below, partial sell when above). The average-entry inventory stop of the previous version is REMOVED: a rebalancer adds to losers by design, so loss vs cost basis is the wrong risk measure (it fired at the Dec-2017 top and halted trading for 5.4 years). Protection is now at the sleeve level: (1) hard cap: LTC holding never exceeds 30% of total account equity — a buy is truncated to the cap, and if price appreciation pushes LTC above 30% of equity it is trimmed to 30% at the bar close; (2) sleeve-drawdown stop: track the sleeve value's high-water mark; when sleeve value falls sleeve_dd_stop_pct below its HWM at a bar close, sell all LTC into sleeve cash (market order at that bar's close) and enter a PAUSED state; (3) re-arm: the pause ends at the first bar close where EITHER cooldown_bars bars have elapsed since the stop OR the LTC close has recovered rearm_recovery_pct from its lowest close during the pause; on re-arm, buy LTC back to 50% of the (now all-cash) sleeve and reset the sleeve HWM to the current sleeve value. There is never a permanent halt. Execution: all orders are market orders at the signal bar's close (next_bar_open_fill substituted by the platform's 24/7 equivalent: the 4H close equals the next bar's open). No leverage, no shorting, no supplementary data, no assumed cash yield. Risk control route (b): signal-based exits (rebalancing, sleeve-drawdown stop) with capped sizing — the single LTC position is at most 30% of equity, far below 100%. Fee arithmetic: a rebalance only fires after LTC moves enough to shift the weight by band_width; with band_width 0.10, w goes 0.50->0.60 when LTC rises ~50% (or falls ~33% for 0.40), so each rebalance trades ~10-20% of the sleeve after a multi-tens-of-percent move; the volatility-harvesting premium of a 50/50 rebalance relative to buy-and-hold over a ±40% oscillation is roughly 1-3% of the sleeve per round trip, versus 0.10% per spot fill on the traded slice (~0.02% of sleeve). Expected rebalances: LTC 4H realized vol ~80-100%/yr implies roughly 4-10 band crossings per year at 0.10, i.e. ~30-70 rebalance fills over 2018-2025.

The analyst found the code correct but the protection contract incompatible with the mechanism: the frozen 25% average-entry inventory stop fired on the first position (bought at the Dec-2017 top) and the permanent halt left 1 trade in 5.4 years, so the band mechanism was never exercised. Following the analyst's concrete reframe, I changed ONLY the protection: (1) removed the average-entry stop (a rebalancer adds to losers by design); (2) protection is now sleeve-level — the existing 30%-of-equity LTC cap plus a sleeve-drawdown stop from the sleeve's high-water mark; (3) every halt re-arms after cooldown_bars or a rearm_recovery_pct bounce from the pause low, re-entering at the 50/50 target with a reset HWM; (4) risk control route (b): signal exits + single position capped at 30% of equity (<100%). Instrument (LTCUSDT spot), 4H bars, long-only, 40% sleeve, 50% target and band stay as the analyst instructed, so the premise (band rebalancing harvests LTC range volatility after fees) is finally tested across 2018-2025. LTC is uncrowded in the corpus and spot is an under-represented, higher-survival venue; no supplementary feeds needed. The sleeve stop range is set wide (25-60% sleeve drawdown, i.e. roughly 45-90% LTC decline from the sleeve peak given 50% weight) so it acts as disaster protection rather than firing on ordinary range swings.

premise_falsified. The failed gate is the enforced year-consistency floor: the final parameters are positive in only 5 of 9 calendar years (56% < 60%). This is a hard failure, so neither promotion nor labelled paper observation is allowed. Drawdown also failed: 32.5% vs the pre-registered 25% (Calmar 0.21). Fees are not the cause: commission is about 1% of gross, and per-cycle figures are -6.6% at the contract defaults and +22.2% for the final parameters, against a 0.25% spot floor (these per-cycle numbers are not fee-comparable for a rebalancer). PF is 0.71 at the defaults and 3.0 for the final parameters. The volatility-harvesting premise is measurably absent. With the band now exercised (61 fills over 2018-2026), the default rebalancer returned -14.1% against about -15.9% for a static 20% LTC hold (LTC buy-and-hold: -79.4%). That is about +0.2%/yr of rebalancing benefit, where the hypothesis predicted several percent. Benchmark correlation is 0.93 and alpha is negative. The final parameters' +78% comes from sleeve-stop and re-entry timing fitted on the development data, with 2020 contributing 44.5% of the profit. Every sensitivity point stays within one Sharpe standard error of the baseline (SE 0.44), and the holdout is inconclusive (Sharpe 0.18 on 4 cycles, t 0.32). Insufficient significance is not the reason for this verdict; the measured absence of the harvesting premium and the hard year-consistency failure are. This verdict covers this operationalization: LTC/USDT single-asset vs cash band rebalancing in a 40% sleeve on Binance spot 4-hour bars, with sleeve-drawdown protection.

This implements the revised protection contract from scratch. The average-entry stop is gone; protection is at the sleeve level (30%-of-equity LTC cap plus the sleeve HWM drawdown stop with re-arm by cooldown or recovery). Risk route (b): rebalancing and stop exits, with the single LTC position at most 30% of equity and leverage 1. All state is kept in scalar attributes: reserve, HWM, pause flag, pause start ns, pause-low close. The cooldown is measured from bar timestamps, not a process counter. Parameter names and defaults follow the optimization plan; the fixed ones are target_weight, sleeve_fraction, min_notional_usd and max_ltc_equity_frac. Nothing is clamped in code, so no _param_bounds. Caveat for the analyst: this is a NETTING account and the sleeve rarely goes flat (only on a sleeve stop), so the positions report books the whole period between stops as ONE trade. Expect total_trades to be very low (single digits) even with dozens of rebalance fills, and expect avg_trade_return_pct and win rate to describe that coarse denominator (lesson 175). Judge the run on fill count and the equity curve, not on the min_trades=30 criterion as the positions ledger counts it.

The fixed 30%-of-equity LTC cap turns the strategy into a per-bar churner whenever the sleeve has grown. I replicated the code's logic in plain Python on the real LTCUSDT 4H closes. On the sandbox-length window (last 2,400 days, roughly 2020-2026) it makes about 960 fills, close to the sandbox's 973 trades. 885 of those fills are cap trims and 71 are band buys, against the 4-10 band crossings a year the hypothesis expected. Here is how it happens. After the 2021 rally the sleeve grows past about 0.9x initial equity. From then on 50% of the sleeve is more than 30% of equity, so LTC sits at the cap. Every up-close above the cap is trimmed and every down-close is topped back up to the cap, and each of these fills only has to clear the $5 minimum. In that regime the strategy is not harvesting band volatility. It is a constant 30%-of-equity LTC holder that pays 0.10% taker plus spread on many small slices, so the hypothesis's fee arithmetic (one ~0.02%-of-sleeve fill per multi-tens-of-percent move) does not hold there. On the full 2017-12 to 2026 history the same simulation gives about 60 fills, inside the predicted 30-70, with only 1 cap fill: the first buy lands at the Dec-2017 top and the sleeve never grows enough to hit the cap. Walk-forward windows that start after 2020 will look like the sandbox regime. Check first: fills by kind (cap trim vs band rebalance) per walk-forward window, and total commission plus spread as a share of gross PnL. If cap churn dominates, a tolerance band on the cap (for example trim only above 0.30 + band_width x 0.3) would be a Research Lead amendment, because max_ltc_equity_frac and its literal 'trim to 30% at the close' rule are fixed by the contract. Ledger note (lesson 133): this is a NETTING account and the sleeve rarely goes flat, so positions-report trade counts and avg_trade_return_pct describe stop-to-stop holding periods, not rebalance fills. Judge the run on fills and the equity curve.

The Research Lead REVISED this hypothesis because the previous premise did not work. Implement the REVISED hypothesis (see the hypothesis description) from scratch — do not reuse the prior strategy. Revision rationale: The analyst found the code correct but the protection contract incompatible with the mechanism: the frozen 25% average-entry inventory stop fired on the first position (bought at the Dec-2017 top) and the permanent halt left 1 trade in 5.4 years, so the band mechanism was never exercised. Following the analyst's concrete reframe, I changed ONLY the protection: (1) removed the average-entry stop (a rebalancer adds to losers by design); (2) protection is now sleeve-level — the existing 30%-of-equity LTC cap plus a sleeve-drawdown stop from the sleeve's high-water mark; (3) every halt re-arms after cooldown_bars or a rearm_recovery_pct bounce from the pause low, re-entering at the 50/50 target with a reset HWM; (4) risk control route (b): signal exits + single position capped at 30% of equity (<100%). Instrument (LTCUSDT spot), 4H bars, long-only, 40% sleeve, 50% target and band stay as the analyst instructed, so the premise (band rebalancing harvests LTC range volatility after fees) is finally tested across 2018-2025. LTC is uncrowded in the corpus and spot is an under-represented, higher-survival venue; no supplementary feeds needed. The sleeve stop range is set wide (25-60% sleeve drawdown, i.e. roughly 45-90% LTC decline from the sleeve peak given 50% weight) so it acts as disaster protection rather than firing on ordinary range swings.