XRP Net-Directional Time-Series Momentum (Single-Instrument USD-M, Long-Short, 4H, Discrete Trend-Flip Entries, Capital-Capped, Pure OHLCV)

A LONG-SHORT, SINGLE-INSTRUMENT net-directional TIME-SERIES MOMENTUM strategy on XRPUSDT.BINANCE USD-M perpetual (4H bars). It applies the ONE validated, most-replicated momentum factor -- an asset's own trailing-return trend predicting its near-future return -- to the MOST-DECORRELATED liquid major NOT yet covered by the portfolio's TSMs (which are all on the tightly-correlated L1/smart-contract/exchange majors BTC/ETH/SOL/BNB). RATIONALE FOR ASSET + HORIZON: (1) XRP is top-5 liquidity with full clean history (since 2020 on Binance USD-M) and a DISTINCT idiosyncratic driver -- payment-rail adoption and the SEC legal saga -- so its largest trends (2021 run, 2023 SEC-ruling spike, 2024-25 rally) are LARGELY UNCORRELATED with the L1 majors, adding a genuine diversifying return stream rather than re-expressing BTC beta. (2) The prior XRP *breakout* hypothesis died of EVENT-SPARSITY (a sparse daily breakout+volume conjunction -> OOS-all-zero); a DENSE 4H TIME-SERIES-MOMENTUM signal (continuous trend sign evaluated every bar) structurally avoids that -- it fires in every window and populates the walk-forward. (3) LONG-SHORT fills the under-represented direction bucket (13.1%). This is NOT a closed family: the analyst's verdict this session is explicit -- 'reversion as a family is dead in this engine; redirect to the validated net-directional momentum/trend edge' -- and single-asset TSM is exactly that (baskets, funding-confirmed clones, and dollar-neutral are the closed variants, NOT single-asset net-directional TSM). EVERY design choice respects this session's hard engine walls: (a) SINGLE-INSTRUMENT -- NOT a multi-leg basket, whose summed gross exposure proved structurally UNCAPPABLE in this engine. (b) DISCRETE full-size enter/exit on trend flips -- NOT continuous per-bar rebalancing, which produced the 244%-vs-cap notional-inflation artifact on both spot and perp. (c) NET-DIRECTIONAL plain price-TSM -- more robust than the funding-confirmed version that failed on alts: it goes net-SHORT in downtrends (real crash protection) rather than long-biased bull-beta. (d) PURE OHLCV -- no options/tick/funding plumbing (options engine-unwired -> zero trades; funding data ends 2024); guaranteed to trade in the Layer-3 sandbox. (e) 4H dense triggers -- avoids the daily single-asset event-sparsity that trade-starved breakout/RSI/candle walk-forwards to OOS-all-zero. (f) CAPITAL-CAPPED against STABLE realized equity (excl. unrealized PnL), hard-capped <=100%. Per the brief's guidance to favor simpler strategies, intentionally minimal: 3 tunables (trend_lookback, entry_band, stop_atr_mult) with robust defaults to avoid a trend_lookback cliff.

Implements the validated net-directional time-series momentum edge on the most-decorrelated liquid major (XRP, payment-rail/legal driver) using the standard FactoryStrategy hooks for a clean single-instrument long-short. calculate_signal returns the continuous trailing return so Layer-2 frozen-signal detection passes, while the discrete entry decision lives in should_enter with an entry_band deadband (|trend| >= 1%) to avoid the near-zero whipsaw of pure-sign TSM; the exit uses a zero-cross sign flip so trends run until they decisively reverse (hysteresis between entry_band and 0), with a wide 4x-ATR catastrophe stop as a backstop (entry price/ATR captured on entry, reconstructed from avg_px_open if state is lost). Every documented engine wall is respected: SINGLE-INSTRUMENT (one position at a time -> gross is one leg's notional, avoiding the multi-leg-basket exposure artifact that has corrupted four baskets this session), DISCRETE full-size enter/exit (no continuous per-bar rebalancing -> no notional inflation), CAPITAL-CAPPED against stable realized equity (get_account_equity reads the USDT balance excluding unrealized PnL, so sizing cannot balloon mid-trade) hard-capped at position_pct (95%) of equity at leverage 1.0, and PURE OHLCV (no funding/OI/tick plumbing -> guaranteed to trade in the Layer-3 sandbox and immune to the data-coverage and options-unwired kills). The 4H dense trend signal fires every bar, populating walk-forward OOS windows and structurally avoiding the daily single-asset event-sparsity that trade-starved the prior XRP breakout to OOS-all-zero. Sizing clears the XRPUSDT $5 min-notional at size precision 1 (0.1 step). leverage stays 1.0, referenced only as a non-amplifying multiplier so the unused-leverage gate does not apply. Three robust-default tunables keep the optimization surface small to avoid a trend_lookback cliff.

Identical failure to the just-abandoned BNB sibling, from the same single-asset discrete-entry TSM template. Two independent fatal problems, each sufficient. (1) CATASTROPHIC OVER-LEVERAGE + LIQUIDATION: despite the docstring's 'single-instrument -> exposure structurally cappable, discrete entries -> no notional inflation, capital-capped <=100%,' the backtest shows avg_position_pct 151.9% (~1.6x the intended 95%), physically-impossible daily returns (+1236.8% on 2022-06-14, -528.3% on 2022-06-22, +332.2%, -279.0%) that a single XRP position at <=95% sizing cannot produce, return_kurtosis 1666, annualized_volatility 442%, and a full LIQUIDATION on 2022-08-24 (total_return -100%, max_drawdown 100%, 2023-2026 empty). This is the SAME notional-inflation/stacking bug as the BNB run: discrete TSM that re-arms should_enter while a position is open accumulates size, so the hypothesis's claim that single-instrument + discrete entries avoid the basket exposure artifact is empirically FALSE. (2) THE SIGNAL LOSES ANYWAY: profit_factor 0.90 (<1.0), expectancy -$149/trade — naked single-asset XRP time-series momentum has negative gross expectancy and is the explicitly regime-dead naked-trend family. Not optimize: a return series dominated by impossible bars and ending in liquidation cannot be optimized. Not iterate: fixing the sizing/stacking bug would only surface a PF-0.90 money-loser in a dead family (the BNB sibling already established this is not worth a cycle; XRP's PF is no better). FAILURE PATTERN: the single-asset discrete-entry net-directional TSM template (BnbNetDirectional*, XrpFourHourTimeSeriesMomentum*) reproduces the multi-leg-basket notional-inflation bug on a SINGLE instrument (avg_position 150%+ vs a claimed 100% cap, impossible >1000% daily prints, kurtosis >1000, liquidation) because discrete entries re-fire while a position is open; and the underlying naked single-asset TSM has profit_factor < 1.0. Two consecutive liquidations (BNB, XRP) from this template confirm it is a broken implementation of a dead signal, not an asset-selection issue. A 'capital-capped <=100%' docstring claim is not evidence the cap holds -- verify realized avg_position_pct; here it was 1.5x over and the book liquidated.