EthSpotSma200RegimeLong

ETH Spot Long-Hold with 200-SMA Regime Filter (Daily, Pure Regime-Flat Exit)

A long-only single-instrument PURE regime-conditional buy-and-hold strategy on ETHUSDT spot (BINANCE_SPOT) using daily bars and OHLCV-only data. This is the SIMPLEST possible strategy in the entire pipeline — just one tunable parameter (the SMA period) and a binary regime decision: hold spot ETH when above 200-day SMA, hold cash when below. No ATR trail. No drawdown threshold. No EMA crossover events. No volume confirmation. No multi-condition AND-gates. Just a single dominant filter: the 200-day SMA regime check. This proposal explicitly addresses the recently-documented architecture-mirror failure pattern (6 failures this session). Rather than mirror an existing architecture, this proposes a structurally NEW mechanism not yet present in the pipeline: pure binary regime-flat exit (versus all existing strategies which use ATR trails, drawdown thresholds, EMA crossover events, or volume breakouts as exit triggers). This is critical because: (a) the failed ETH Spot Drawdown showed that ETH spot mean-reversion mechanics don't work — the 30-day drawdown trigger + 5% trail is too complex for ETH's volatility, (b) but ETH Golden Cross perp works at Sharpe 4.04 — proving ETH responds to simple regime-based entry/exit on daily timeframe. This proposal extracts the ESSENCE of what works on ETH (regime filter) and discards the volatility-sensitive trailing-stop machinery that doesn't generalize. The asset selection is justified by ETH-SPECIFIC empirical evidence: ETH has TWO working perp strategies (Golden Cross Sharpe 4.04, 4H Volume Breakout pending) showing ETH responds cleanly to daily regime filtering. The mechanism selection is justified by the explicit failure mode of ETH Spot Drawdown — by REMOVING the ATR trail and drawdown threshold (which were the failure points), and KEEPING only the 200-SMA filter (which is the proven element). Existing ETH strategies use ATR-based exits or EMA crossovers; this uses a single binary regime check. Genuinely orthogonal exit mechanism. Single parameter (200-SMA period) means no overfitting risk and clean walk-forward sensitivity. Calibrated for ~3-8 entries per market cycle (very low frequency, very high per-position holding period).

Extracts the proven element (daily regime filtering, which works for ETH per the Golden Cross perp at Sharpe 4.04) and discards the volatility-sensitive trailing-stop / drawdown machinery that caused the ETH Spot Drawdown failure. calculate_signal() returns the continuous % distance of close from the SMA, so the decision variable varies every bar and is returned in natural units; the binary regime decision lives in should_enter() (buy when distance > 0) and should_exit() (flat to cash when distance < 0), giving a genuinely orthogonal pure regime-flat exit mechanism not present elsewhere in the pipeline. Being long-only with no leverage and no funding edge, the correct venue is BINANCE_SPOT (CASH account, leverage hard-capped at 1.0); leverage is left at 1.0 and intentionally not referenced in sizing to avoid the leverage_set_but_unused gate. Only SELL-to-exit is used (never SELL-to-enter), respecting the spot CASH constraint. A single parameter (sma_period=200) means minimal overfitting surface and clean walk-forward sensitivity, and the very low turnover (~3-8 regime flips per cycle) keeps the ~0.20% spot round-trip fee negligible against multi-month holding periods.

Structurally uncompetitive beta play that single-parameter optimization cannot fix. The 200-SMA regime-flat long carries a 71.9% max drawdown (CI to 89%) — far above any promotion threshold — because the slow SMA exit cannot react to ETH's fast sub-SMA crashes; the lone tunable (SMA period) only trades whipsaw against drawdown and cannot bring risk near promotable levels. The headline Sharpe 3.19 is an outlier artifact, not a risk-adjusted edge: sharpe_ci_low is -3.33 (CI [-3.33, 6.35] straddles 0), annualized_volatility is 2607%, and the entire record hinges on a single 2020-2021 bull-cycle hold (2021 +828%, one trade ~+850%) with 2018 -41% and 2020 -47% otherwise. It is regime-filtered ETH buy-and-hold — beta, not differentiated alpha (CAGR 33.8% vs buy-hold ~27%). This is not an iterate case: the only fixable lever is the drawdown, which would require the ATR-trail / drawdown-stop exit machinery the hypothesis explicitly rejects as its entire reason for existing, so fixing it contradicts the premise rather than refining it. With one parameter, the 2-hour walk-forward has nothing meaningful to tune. FAILURE PATTERN: single-parameter SMA-regime buy-and-hold on spot majors produces a high but outlier-driven Sharpe (one bull-cycle trade) with a 70%+ drawdown and a Sharpe CI straddling 0 — it is beta dressed as alpha and the lone SMA knob cannot reduce the drawdown to promotable levels.

The simplest strategy in the pipeline: a long-only, single-instrument, pure regime-conditional buy-and-hold on ETHUSDT spot (BINANCE_SPOT) using daily bars and OHLCV only. It holds spot ETH whenever the daily close is above the 200-day SMA and holds cash whenever the close is below it. There is no ATR trail, no drawdown threshold, no EMA crossover event, and no volume confirmation — just a single binary regime decision driven by one tunable parameter (the SMA period). Deploys 95% of equity when in the bullish regime.

Simplest possible design, faithfully implemented; single parameter means genuinely low overfitting risk

Regime filter does improve on pure ETH buy-and-hold (CAGR ~33.8% vs ~27%, max DD 72% vs ~94%)

max_drawdown 71.9% (CI to 89%) — far above any promotion threshold; the slow 200-SMA exit structurally cannot avoid ETH's fast sub-SMA crashes, and the single SMA parameter cannot fix it

Sharpe 3.19 is an outlier artifact: sharpe_ci_low -3.33 (CI straddles 0 massively), annualized_volatility 2607%, return_skew 4.78 — entire result hinges on one 2020-2021 bull-cycle trade (2021 +828%)

Fundamentally ETH beta, not differentiated alpha — regime-filtered buy-and-hold

Only 27 trades over 9 years; with one parameter the 2-hour optimization has almost nothing to tune and cannot address the drawdown or significance

Negative in 2018 (-41%) and 2020 (-47%); strip 2021 and there is no edge

EthSpotSma200RegimeLong distilled ETH down to its simplest working element — a single 200-day SMA regime filter — deliberately discarding the trailing-stop and drawdown machinery blamed for the failed ETH Spot Drawdown strategy. Faithfully implemented with just one tunable parameter, it posted an eye-catching +834% and Sharpe 3.19, but the analyst found this to be regime-filtered ETH beta rather than differentiated alpha: a 71.9% max drawdown, a Sharpe CI straddling zero, and a record resting entirely on the 2020-2021 bull cycle with losing 2018 and 2020 years. Judging it non-iterable — the only fixable lever (drawdown) demanded the very exit machinery the premise rejected, and one parameter gave optimization nothing to tune — the analyst abandoned it at the backtest-review gate. It ended after a single iteration as abandoned, never reaching optimization, analysis, or risk review.

Single-parameter SMA-regime buy-and-hold on spot majors is beta dressed as alpha: it produces a high but outlier-driven Sharpe carried by one bull-cycle trade, a 70%+ drawdown, and a Sharpe CI straddling zero — and the only fixable lever (drawdown) would require exactly the ATR-trail/drawdown-stop machinery this hypothesis was built to reject, so fixing it contradicts the premise.

It was abandoned at the pre-optimization backtest-review gate (verdict: abandon) as a structurally uncompetitive beta play — failing the max_drawdown, total_trades, and sharpe_ci_low criteria — because the slow 200-SMA exit cannot react to ETH's fast sub-SMA crashes and the lone SMA parameter cannot bring risk to promotable levels, so optimization and all later stages were never reached.

The simplest possible strategy in the pipeline: a long-only single-parameter regime-conditional buy-and-hold on ETHUSDT.BINANCE_SPOT daily bars (OHLCV only) that held spot ETH while price was above its 200-day SMA and moved fully to cash when it fell below — a pure binary regime-flat exit with no ATR trail, drawdown threshold, crossover event, or volume gate.

Over ~9 years (2017-08-17 to 2026-05-18) it took only 27 trades with a 30% win rate: total return +834%, CAGR 33.8%, headline Sharpe 3.19 (CI [-3.33, 6.35]), profit factor 4.08, but an uninvestable 71.9% max drawdown (CI to 89%) and 2607% annualized vol — with the entire record hinging on a single 2020-2021 bull-cycle hold (2021 +828%) and negative 2018 (-41%) and 2020 (-47%).