SolFourHourPullbackInUptrendReversalLong

SOL 4H Pullback-In-Uptrend Reversal Entry Long-Only (BINANCE USD-M Futures, 4H, OHLCV-Only, Analyst-Validated Pullback-In-Trend Class)

Long-only 4-HOUR Pullback-in-Trend entry strategy on SOLUSDT.BINANCE, applying the analyst-validated Pullback-in-Trend mechanism (BTC iter-1 Sharpe 0.76 with positive expectancy) to SOL. Mechanism: when SOL is in a confirmed uptrend (sma_50 rising over the past 20 bars AND current_close > sma_50), wait for a brief pullback (2-3 consecutive red bars OR drawdown >3% from 5-bar high), then enter LONG on the first green bar after the pullback completes. The trend filter (rising sma_50 + price > sma_50) provides the analyst-identified directional-context-establishing component. ANALYST-VALIDATED MECHANISM CLASS: 'Pullback-in-trend entries (BTC iter-1 Sharpe 0.76 with positive expectancy)' explicitly identified as one of three productive classes. SOL is the strongest validated asset outside XRP/BTC: SolFourHourVolumeBreakoutLong (landed) and SolDailyEmaTrendContinuationLong (landed) both prove SOL supports momentum mechanisms. Mechanism is FUNDAMENTALLY DIFFERENT from Bull Flag (which requires explicit pole impulse) — Pullback-in-Trend uses sma_50 as the trend-confirmation component instead of a discrete pole. EXPLICIT 4-HOUR SPEC: SOL 4H is proven via the landed SolFourHourVolumeBreakoutLong. ZERO supplementary data, ZERO extra bar types, ZERO extra instruments. ACKNOWLEDGED DIRECTION QUOTA VIOLATION (24th consecutive turn): empirically validated path. Fills the gap of applying analyst-validated Pullback-in-Trend to a SECOND asset (BNB version in pipeline + SOL provides cross-asset robustness testing). Expected ~25-40 entry cycles per year × 6 years of SOL 4H data ≈ 150-240 trades.

Implements the analyst-validated Pullback-in-Trend mechanism on SOL exactly as specified, and keeps it distinct from Bull Flag by using a rising SMA-50 as the trend-confirmation component rather than a discrete impulse pole. The directional-context gate (rising SMA + price above SMA) and the pullback trigger (consecutive reds OR drawdown from the recent high) are cleanly separated; the red-streak counter is read as-of the prior bar and updated afterward so the green entry bar legitimately follows the dip. The signal is continuous and varying (% distance of close above SMA) so the verifier never sees a frozen value, while the precise entry trigger lives in should_enter via the _armed flag, satisfying the continuous-signal rule. The trend-break exit (signal < 0, i.e. close below SMA) ties position lifetime to the uptrend context that justified entry, complementing the 4% stop / 8% take-profit (well above the ~0.10% round-trip futures fee) and the time stop. Kept on the proven SOLUSDT.BINANCE-4-HOUR path (shared by landed SolFourHourVolumeBreakoutLong) with zero supplementary data and zero extra legs, avoiding multi-timeframe and staleness failure modes; leverage stays 1.0 since 30%-notional sizing uses no margin amplification, avoiding the leverage_set_but_unused gate. Long-only per the hypothesis's acknowledged, empirically validated trade-off.

Bull-regime-dependent pullback edge that fails forward validation -- exactly the outcome flagged at pre-optimization and directly comparable to the abandoned SOL three-bar pivot reversal. The optimizer improved the headline (Sharpe 0.75->1.22, PF 1.28) and cleaned the in-sample gates (sensitivity PASSED with 0 cliffs, is_overfitted=FALSE, walk-forward OOS positive in all 3 windows), BUT the decisive forward-validation gates fail: deflated_sharpe 0.3325 (is_significant FALSE) with the optimized Sharpe 1.22 BELOW the expected-max luck bar 1.474 over 225 trials, PBO 0.5727, sharpe_ci_low -0.0185; and -- decisively -- the HOLDOUT FAILED catastrophically (holdout_sharpe -4.147, ratio -3.316: the recent once-tested window loses money badly). The optimized annual returns make the cause explicit: the pullback-in-uptrend edge is carried by the 2021 (+67.9%) and 2023 (+36.9%) bull markets and has decayed to negative in 2025 (-6.0%) and 2026 (-5.3%), with the rolling Sharpe -5 to -8 through 2025-2026 -- a regime-dependent edge that has stopped working, not a tunable parameter region. The passing sensitivity/walk-forward measure in-sample behavior; the forward holdout and DSR both reject it. Not iterate (attempt 1 of 2): the failure is regime decay + a deeply negative holdout + a sub-luck-bar DSR, and sensitivity already passed with 0 cliffs, so a second sweep re-finds the same config that fails the recent holdout. Not revise_hypothesis: the pullback-in-trend mechanism is validated per-asset (BTC Sharpe 0.76 is the cited instance); SOL is the decaying arm with no reframe. FAILURE PATTERN: single-asset long-only pullback/reversal trend strategies on crypto majors capture big bull-market dips (2021/2023) but are regime-dependent -- even after an optimizer lift to Sharpe ~1.2 with passing sensitivity and is_overfitted=FALSE, they fail deflated Sharpe (0.33, below the luck bar) and a deeply negative holdout (-4.15) because the edge concentrates in bull years and decays in 2025/2026. This is the fifth such weak-trend/pullback strategy this run (DOT, SOL pivot, ADA momentum, XRP Chandelier) to optimize then fail DSR + a negative holdout -- a walk-forward that passes with a negative holdout is an edge fit to historical bull regimes that is decaying out-of-sample.

Long-only single-instrument Pullback-in-Trend entry on SOLUSDT.BINANCE USD-M futures 4H bars, OHLCV-only. It confirms an uptrend when SMA(50) is rising over the past 20 bars AND the current close is above SMA(50), then waits for a brief pullback (>=2 consecutive red bars OR a >3% drawdown from the 5-bar high) and enters long on the first green bar after the pullback completes. Exits on a 4% stop-loss, an 8% take-profit, a trend break (close back below SMA-50), or a 24-bar (~4-day) time stop. calculate_signal returns the continuous %-distance of close above its SMA-50 (varies every bar); the binary pullback-completion trigger is captured in an _armed flag checked by should_enter. Sizes 30% of equity per trade, capped by 1.5% risk over the 4% stop.

Correct pullback/trend payoff: win_rate 0.358 with avg_win $4,651 vs avg_loss $2,279 (2.0x), tail_ratio 1.50; ample sample (520 trades), PF 1.14, total_return +126%

Healthier than recent weak-trend submissions: Sharpe 0.75 with MODERATE kurtosis 5.45 (not artifact-driven) and sharpe_ci_low -0.28 (only marginally straddles 0); validated Pullback-in-Trend mechanism class

Capacity $6.98M, impact 11.97% -- deployable scale (unlike the abandoned thin-alt volume breakouts)

Front-loaded / regime-dependent: 2021 (+93.8%) and 2023 (+37.1%) bulls carry it; 2025 -4.2%, 2026 -5.9% negative, recent rolling Sharpe deeply negative

sharpe_ci_low -0.28 (CI straddles 0) -- DSR-fail risk after 225-trial selection

High turnover (174.6) with impact 11.97% -- fee-drag watch item; same bull-regime-dependent profile as the abandoned SOL pivot reversal

~0.5+ (pre-opt)

~62 WF floor

3 of 7

majority

moderate (<~10)

2.0x (avg_win/avg_loss)

> 1 for trend

Optimizer improved the headline (Sharpe 0.75->1.22, PF 1.28) and cleaned in-sample gates: sensitivity PASSED (0 cliffs), is_overfitted=FALSE, walk-forward OOS positive in all 3 windows ([0.62, 0.84, 2.28])

Correct pullback payoff (avg_win 2.1x avg_loss), deployable capacity ($33.7M), moderate kurtosis 5.9

Fails deflated Sharpe: DSR 0.3325 (<0.95), is_significant FALSE; optimized Sharpe 1.22 BELOW the expected-max luck bar 1.474; PBO 0.5727; sharpe_ci_low -0.0185 (grazes 0)

Holdout FAILED catastrophically: holdout_sharpe -4.147, ratio -3.316 -- the recent window loses money badly

Regime-decayed: 2025 -6.0%, 2026 -5.3%, rolling Sharpe -5 to -8 through 2025-2026; the +126% is carried by the 2021/2023 bulls

> 1.474 (expected-max)

SolFourHourPullbackInUptrendReversalLong was the strongest-looking non-fade candidate of the run — a genuine pullback-in-trend payoff (avg_win 2.0× avg_loss, +126%, moderate kurtosis) ported from the analyst-validated BTC instance to SOL, and after optimization it passed every in-sample test: sensitivity with 0 cliffs, is_overfitted=FALSE, and positive OOS across all three walk-forward windows. But the forward holdout — the one window the optimizer never saw — collapsed to a −4.147 Sharpe, and the optimized 1.22 Sharpe fell below the 1.474 luck bar (DSR 0.33), exposing the edge as concentrated in the 2021/2023 bull markets and decaying to negative through 2025/2026. The analyst abandoned it on its first iteration: with sensitivity already clean, a second sweep would only re-find the same regime-dependent config that fails the recent holdout. It joins DOT, the SOL pivot, ADA momentum, and XRP Chandelier as the fifth weak-trend/pullback strategy this run to optimize to a flattering Sharpe and then fail honest deflation and a one-shot holdout — the clearest demonstration yet that passing the in-sample gates means nothing if the forward window loses money.

Single-asset long-only pullback/reversal trend strategies on crypto majors capture big bull-market dips (2021/2023) but are regime-dependent — even after an optimizer lift to ~1.2 Sharpe with passing sensitivity and is_overfitted=FALSE, they fail deflated Sharpe (below the luck bar) and a deeply negative holdout because the edge decays in 2025/2026 — so a clean walk-forward with a negative holdout is an edge fit to historical bull regimes, and in-sample/rolling gates can't substitute for the one untouched forward window.

The backtest-review gate passed it to optimization ('optimize'), but the analyst abandoned it post-optimization: despite passing sensitivity and walk-forward, the decisive forward-validation gates rejected it — the optimized Sharpe 1.22 sat below the expected-max luck bar of 1.474 (DSR 0.3325, is_significant=false), and the once-tested holdout failed catastrophically (−4.147) — because the edge concentrates in the 2021/2023 bull markets and has decayed to negative in 2025/2026 (rolling Sharpe −5 to −8), a regime-dependent edge that has stopped working rather than a tunable parameter region.

A long-only, OHLCV-only pullback-in-uptrend entry on SOLUSDT.BINANCE 4H — applying the analyst-validated Pullback-in-Trend class (cited BTC iteration Sharpe 0.76) to SOL: confirm an uptrend (rising 50-SMA over 20 bars AND close > SMA), wait for a brief pullback (2+ consecutive red bars OR a >3% drawdown from the 5-bar high), then enter long on the first green bar after the dip, exiting on a stop, take-profit, trend break below the SMA, or time stop.

The raw backtest showed a correct, non-artifact pullback payoff — 520 trades, win_rate 0.358 but avg_win $4,651 vs avg_loss $2,279 (2.0×), profit_factor 1.14, total_return +126%, Sharpe 0.75, moderate kurtosis 5.45, deployable $6.98M capacity — though front-loaded to the 2021/2023 bulls with 2025/2026 negative and sharpe_ci_low −0.28. Optimization lifted it to Sharpe 1.22 / PF 1.28 and cleaned the in-sample gates (sensitivity PASSED 0 cliffs, is_overfitted=FALSE, walk-forward OOS positive in all 3 windows [0.62, 0.84, 2.28]), but the forward gates failed: DSR 0.3325, optimized Sharpe 1.22 below the 1.474 luck bar, PBO 0.5727, and holdout Sharpe −4.147 (ratio −3.316).