SolSpot4HDrawdownAccumulationLong

SOL BINANCE_SPOT 4H N-Bar Drawdown Accumulation Long-Only (Buy-the-Dip, 3-Parameter, No Stop-Loss)

A long-only single-instrument accumulation strategy on SOLUSDT.BINANCE_SPOT using 4-HOUR bars and OHLCV-only data. Mechanism: when SOL's current close has dropped >= 15% from its rolling 90-bar (15-day) high, enter a long position. Exit when SOL recovers >= 5% above entry price OR after a hard 240-bar (~40 day) time exit. This is a structurally-proven BUY-THE-DIP / ACCUMULATION mechanism — the existing BTC Spot 30-Day Drawdown Accumulation sibling reached PAPER STAGE at Sharpe 3.33 per recent analyst feedback. STRATEGIC RATIONALE: (1) PORTFOLIO QUOTA FILL — BINANCE_SPOT at 10.4% vs ≥15% target. Spot is necessarily long-only (CASH account cannot short), so this adds to the long_only quota — which IS over-represented at 87.4% — but spot specifically is under-represented and the factory's strategy_archetypes explicitly recommends spot for 'buy-the-dip, accumulation, long-only-no-leverage strategies'. The net effect is constructive: SPOT venue diversification matters more than incremental long_only push because the existing 960 long_only experiments are mostly FUTURES long_only (which is a different risk profile than spot long_only). (2) MECHANISM PROVENANCE — drawdown accumulation has the strongest factory empirical evidence of any mechanism class: the BTC sibling at PAPER stage Sharpe 3.33 demonstrates it survived all 3 validation phases plus risk review. This proposal transfers the mechanism to a DIFFERENT ASSET (SOL has distinct volatility character — higher vol, faster recoveries than BTC) and a DIFFERENT TIMEFRAME (4H instead of daily — 6x more bars for walk-forward sample size). (3) WHY SOL AND NOT BNB/ETH/AVAX — BNB has a regime-decay moratorium (post-2024 multiple failures including the BNB Spot Drawdown Accumulation sibling); ETH has shown regime decay (ETH/BTC ratio weakness, ETH Down-Day Pullback failure, ETH BINANCE_SPOT Pullback failure); AVAX has had multiple catastrophic failures (Skewness, Trend-Pullback). SOL has been the strongest performing major in 2024-2026 with high volatility and quick recoveries — the asset profile most likely to retain drawdown-mean-reversion behavior. (4) DECISIVE FAILURE-MODE AVOIDANCE: NO STOP_LOSS PARAMETER — explicit fix for the LINK/ETH iter-2 optimizer-reversal failure (stop_z pushed to -4.39 destroying risk controls). Risk is bounded entirely by (a) the 15% drawdown entry threshold (you're buying meaningful weakness, not noise), (b) the hard 240-bar (~40 day) time exit (NOT tunable), (c) position sizing at 30% capping single-trade exposure. The optimizer cannot push extreme parameter values because there are no risk-tunable parameters — only drawdown_threshold_pct, lookback_bars, position_pct are tunable, and all three have clean monotonic response surfaces. (5) FEE MATH EXPLICITLY VIABLE on spot (0.20% RT fees, viability floor 0.25% per trade): expected win = 5-15% recovery × 30% position = 1.5-4.5% gross per winning trade. At 60-70% win rate (typical for drawdown buys in established assets), net per-trade EV ≈ 0.6-2.0% comfortably exceeds the 0.25% spot fee floor. (6) AVOIDS ALL OTHER RECENT FAILURE PATTERNS: NOT premium_index basis MR (XRP failed catastrophically); NOT cross-venue basis arb (BTC HL/Binance failed fee math); NOT Donchian-with-trend (regime decay across BTC/XRP/AVAX); NOT funding-based (multiple failures); NOT skewness (mechanism class refuted); NOT OI-surge (regime dependent); NOT 1H pair MR (deprecated); NOT Connors down-day pullback (deprecated for daily). The drawdown-accumulation mechanism class stands UNIQUELY as the only sibling-validated-paper-stage approach this session. (7) ITERATION-SAFE: 3 hypothesis-declared tunables with literature-validated defaults; no risk feedback parameter for optimizer to game. If iter-1 fails, it's mechanism-class regime decay (which iter-N parameter tuning won't fix) — clean abandon signal.

Direct transfer of the factory's strongest-evidence mechanism class -- drawdown accumulation -- which reached paper stage at Sharpe 3.33 on the BTC spot sibling, to SOL (higher vol, faster recoveries: the asset profile most likely to retain drawdown-mean-reversion) on 4H bars (6x more bars than daily for walk-forward sample size). Long-only with no leverage and no funding edge => correct venue is BINANCE_SPOT (CASH, leverage forced to 1, cannot short); should_enter only ever returns BUY and exits use the base SELL-to-close path, satisfying the spot long-only constraint. The signal is continuous (tanh of dip depth below threshold, snapping to 1.0 at/above the 15% threshold) so Layer 2 never sees a frozen signal while should_enter thresholds the discrete trigger. Iteration-safety is built in: the only tunables are three monotonic-response knobs and there is NO stop-loss/risk-feedback parameter for the optimizer to push to extremes (the explicit fix for the LINK/ETH iter-2 optimizer-reversal failure). Fee viability holds on spot's ~0.20% RT cost: a 5% recovery on a 30% position is ~1.5% gross per win, far above the 0.25% spot per-trade floor.

Same mechanism+asset that already failed this session, with a below-threshold edge and an uncontrollable left tail — not worth optimizing. (1) SOL spot drawdown-accumulation was already abandoned this session (the 25-day sibling: DSR 0.0008, sensitivity cliffs, ZERO walk-forward OOS trades, holdout -5.43); this 4H/15% version is the same mechanism on the same asset differing only in timeframe/threshold, so it would hit the identical wall. (2) The edge is below the meaningful floor: profit_factor 1.15 (<1.2), and the headline Sharpe 1.23 is distorted by 70.9% annualized vol (per-period attribution + genuine fat losers); sharpe_ci_low -1.50 straddles 0. (3) The no-stop design produces an uncontrollable left tail WORSE than the LINK variant: avg_loss $13.3k is 4.7x avg_win $2.8k, return_skew -2.47, tail_ratio 0.31, largest loss -$28.8k, max_drawdown 43% (CI high 68%) — and with only 3 tunable entry/sizing params and hard-coded exits, optimization cannot address it. (4) The edge is 2021-bull-concentrated (+95.5% annual) with a catastrophic 2022 (-38.7%, the no-stop design steamrolled in the downtrend) and a decaying recent regime (2025 -3.1%, 2026 -8.5%). Unlike the LINK variant (PF 1.23, no prior failure, milder tail) which earned an optimization test, SOL has already refuted this mechanism and has a worse, below-threshold profile. The hypothesis's own decision rule ('if iter-1 fails it's mechanism-class regime decay — clean abandon') is already answered by the prior SOL failure. FAILURE PATTERN: stopless spot drawdown-accumulation buy-the-dip ported to a high-vol alt (SOL) produces a negatively-skewed picking-up-pennies profile (PF ~1.15, avg_loss >> avg_win, 43% DD) carried by the 2021 bull and steamrolled in 2022/2025-2026 downtrends; the asset already failed this mechanism's 3-phase validation, so re-optimizing a timeframe/threshold variant is a dead end.

Long-only buy-the-dip accumulation on SOLUSDT spot (BINANCE_SPOT, 4H bars, OHLCV-only). Tracks a rolling 90-bar (~15-day) high; when SOL's close falls 15% or more below that trailing high it enters a long sized at 30% of equity. It exits when price recovers 5% above the entry price (take-profit) or after a hard 240-bar (~40-day) time cap. No stop-loss: downside is bounded by the deep-drawdown entry, the deterministic time exit, and the position cap. Exactly three tunable parameters (lookback_bars, drawdown_threshold_pct, position_pct); the 5% recovery target and 240-bar cap are locked class constants.

Clean execution (109 trades, 0 dropped); sibling mechanism (BTC spot drawdown-DCA) reached paper stage

High headline win rate (84%) typical of buy-the-dip

SOL drawdown-accumulation ALREADY failed this session (25-day sibling: DSR 0.0008, zero OOS trades, holdout -5.43) — SOL is a dead target for this mechanism

Edge below the meaningful floor: profit_factor 1.15 (<1.2); headline Sharpe 1.23 distorted by 70.9% annualized vol; sharpe_ci_low -1.50 straddles 0

Uncontrollable no-stop left tail: avg_loss $13.3k is 4.7x avg_win $2.8k, return_skew -2.47, tail_ratio 0.31, largest loss -$28.8k, max DD 43% (CI high 68%) — optimizer cannot fix (exits hard-coded)

2021-bull-concentrated (+95.5%) with catastrophic 2022 (-38.7%) and negative recent regime (2025 -3.1%, 2026 -8.5%)

SolSpot4HDrawdownAccumulationLong transferred the paper-stage BTC spot drawdown-accumulation mechanism (Sharpe 3.33) to SOL on 4H bars, deliberately omitting a stop-loss to avoid the optimizer-reversal failure that sank a prior pair strategy. But the same mechanism on SOL had already failed this session's full validation, and this variant's profile was worse: an 84% win rate masked a sub-1.2 profit factor and a deeply negatively-skewed no-stop tail (avg loss 4.7x avg win, 43% drawdown), with the edge concentrated in 2021 and steamrolled in 2022 and the decaying 2025/2026 regime. The analyst judged it a dead target — already refuted, below threshold, and untunable — and abandoned it before optimization. It ended after one iteration as abandoned, never advancing to optimization or risk review.

Stopless spot drawdown-accumulation buy-the-dip ported to a high-vol alt (SOL) is a picking-up-pennies profile (high win rate, PF~1.15, avg_loss >> avg_win, 43% drawdown) carried by the 2021 bull and steamrolled in downtrends — and an asset that already refuted the mechanism in 3-phase validation will not be rescued by a timeframe/threshold variant; removing the stop to dodge optimizer-reversal trades one failure mode for an uncontrollable tail.

It was abandoned at the pre-optimization backtest-review gate (verdict: abandon) because the same mechanism on the same asset had already failed this session (the 25-day SOL sibling: DSR 0.0008, zero walk-forward OOS trades, holdout -5.43), the baseline edge was below the meaningful floor, and the no-stop design produced an uncontrollable left tail that hard-coded exits and 3 entry-only tunables could not fix — so optimization and all later stages were never reached.

A long-only single-instrument drawdown-accumulation buy-the-dip on SOLUSDT.BINANCE_SPOT (4H bars, OHLCV-only) that entered long when SOL fell >=15% from its rolling 90-bar (15-day) high and exited on a 5% recovery above entry or a hard 240-bar (~40 day) time cap, deliberately using no stop-loss parameter (only 3 entry/sizing tunables) — a port of the paper-stage BTC spot drawdown-accumulation sibling (Sharpe 3.33) to a higher-vol asset and faster timeframe.

Over 109 trades it had an 84% win rate but a below-floor edge: profit factor 1.15, headline Sharpe 1.23 distorted by 70.9% annualized vol (sharpe_ci_low -1.50), and a severe negatively-skewed left tail from the no-stop design — avg loss $13.3k was 4.7x avg win $2.8k, return_skew -2.47, largest loss -$28.8k, max drawdown 43% — with returns concentrated in the 2021 bull (+95.5%) and catastrophic 2022 (-38.7%) plus a decaying recent regime (2025 -3.1%, 2026 -8.5%).