EthFourHourVolumeConfirmedBreakoutLong

ETH 4H Volume Breakout Long

A LONG-ONLY momentum-CONFIRMATION breakout strategy on ETHUSDT.BINANCE perpetual futures (Binance USD-M) using 4-HOUR bars. Mechanically equivalent to the BTC/BNB/SOL 4H volume-breakout strategies already in pipeline — same entry triggers, same volume requirement, same exit logic, same parameters — applied to ETH as the fourth major instrument completing the 4H-volume-breakout coverage. ETH is the natural completion of the cross-instrument validation matrix: BTC (largest market cap), BNB (exchange token), SOL (high-volatility alt), DOGE/XRP/ADA (daily-bar variants), and ETH (the missing second-largest market cap) — each instrument's pipeline strategy uses identical mechanism parameters to test cross-instrument generalization, not parameter mining. ETH is selected because: (1) ETH has only ONE pipeline strategy (Golden Cross Momentum Long on daily bars) — adding a 4H volume breakout adds within-instrument diversification at a different timeframe AND different signal type; (2) ETH precision is confirmed in the map (proven by the golden cross strategy running); (3) ETH has 6+ years of clean Binance USD-M 4H bar data with no deprecation risk; (4) ETH is the second-most-liquid crypto perpetual after BTC, supporting clean volume-based signals; (5) ETH has produced episodic high-volume breakouts on multiple narrative cycles (Q1 2021 DeFi summer, Sep 2022 Merge, Q1 2024 ETF approval, Q4 2024 institutional flows) — exactly the setup this mechanism targets. Built entirely from the session's EMPIRICALLY VALIDATED working pattern: 4H bars (validated frequency, generates 50-150+ trades/year), momentum CONFIRMATION direction (NOT pullback — 3x falsified; NOT counter-trend — 2x falsified), volume-spike confirmation (working mechanism class with 3 successes already), breakout-failure exit (LINK weekly lesson), outlier-dependency abandonment rule. Structurally distinct from every pipeline strategy by virtue of instrument-timeframe combination not yet covered. Cross-instrument parameter consistency: SAME 4H lookback structure, SAME 1.5× volume threshold, SAME exit logic as BTC/BNB/SOL versions — testing mechanism generalization to ETH, not parameter mining.

Implements the hypothesis as a verbatim cross-instrument transfer of the session's empirically validated 4H-volume-breakout mechanism (momentum CONFIRMATION, not pullback or counter-trend, both of which were falsified) to ETH, the missing second-largest-cap major that completes the BTC/BNB/SOL/ETH matrix. Every entry trigger (20-bar Donchian breakout), the 1.5x volume-spike filter, the 50-SMA trend gate, and the exit logic (TP/SL/trend-break/time-stop) are copied from the landed family with NO re-tuning, so this tests mechanism generalization rather than parameter mining. Signal design follows the passing family pattern: calculate_signal returns a continuous, bar-varying blend (volume surge + channel penetration + ROC) to satisfy Layer-2 frozen-signal detection, while the discrete breakout+volume+trend conjunction is precomputed into _entry_ok and read by should_enter. Sizing is capital-relative via get_account_equity() with a risk cap, and 4H holds spanning days clear the ~0.10% USD-M round-trip fee. Venue is BINANCE USD-M futures for consistency with the sibling family (the cross-instrument matrix is defined on futures); leverage is left at 1.0 and is not referenced to amplify sizing, so there is no leverage_set_but_unused risk.

The 4H volume-confirmed breakout mechanism does NOT generalize to ETH — this is the asset-mismatch failure mode the architecture is already known to have (the hypothesis itself cites LINK -7.2% on smooth-volume assets). As a deliberate fixed-parameter cross-instrument test, the backtest IS the experiment, and the answer is negative: over a healthy 204-trade sample ETH yields Sharpe 0.325 with sharpe_ci_low -0.770 (CI straddles zero), profit_factor 1.09, win_rate 34%, NEGATIVE alpha (-0.0038), and omega 1.10 — qualitatively far weaker than every promoted sibling (SOL 4.42, BTC 3.74, ADA 2.41, BNB 1.74, all with positive CI lower bounds). The thin edge is fee-fragile (impact_cost_pct 9.72%, commission 3.41% of gross, capacity only $10.6M, expectancy just $54/trade) and decays through the recent regime (2022 -3.0, 2023 -2.8, 2025 0.0, 2026 -3.9). Not optimize: a base Sharpe of 0.32 with a CI through zero is far below the 225-trial expected-max luck bar, so optimization would predictably fail the deflated-Sharpe/holdout gates (the repeated pattern this session), and re-tuning ETH parameters would be exactly the parameter mining the hypothesis's fixed-parameter generalization design forbids. Not iterate: there is no code bug — the mechanism faithfully implements the family template and simply does not fit ETH, so there is no specific developer fix. The cross-instrument matrix result stands: the mechanism works on BTC/BNB/SOL/ADA but not ETH; abandon this instance.

Long-only 4-hour volume-confirmed Donchian breakout on ETHUSDT.BINANCE USD-M perpetual futures. Enters long when the bar closes above the highest high of the prior 20 bars AND bar volume exceeds 1.5x the trailing 20-bar average volume AND price is above its 50-bar SMA. Exits on a 12% take-profit, 6% stop-loss, close below the trend SMA (breakout failure), or a 36-bar (~6-day) time stop. Position sized as the min of 18% notional and a 1.5%-risk/stop-distance cap. Identical mechanism and parameters to the landed BTC/BNB/SOL 4H volume-breakout siblings, applied to ETH. Leverage 1.0.

Clean, faithful implementation: 204 longs / 0 shorts, healthy sample, zero plumbing drops; a legitimate fixed-parameter cross-instrument generalization test.

Marginally positive over the full sample (total_return +17.6, expectancy +54/trade) — not a catastrophic loser.

Mechanism does not generalize to ETH: Sharpe 0.325 with sharpe_ci_low -0.770 (CI straddles 0), vs siblings SOL 4.42 / BTC 3.74 / ADA 2.41 / BNB 1.74 all with CI low >0.

Barely an edge: profit_factor 1.09, win_rate 34%, negative alpha (-0.0038), omega 1.10 — and fee-fragile (impact_cost_pct 9.72%, commission 3.41% of gross, capacity only $10.6M).

Decaying: positive only 2020/2021/2024; negative/flat in 2022, 2023, 2025, 2026.

Base Sharpe 0.32 sits far below the 225-trial deflation luck bar — optimization is very unlikely to clear DSR/holdout and would be ETH-specific parameter mining against the hypothesis's own no-re-tuning generalization design.

EthFourHourVolumeConfirmedBreakoutLong was designed to complete the 4H volume-breakout cross-instrument matrix, applying the exact BTC/BNB/SOL family parameters to ETH as a generalization test rather than a new tuning exercise. Its 204-trade backtest was faithfully implemented and marginally profitable (+17.6 total return, +$54/trade), but the edge was thin and unstable: Sharpe 0.325 with a CI straddling zero, PF 1.09, 34% win rate, negative alpha, heavy fee fragility, and decay through 2022–2026. Against siblings whose Sharpes ran 1.74 to 4.42 with positive CI lower bounds, the analyst judged that the mechanism simply does not fit ETH and abandoned it at backtest review on iteration 1 — declining to optimize because a base Sharpe of 0.32 would predictably fail deflation and re-tuning would violate the hypothesis's own no-re-tuning premise. The cross-instrument conclusion stands: the mechanism works on BTC/BNB/SOL/ADA but not ETH.

The 4H volume-confirmed breakout mechanism does not generalize to ETH — this is the known asset-mismatch failure mode (cited via the LINK smooth-volume lesson), and since re-tuning ETH parameters would be exactly the parameter mining the fixed-parameter design forbids, a negative cross-instrument result should simply stand as abandon.

The analyst abandoned it at the pre-optimization backtest-review gate: because it was a fixed-parameter generalization test, the backtest itself was the experiment, and the answer was negative — ETH's Sharpe 0.325 (CI through zero) sits far below its promoted siblings (SOL 4.42, BTC 3.74, ADA 2.41, BNB 1.74, all with positive CI lower bounds), so optimization was not run.

A long-only ETHUSDT.BINANCE 4H volume-confirmed Donchian breakout (close above the prior 20-bar high, volume ≥1.5× its 20-bar average, above the 50-SMA) — a deliberate fixed-parameter cross-instrument generalization test of the landed BTC/BNB/SOL/ADA breakout family, with no re-tuning for ETH.

Over a healthy 204-trade sample (204 longs, 0 shorts) it was only marginally positive — total return +17.6, expectancy +$54/trade — but weak on risk-adjusted terms: Sharpe 0.325 with sharpe_ci_low -0.770 (CI straddling zero), profit factor 1.09, 34% win rate, negative alpha (-0.0038), omega 1.10, and fee-fragile (impact cost 9.72%, commission 3.41% of gross, capacity only $10.6M), profitable only in 2020/2021/2024 and flat-to-negative in 2022, 2023, 2025, 2026.