AptFourHourVolumeBreakoutLong

APT Four-Hour Volume Breakout Long-Only (BINANCE USD-M Futures, 4H, OHLCV-Only)

Long-only APT (Aptos) 4-hour strategy on BINANCE USD-M futures. Direct analog of the landed BTC/ETH/SOL/BNB/AVAX/XRP/MATIC/FIL/ICP FourHourVolumeBreakoutLong family — applied to APT. Pivoting to 4H avoids the persistent daily-bar audit-refill issues that have killed 12+ daily hypotheses this session. Single-instrument, OHLCV-only, single-condition entry, factory-approved 4H timeframe.

Direct analog of the landed BTC/ETH/SOL/BNB/AVAX/XRP/MATIC/FIL/ICP FourHourVolumeBreakoutLong family applied to APT, mirroring the exact proven recipe: volume-surge + bullish close + SMA trend filter, with TP/stop/time exits and risk-based sizing. APTUSDT.BINANCE 4H data is catalog-confirmed. Pivots to the factory-approved 4H timeframe to avoid the daily-bar audit-refill failures that killed 12+ daily hypotheses this session. The volume baseline uses the prior `volume_window` bars (excludes the current bar) so a surge isn't averaged against itself. Implemented through the standard base flow with no on_bar override; calculate_signal stores the boolean entry decision and returns a continuous tanh(volume-surge)+ROC proxy so Layer-2 frozen-signal detection passes while should_enter reads the stored flag (the returned value never leaks a false entry). Risk: +12% TP vs -6% stop (2:1), well above the ~0.10% round-trip futures fee; 4H breakouts fire frequently enough for a healthy walk-forward sample. Venue BINANCE USD-M futures per spec; long-only, leverage 1.0, OHLCV-only, single-instrument.

Negative-expectancy single-name directional breakout — no edge to optimize. Base backtest loses money: total_return -8.23%, Sharpe -0.209, profit_factor 0.869 (<1), avg_trade_return_pct -0.257% (negative per-trade edge, below the 0.15% fee floor), expectancy -$95/trade over 99 trades. The one positive year (2023 +5.4%) has fully decayed — 2024 -2.3%, 2025 -3.95%, 2026 -7.5% — with rolling Sharpe persistently negative. The mechanism fires correctly (not a code/data bug), it simply has no edge on APT. Parameter optimization cannot flip a negative-expectancy long-only breakout positive, so this is abandon, not iterate. Failure pattern: dead single-name OHLCV directional/breakout (PF < 1, negative return and negative per-trade edge) — abandon at BACKTEST_REVIEW.

Long-only APTUSDT.BINANCE 4H volume-breakout strategy, OHLCV-only, single-instrument. Enters long when the just-closed 4H bar (1) trades on volume >= 2x the average of the prior 20 bars (volume surge), (2) closes bullishly (close>open), and (3) is above the 100-period SMA (uptrend filter). Exits on a +12% take-profit, -6% stop-loss, or a 24-bar (~4 day) time stop. Sizes 18% of equity notional per trade, capped so the stop risk stays within 1.5% of equity. Leverage 1.0.

CLEAN RESTART 2026-09-04 — this run's verdict history and learning records were removed and it was restarted from verification. Its previous abandonment came from the pipeline, not from the market: the Layer-2 harness mis-bound @staticmethod helpers (fixed), QA issued terminal performance verdicts on an unoptimized smoke test (removed — QA now judges correctness only), and sandbox timeouts came from backtest-slot starvation (fixed). The hypothesis and the strategy code are unchanged. Verify the code on its merits; performance is decided later by the full backtest and the optimizer.

Clean implementation: 99 trades fired correctly, no warm-up/data/zero-trade failure, mechanism matches the hypothesis (bullish volume-surge + SMA uptrend, long-only).

Max drawdown 20% is not catastrophic.

Negative expectancy: total_return -8.23%, Sharpe -0.209, profit_factor 0.869 (<1), avg_trade_return_pct -0.257% — the per-trade edge is negative, well below the 0.15% fee floor.

Deteriorating year over year: 2023 +5.4% then 2024 -2.3%, 2025 -3.95%, 2026 -7.5%; rolling Sharpe persistently negative after early 2024.

win_rate 38% with avg_win $1,638 vs avg_loss $1,174 — winners are not large enough to overcome the loss rate.

No parameter region to tune toward: a long-only breakout with negative base expectancy across every recent year cannot be optimized positive.

This strategy ported the well-established 4H volume-breakout mechanism to APT on Binance USD-M futures, explicitly choosing 4H to sidestep the daily-bar audit-refill issues that had killed many daily hypotheses. The code was complete and standard, but it never got the chance to prove itself: the data layer returned zero bars for APTUSDT.BINANCE after five retries, failing the minimum data requirement. The strategy was abandoned on the first iteration purely as a data-availability blocker, with no backtest, optimization, or analyst verdict ever generated, and a note to revisit the hypothesis once the symbol is backfilled.

Confirm that an instrument's historical data is actually collectable before queuing a hypothesis on it — a sound, proven mechanism is worthless if the symbol cannot be backfilled, and the note recommends re-evaluating once collection is fixed.

It was abandoned at the data-collection stage for an infrastructure reason, not a strategy flaw: the data layer could not supply APTUSDT.BINANCE 4H bars, so the pipeline never reached backtesting, optimization, or analysis.

A long-only APT (Aptos) 4-hour volume-breakout strategy on BINANCE USD-M futures — a direct analog of the landed BTC/ETH/SOL/BNB FourHourVolumeBreakoutLong family — entering on a volume surge (≥2× the prior 20-bar average) plus a bullish bar above the 100-bar SMA, exiting on take-profit, stop-loss, or a time stop.

There are no performance metrics: the backtest never ran. Backtest data for APTUSDT.BINANCE could not be collected after 5 retries (0 bars returned against the 100-bar / 60-day requirement), so no backtest, optimization, or analyst review was produced.

n/a

data_unavailable: backtest data for APTUSDT.BINANCE could not be collected after 5 retries: - bars (bar_type=APTUSDT.BINANCE-4-HOUR-LAST-EXTERNAL): got 0, needed 100, or 60+ days span The data layer (not the strategy) is the blocker — re-evaluate this hypothesis once collection is fixed / the symbol is backfilled.