zenith-lowtf — Can V5's edge run on a lower timeframe?

BTC-first → 10-asset portfolio → out-of-sample. Conservative costs (~0.21% round-trip). Frozen config: 5m, rr×1.0, stop 4.0 ATR.

Verdict

The lower-TF edge is real and survives out-of-sample (5m OOS PF 1.418, t=2.49 — degrades gracefully from IS, not overfit) and it takes +19% more trades, the original goal. But it does not beat V5: the 15m baseline holds a higher PF (1.825 vs 1.418) at roughly half the drawdown, both IS and OOS. The binding constraint is edge-density/correlation of the pullback entry — not the timeframe. Recommendation: keep V5; pursue a second regime-orthogonal 15m entry + correlation gate (V6), not a naive lower-TF port.

Portfolio results — 10 assets, conservative costs

RunTradesWin%PFReturnMaxDDt-stat
IN-SAMPLE (Binance, 2020 → 2024-04)
15m baseline (V5)60062.7%1.825218.9%-11.44%5.11
5m locked71160.3%1.535191.2%-18.35%4.12
OUT-OF-SAMPLE (Binance, 2024-04 → 2026-05) — frozen config, run once
15m baseline (V5)31561.3%1.82560.4%-8.07%3.79
5m locked37659.0%1.41843.5%-10.64%2.49

In-Sample equity 2020–2024

Out-of-Sample equity 2024–2026

Integrity — Look-ahead audited clean (all HTF/regime series shift(1); engine forward-only). Fees taker both sides + 0.05%/side slippage. OOS run once at frozen params, never re-tuned. Bybit OOS dropped (REST retention truncation); primary OOS = Binance full history, balanced 10-asset panel (227,808 bars each). Noise-band law verified: peak-edge stop widens 2.5→4.0→6.0 across 15m→5m→3m (√TF).