Sample · abbreviated deliverable
Pre-registered daily BTC/USDT trend hypothesis (Donchian + ATR trail, long-only, EMA filter). Full experiment report on disk: H001_20260810T082405Z (research lab, Aug 2026).
Evidence verdict: insufficient for live capital
| Symbol / TF | BTC/USDT · 1d · Binance spot (public OHLCV) |
|---|---|
| Cost model | 10 bps fee + 5 bps slippage per side (pre-registered) |
| Walk-forward | 730 train / 180 test bars, step 180 |
| Null test | 100 permutations, pass if OOS PnL > 95th percentile of null |
| Min OOS trades gate | ≥ 15 (pre-registered) |
| Trades | 13 |
|---|---|
| Expectancy / trade | 8.76 (currency units on 1,000 start) |
| Total OOS PnL | 113.92 |
| Win rate | 46.15% |
| Profit factor | 2.48 |
| Max DD (trade equity path) | −1.95% |
Raw OOS metrics can look acceptable. Gates still failed — see below.
| oos_expectancy_gt_0 | PASS |
|---|---|
| oos_total_pnl_gt_0 | PASS |
| null_test_pass | FAIL — observed 113.9 < threshold 455.5; null mean 148.8 |
| sample_adequate | FAIL — 13 trades < 15 required |
Do not tune H001 parameters on this OOS. Either archive hypothesis or open a new ID with pre-registered changes. If continuing trend family: lengthen history requirement or tighten null test before any paper trading — still no live capital until all gates pass.
This page is a marketing sample, not a client deliverable. Numbers are copied from the archived experiment report only. No performance promise is made for any strategy.
Buy the founding review — €79