Does anything beat buy & hold on USD/CNY?
Every setup we tested on USD/CNY — ranked out-of-sample, corrected for multiple testing, and forward-tracked in public from the day this page published. The honest answer is the headline.
Beat buy-and-hold out-of-sample AND cleared the multiple-testing hurdle.
Beat buy-and-hold out-of-sample AND full-sample out-of-sample Sharpe 1.47 cleared the selection hurdle 1.33. Buy-and-hold benchmark: -0.8% CAGR over 24.8 years (+0.1% CAGR in the out-of-sample window).
Educational research from historical backtests — not investment advice. Past performance does not predict future results.
USD/CNY: DMI Direction cleared every bar we set — a rare verdict, not a promise
Currency pairs like USD/CNY are mean-reverting macro instruments with near-zero long-run drift, which cuts both ways: buy-and-hold offers little to beat, but that also makes any outperformance harder to trust, since noise strategies cluster right around zero. Out of 677 indicator configurations we tested on USDCNY, DMI Direction on the daily timeframe passed all three of our checks. It outperformed over the full window, outperformed out-of-sample, and its out-of-sample Sharpe of 1.47 cleared the multiple-testing hurdle of 1.33. That last check matters most, because in a driftless market the best of hundreds of tries is usually luck wearing a lab coat.
Read the figures plainly. Over 7.4 out-of-sample years, the setup produced +4.6% annualized alpha (+1.9% over the full window) against a buy-and-hold CAGR of -0.8%, across 171 trades with a 15.8% win rate and a -24.7% maximum drawdown. For context, only 48.7% of setups on this pair beat the benchmark at all. Clearing the hurdle means the result is unlikely to be a selection artifact — it does not mean it will persist. Currency regimes turn with central-bank cycles and carry conditions, and an edge measured in one regime can fade in the next. Past performance does not predict future results.
Every figure above is computed from our own backtests — nothing is estimated or invented. Hypothetical results; not investment advice.
The setups that ranked, and what they are
Fully mechanical — no discretionary steps, no hidden filters. We name the indicator and publish its record; we keep the settings.
DMI Direction
What it is: A trend-following rule: it holds while a smoothed trend reference is pointing up and goes flat when it turns. Long/flat only — no leverage, no shorting. The parameters we tested are not published.
How it was tested: signals evaluated at daily-bar close, the position changes on the NEXT bar, 0.08% cost per side. Same engine, same cost model, same out-of-sample split as every other number on this site.
Out-of-sample (last ~30% of the window, never used to pick this setup): Sharpe 1.47 · alpha +4.6% · 42 trades over 7.4 yrs.
ADXR
What it is: A trend-following rule: it holds while a smoothed trend reference is pointing up and goes flat when it turns. Long/flat only — no leverage, no shorting. The parameters we tested are not published.
How it was tested: signals evaluated at daily-bar close, the position changes on the NEXT bar, 0.08% cost per side. Same engine, same cost model, same out-of-sample split as every other number on this site.
Out-of-sample (last ~30% of the window, never used to pick this setup): Sharpe 1.33 · alpha +3.9% · 41 trades over 7.4 yrs.
ADX Strong Trend
What it is: A trend-following rule: it holds while a smoothed trend reference is pointing up and goes flat when it turns. Long/flat only — no leverage, no shorting. The parameters we tested are not published.
How it was tested: signals evaluated at daily-bar close, the position changes on the NEXT bar, 0.08% cost per side. Same engine, same cost model, same out-of-sample split as every other number on this site.
Out-of-sample (last ~30% of the window, never used to pick this setup): Sharpe 1.26 · alpha +3.4% · 28 trades over 7.4 yrs.
Since publication — including if it loses
The setup is ahead so far — over a window this short that is mostly noise. Marked to market nightly from real prices, rules frozen at publication, as of 2026-09-11. Currently FLAT.
We tested 677 setups (indicator × parameters × timeframe) on USD/CNY. Only setups with ≥30 trades qualify (532 did). Setups are ranked by out-of-sample Sharpe — the last ~30% of history, which standard-parameter rules never saw during selection. Because picking the best of 677 tries mines even the holdout, the VALIDATED verdict additionally requires the top setup’s OOS Sharpe to clear a selection hurdle of 1.33 (√(2 ln N)/√T) AND positive alpha in both windows. Of the eligible setups, 48.7% had positive out-of-sample alpha (median OOS Sharpe 0.03) — the table below is truncated, but this summary covers all of them. How we test: methodology. The scoring contract is published in full; the parameters of each individual setup are not.
Top 20 of 532 eligible setups
Ranked by out-of-sample Sharpe. Full + out-of-sample columns, costs included. Hypothetical.
| # | Setup | TF | Total ret | Sharpe | Max DD | Win | Trades | α vs B&H | OOS Sharpe | OOS α | OOS trades |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | DMI Direction | Daily | +30.9% | 0.39 | -24.7% | 15.8% | 171 | +1.9% | 1.47 | +4.6% | 42 |
| 2 | ADXR | Daily | +36.1% | 0.59 | -15.3% | 19.0% | 121 | +2.1% | 1.33 | +3.9% | 41 |
| 3 | ADX Strong Trend | Daily | +34.9% | 0.63 | -9.2% | 23.8% | 80 | +2.0% | 1.26 | +3.4% | 28 |
| 4 | ADX / DMI | Daily | +32.5% | 0.55 | -13.5% | 19.3% | 109 | +2.0% | 1.2 | +3.5% | 34 |
| 5 | Parabolic SAR | Daily | -5.0% | -0.08 | -36.1% | 14.4% | 368 | +0.6% | 1.11 | +3.6% | 80 |
| 6 | Fisher Transform | Daily | -23.9% | -0.44 | -44.3% | 13.1% | 641 | -0.3% | 0.94 | +3.0% | 189 |
| 7 | DeMarker (21) | Daily | +0.7% | 0.02 | -32.6% | 13.2% | 212 | +0.8% | 0.92 | +2.8% | 55 |
| 8 | DeMarker (7) | Daily | -14.2% | -0.19 | -39.4% | 12.9% | 441 | +0.2% | 0.89 | +2.8% | 119 |
| 9 | Parabolic SAR (fast) | Daily | -22.6% | -0.42 | -45.2% | 13.9% | 547 | -0.2% | 0.88 | +2.9% | 130 |
| 10 | Exponential Hull MA | Weekly | +15.4% | 0.29 | -11.3% | 26.7% | 75 | +1.4% | 0.85 | +2.2% | 20 |
| 11 | VuManChu Cipher B | Weekly | +14.7% | 0.32 | -4.2% | 52.9% | 34 | +1.4% | 0.85 | +2.1% | 16 |
| 12 | Elder Impulse | Weekly | +5.1% | 0.13 | -14.0% | 34.2% | 79 | +1.0% | 0.83 | +1.9% | 23 |
| 13 | True Strength Index | Weekly | +12.5% | 0.24 | -15.6% | 23.3% | 43 | +1.3% | 0.8 | +2.3% | 11 |
| 14 | DeMarker (14) | Daily | -4.6% | -0.05 | -30.6% | 15.5% | 264 | +0.6% | 0.79 | +2.4% | 80 |
| 15 | MACD | Weekly | +12.7% | 0.24 | -16.5% | 25.6% | 39 | +1.3% | 0.79 | +2.3% | 10 |
| 16 | Super Smoother (Ehlers) | Weekly | +19.2% | 0.35 | -9.2% | 35.5% | 62 | +1.5% | 0.79 | +2.2% | 17 |
| 17 | Derivative Oscillator | Weekly | +10.3% | 0.23 | -15.2% | 29.7% | 74 | +1.2% | 0.78 | +1.8% | 20 |
| 18 | Percentage Price Osc. | Weekly | +12.0% | 0.23 | -16.7% | 23.8% | 42 | +1.3% | 0.77 | +2.2% | 10 |
| 19 | PPO Cross | Weekly | +12.0% | 0.23 | -16.7% | 23.8% | 42 | +1.3% | 0.77 | +2.2% | 10 |
| 20 | TEMA 30 Trend | Weekly | +15.2% | 0.31 | -9.8% | 32.3% | 62 | +1.4% | 0.73 | +1.9% | 19 |
Hypothetical backtests with 0.08%/side costs. Not investment advice — see the full disclaimer.
These are historical backtests of mechanical rules. They are educational research, not investment advice, not a recommendation, and not tailored to you. Educational information only — not investment advice. Hypothetical backtested results; past performance does not guarantee future results. Trading involves risk of loss.