Does anything beat buy & hold on CHF/JPY?
Every setup we tested on CHF/JPY — 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.
No setup beat simply holding once tested honestly. We say so plainly.
Its best setup's out-of-sample profit factor (0.97) is below 1 — it lost money per trade on unseen data. Buy-and-hold benchmark: +3.8% CAGR over 23.2 years (+9.4% CAGR in the out-of-sample window).
Educational research from historical backtests — not investment advice. Past performance does not predict future results.
CHF/JPY: 687 Indicators Tested, None Earned Their Keep
Currency pairs like CHF/JPY should be fertile ground for indicators. There's almost no long-run drift to compete against — CHFJPY's buy-and-hold CAGR of +3.8%% is a low bar — and mean reversion is the textbook behavior of macro-anchored exchange rates. Yet across 687 tested setups, none cleared that bar honestly. The patterns that looked tradeable in-sample fell apart on data they hadn't seen. When an asset barely trends and timing signals still can't add value, the plainest explanation is that the apparent edges were noise dressed up as structure.
The best performer here, Fisher Transform on the daily timeframe, posted an out-of-sample Sharpe of 2.57 against a required hurdle of 1.37 — a hurdle that exists because picking the winner from 687 attempts is data-mining by construction. Its out-of-sample alpha of +5.4%% across 558 trades over 7.0 years tells you what the in-sample fit concealed. Only 1.0%% of setups beat holding at all, before any statistical correction. Read this as a snapshot, not a verdict on the future: regimes shift, and past results describe what happened, not what will.
Every figure above is computed from our own backtests — nothing is estimated or invented. Hypothetical results; not investment advice.
The least-bad setups — shown with their failure numbers
Nothing here earned a verdict — these are the best of a losing field, published so you can see exactly how "best" still failed.
Fisher Transform
Mechanical rule (exactly as backtested): Long while the Fisher Transform turns up above its trigger. Signals are evaluated at daily-bar close, the position changes on the NEXT bar, 0.08% cost per side, long/flat only — no leverage, no shorting.
Out-of-sample (last ~30% of the window, never used to pick this setup): Sharpe 2.57 · alpha +5.4% · 169 trades over 7.0 yrs.
Parabolic SAR (fast)
Mechanical rule (exactly as backtested): Variant — faster Parabolic SAR (step 0.04); long while price holds above SAR. Signals are evaluated at daily-bar close, the position changes on the NEXT bar, 0.08% cost per side, long/flat only — no leverage, no shorting.
Out-of-sample (last ~30% of the window, never used to pick this setup): Sharpe 2.25 · alpha +3.6% · 134 trades over 7.0 yrs.
DMI Direction
Mechanical rule (exactly as backtested): Long whenever +DI is above -DI — pure directional movement, no strength filter. Signals are evaluated at daily-bar close, the position changes on the NEXT bar, 0.08% cost per side, long/flat only — no leverage, no shorting.
Out-of-sample (last ~30% of the window, never used to pick this setup): Sharpe 2.2 · alpha +3.4% · 80 trades over 7.0 yrs.
Since publication — including if it loses
The forward record is just getting started — the gap between the two is the honest score. Marked to market nightly from real prices, rules frozen at publication, as of 2026-07-02. Currently LONG.
We tested 687 setups (indicator × parameters × timeframe) on CHF/JPY. Only setups with ≥30 trades qualify (574 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 687 tries mines even the holdout, the VALIDATED verdict additionally requires the top setup’s OOS Sharpe to clear a selection hurdle of 1.37 (√(2 ln N)/√T) AND positive alpha in both windows. Of the eligible setups, 1.0% had positive out-of-sample alpha (median OOS Sharpe 0.49) — the table below is truncated, but this summary covers all of them. Full recipe: methodology · the engine’s contract lives in the repo as STRATEGY_METHODOLOGY.md.
Top 20 of 574 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 | Fisher Transform | Daily | +822.6% | 1.2 | -27.6% | 39.1% | 558 | +6.2% | 2.57 | +5.4% | 169 |
| 2 | Parabolic SAR (fast) | Daily | +596.5% | 1.05 | -31.3% | 40.4% | 448 | +4.9% | 2.25 | +3.6% | 134 |
| 3 | DMI Direction | Daily | +419.3% | 0.92 | -25.7% | 33.7% | 273 | +3.5% | 2.2 | +3.4% | 80 |
| 4 | DeMarker (7) | Daily | +449.0% | 0.95 | -29.9% | 38.0% | 416 | +3.8% | 2.0 | +2.1% | 124 |
| 5 | Parabolic SAR | Daily | +499.0% | 0.97 | -26.1% | 41.4% | 278 | +4.2% | 1.95 | +1.8% | 82 |
| 6 | Vortex (7) | Daily | +325.5% | 0.81 | -28.6% | 38.7% | 426 | +2.6% | 1.69 | +0.6% | 138 |
| 7 | Woodies CCI | Daily | +229.5% | 0.7 | -31.5% | 36.3% | 493 | +1.4% | 1.58 | -1.3% | 158 |
| 8 | DeMarker (14) | Daily | +193.3% | 0.62 | -35.3% | 35.6% | 295 | +0.9% | 1.53 | -0.5% | 85 |
| 9 | Fisher Center-of-Gravity | Daily | +113.4% | 0.51 | -29.4% | 44.4% | 601 | -0.5% | 1.52 | -2.6% | 191 |
| 10 | DeMarker (21) | Daily | +180.9% | 0.59 | -33.2% | 34.1% | 267 | +0.7% | 1.51 | -0.6% | 65 |
| 11 | Impulse MACD | Daily | +121.4% | 0.56 | -38.1% | 35.7% | 378 | -0.4% | 1.5 | -2.7% | 136 |
| 12 | CCI (14) | Daily | +173.7% | 0.56 | -27.4% | 36.8% | 359 | +0.6% | 1.41 | -1.2% | 107 |
| 13 | Random Walk Index | Daily | +169.1% | 0.56 | -30.4% | 38.7% | 297 | +0.5% | 1.37 | -1.1% | 90 |
| 14 | Random Walk Index | Daily | +169.1% | 0.56 | -30.4% | 38.7% | 297 | +0.5% | 1.37 | -1.1% | 90 |
| 15 | Aroon | Daily | +137.9% | 0.5 | -32.3% | 42.0% | 205 | -0.0% | 1.36 | -1.5% | 63 |
| 16 | Aroon Oscillator | Daily | +137.9% | 0.5 | -32.3% | 42.0% | 205 | -0.0% | 1.36 | -1.5% | 63 |
| 17 | Vortex | Daily | +176.3% | 0.58 | -30.2% | 39.7% | 295 | +0.6% | 1.34 | -1.3% | 89 |
| 18 | Projection Bands | Daily | +53.2% | 0.29 | -21.0% | 61.8% | 191 | -2.0% | 1.25 | -5.0% | 55 |
| 19 | Markov Regime | Daily | +124.5% | 0.46 | -23.7% | 66.7% | 48 | -0.3% | 1.21 | -0.1% | 13 |
| 20 | Connors RSI-2 | Weekly | >+999% | 0.21 | -27.0% | 57.4% | 47 | +21.0% | 1.21 | -5.1% | 15 |
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.