Does anything beat buy & hold on Hang Seng?
Every setup we tested on Hang Seng — 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 only beat buy-and-hold in one window — a regime artifact, not a strategy. Buy-and-hold benchmark: +5.9% CAGR over 39.6 years (+0.2% CAGR in the out-of-sample window).
Educational research from historical backtests — not investment advice. Past performance does not predict future results.
Hang Seng: Nothing Beat Buy-and-Hold, and That Is the Honest Answer
Broad, diversified instruments like Hang Seng are where indicator strategies go to disappoint. We ran 761 setups against HSI, and none cleared the bar once scored honestly — on data the strategy never saw. The best of the batch, Bullish Marubozu on the weekly timeframe, posted an out-of-sample Sharpe of 0.58, short of the 1.06 hurdle we require before calling anything real. For an index fund this is the expected result: whatever inefficiency exists in single names tends to average away in the basket, leaving buy-and-hold's +5.9% annualized return as the number nothing here managed to beat.
Read these figures with the selection problem in mind. Test 761 indicators, keep the best, and the winner looks impressive by construction — which is exactly why the hurdle exists instead of applause for a lucky draw. Here, only 29.4% of setups outperformed buy-and-hold even in-sample, and the top candidate produced +3.7% annual alpha over 11.9 unseen years, across 74 trades with a 60.8% win rate and a -27.5% drawdown. That pattern reads as noise, not signal. Markets also change, so even a genuine past edge can fade. This page documents what failed — useful to know before assuming something works.
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.
Bullish Marubozu
Mechanical rule (exactly as backtested): Full-bodied bull candle (no wicks) — enter long, hold a short horizon. Signals are evaluated at weekly-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 0.58 · alpha +3.7% · 13 trades over 11.9 yrs.
Laguerre RSI
Mechanical rule (exactly as backtested): Ehlers' fast, smooth RSI — buy up through 0.2, exit above 0.8. Signals are evaluated at weekly-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 0.48 · alpha +3.9% · 21 trades over 11.9 yrs.
Nadaraya-Watson Envelope
Mechanical rule (exactly as backtested): Causal Gaussian kernel-regression envelope (LuxAlgo concept) — long when price dips below the lower band. 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 0.44 · alpha +3.1% · 22 trades over 11.6 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 2026-06-29. Currently FLAT.
We tested 761 setups (indicator × parameters × timeframe) on Hang Seng. Only setups with ≥30 trades qualify (697 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 761 tries mines even the holdout, the VALIDATED verdict additionally requires the top setup’s OOS Sharpe to clear a selection hurdle of 1.06 (√(2 ln N)/√T) AND positive alpha in both windows. Of the eligible setups, 29.4% had positive out-of-sample alpha (median OOS Sharpe 0.01) — 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 697 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 | Bullish Marubozu | Weekly | +342.7% | 0.44 | -27.5% | 60.8% | 74 | -2.1% | 0.58 | +3.7% | 13 |
| 2 | Laguerre RSI | Weekly | +403.5% | 0.49 | -23.9% | 63.5% | 63 | -1.8% | 0.48 | +3.9% | 21 |
| 3 | Nadaraya-Watson Envelope | Daily | -12.8% | 0.04 | -57.9% | 62.0% | 50 | -6.4% | 0.44 | +3.1% | 22 |
| 4 | Ultimate Oscillator | Daily | +105.6% | 0.2 | -58.1% | 75.0% | 40 | -4.2% | 0.42 | +4.4% | 12 |
| 5 | DEMA 20/50 Cross | Weekly | +799.7% | 0.43 | -39.2% | 56.2% | 32 | -0.2% | 0.37 | +4.2% | 10 |
| 6 | Schaff Trend Cycle | Weekly | +5.5% | 0.06 | -49.3% | 54.4% | 57 | -5.8% | 0.37 | +2.6% | 18 |
| 7 | Bollinger 50 (x2.5) Break | Daily | +34.8% | 0.24 | -10.5% | 48.8% | 86 | -5.3% | 0.37 | +1.4% | 27 |
| 8 | Adaptive Supertrend | Weekly | >+999% | 0.52 | -35.2% | 51.4% | 70 | +1.2% | 0.36 | +3.7% | 18 |
| 9 | KAMA 10/30 Cross | Daily | >+999% | 0.51 | -35.3% | 47.2% | 127 | +0.6% | 0.35 | +3.5% | 31 |
| 10 | RSI Mean-Reversion | Daily | +29.3% | 0.12 | -61.4% | 64.5% | 62 | -5.4% | 0.35 | +3.0% | 21 |
| 11 | Holy Grail Confluence | Daily | +64.8% | 0.17 | -44.7% | 64.5% | 31 | -4.8% | 0.34 | +3.4% | 15 |
| 12 | Keltner 20 Break | Daily | +395.6% | 0.55 | -25.3% | 41.6% | 361 | -1.8% | 0.33 | +1.8% | 87 |
| 13 | T3 8/21 Cross | Weekly | +471.0% | 0.36 | -50.5% | 59.5% | 37 | -1.4% | 0.32 | +3.3% | 11 |
| 14 | Ehlers Relative Vigor | Weekly | +521.0% | 0.37 | -50.6% | 50.0% | 172 | -1.2% | 0.32 | +3.3% | 53 |
| 15 | DeMarker (21) | Weekly | +403.0% | 0.33 | -47.1% | 55.1% | 69 | -1.8% | 0.32 | +3.3% | 18 |
| 16 | Coral Trend | Weekly | +447.9% | 0.34 | -57.4% | 55.3% | 47 | -1.6% | 0.32 | +3.2% | 12 |
| 17 | DEMA 100 Trend | Weekly | +636.1% | 0.43 | -34.4% | 49.3% | 69 | -0.8% | 0.32 | +3.1% | 19 |
| 18 | Donchian 55/20 | Daily | +168.1% | 0.26 | -41.8% | 44.3% | 88 | -3.5% | 0.32 | +2.6% | 21 |
| 19 | Hull Suite | Weekly | +542.5% | 0.37 | -48.0% | 55.0% | 40 | -1.1% | 0.31 | +3.3% | 12 |
| 20 | WMA 10/30 Cross | Weekly | +304.7% | 0.3 | -56.4% | 54.2% | 48 | -2.3% | 0.31 | +3.0% | 12 |
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.