Does anything beat buy & hold on Natural Gas (NG)?
Every setup we tested on Natural Gas (NG) — 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 in both windows — but can't be told apart from selection luck.
Beat buy-and-hold in both the full window and out-of-sample but its OOS Sharpe 1.38 did not clear the 2.42 selection hurdle (best-of-N luck cannot be ruled out). Buy-and-hold benchmark: +0.5% CAGR over 8.2 years (+2.5% CAGR in the out-of-sample window).
Educational research from historical backtests — not investment advice. Past performance does not predict future results.
Natural Gas: the winner beat buy-and-hold in both windows — and still can't shake the luck problem
In Natural Gas, buy-and-hold is a harder benchmark than it looks: roll costs drag on returns, macro regimes flip the rules mid-game, and prices can go nowhere for years. Against that backdrop, we ran 1,519 indicator configurations on NG. The best, Laguerre RSI on the 4-hour timeframe, beat buy-and-hold in both windows — +18.6% annualized edge over the full period, +37.5% out of sample against a buy-and-hold CAGR of +0.5%. Over 2.5 out-of-sample years it took 105 trades, won 63.8% of them, and drew down -24.6% at worst.
Here is the honest read. When you test 1,519 setups and keep the best one, the winner looks good by construction — even on pure noise. Our correction for that selection effect sets a Sharpe hurdle of 2.42, and this setup's out-of-sample Sharpe of 1.38 falls short. Statistically, we cannot distinguish it from the luckiest of hundreds of tries. It also has company: 14.7% of setups edged buy-and-hold, which tends to happen when one regime dominated the sample. Of everything tested, 1,406 produced enough trades to grade at all. Macro regimes rotate, and a rule tuned to one rarely survives 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.
Top setups as mechanical rules
Exactly as the backtest defined them — no discretionary steps, no hidden filters.
Laguerre RSI
Mechanical rule (exactly as backtested): Ehlers' fast, smooth RSI — buy up through 0.2, exit above 0.8. Signals are evaluated at 4-hour-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 1.38 · alpha +37.5% · 31 trades over 2.5 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 1-hour-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 1.32 · alpha +19.6% · 25 trades over 2.3 yrs.
VWAP Bands
Mechanical rule (exactly as backtested): Rolling VWAP ± standard-deviation bands — buy below the lower band, exit at VWAP. Signals are evaluated at 1-hour-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 1.31 · alpha +36.5% · 88 trades over 2.3 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 1,519 setups (indicator × parameters × timeframe) on Natural Gas (NG). Only setups with ≥30 trades qualify (1,406 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 1,519 tries mines even the holdout, the VALIDATED verdict additionally requires the top setup’s OOS Sharpe to clear a selection hurdle of 2.42 (√(2 ln N)/√T) AND positive alpha in both windows. Of the eligible setups, 14.7% had positive out-of-sample alpha (median OOS Sharpe 0.02) — 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 1,406 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 | Laguerre RSI | 4-Hour | +321.5% | 0.97 | -24.6% | 63.8% | 105 | +18.6% | 1.38 | +37.5% | 31 |
| 2 | Nadaraya-Watson Envelope | 1-Hour | +114.5% | 0.78 | -14.6% | 52.9% | 85 | +10.1% | 1.32 | +19.6% | 25 |
| 3 | VWAP Bands | 1-Hour | +68.3% | 0.41 | -38.5% | 51.4% | 282 | +6.6% | 1.31 | +36.5% | 88 |
| 4 | Murrey Math Lines | 1-Hour | +88.8% | 0.43 | -40.3% | 60.3% | 136 | +8.3% | 1.24 | +35.5% | 46 |
| 5 | Bollinger Mean-Reversion | 1-Hour | +43.4% | 0.32 | -35.2% | 53.2% | 280 | +4.4% | 1.18 | +33.8% | 84 |
| 6 | Fibonacci Bands | 1-Hour | +43.4% | 0.32 | -35.2% | 53.2% | 280 | +4.4% | 1.18 | +33.8% | 84 |
| 7 | MA Envelope | 1-Hour | +144.0% | 0.59 | -40.3% | 53.4% | 161 | +12.0% | 1.13 | +34.4% | 50 |
| 8 | Intraday Momentum Index | 4-Hour | +151.2% | 0.53 | -37.4% | 63.0% | 54 | +11.4% | 1.08 | +30.8% | 18 |
| 9 | SMC: Liquidity Sweep | 4-Hour | +142.2% | 0.49 | -42.3% | 68.8% | 48 | +10.9% | 1.05 | +35.0% | 17 |
| 10 | WaveTrend (8/6/4) | 4-Hour | +267.5% | 0.67 | -30.9% | 61.4% | 57 | +16.7% | 1.05 | +33.3% | 21 |
| 11 | VWAP Bands | 4-Hour | +133.7% | 0.55 | -25.0% | 66.7% | 63 | +10.4% | 1.03 | +26.1% | 20 |
| 12 | Markov Regime | 4-Hour | +24.4% | 0.23 | -45.1% | 50.0% | 136 | +2.2% | 1.03 | +15.9% | 43 |
| 13 | Connors RSI-2 | 4-Hour | +88.2% | 0.45 | -36.2% | 60.8% | 189 | +7.5% | 0.99 | +18.4% | 58 |
| 14 | Projection Bands | 1-Hour | -27.5% | -0.01 | -64.6% | 52.2% | 515 | -4.7% | 0.98 | +28.8% | 153 |
| 15 | Laguerre RSI | 1-Hour | +30.2% | 0.27 | -48.9% | 50.4% | 377 | +3.1% | 0.93 | +23.3% | 120 |
| 16 | Ehlers Roofing Filter | 1-Hour | +21.3% | 0.24 | -55.5% | 39.3% | 244 | +2.1% | 0.91 | +31.5% | 68 |
| 17 | Stochastic Momentum Index | 4-Hour | +230.9% | 0.67 | -28.5% | 68.5% | 73 | +15.2% | 0.9 | +24.1% | 25 |
| 18 | Zero-Lag LSMA | 4-Hour | +56.6% | 0.32 | -42.0% | 39.5% | 228 | +5.1% | 0.88 | +27.3% | 66 |
| 19 | DeMarker | 4-Hour | +23.4% | 0.22 | -37.4% | 58.0% | 50 | +2.1% | 0.85 | +14.2% | 19 |
| 20 | Demand Index | 4-Hour | +173.1% | 0.57 | -29.5% | 63.2% | 125 | +12.5% | 0.82 | +18.8% | 46 |
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.