Does anything beat buy & hold on 20Y Treasuries (TLT)?
Every setup we tested on 20Y Treasuries (TLT) — 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 1.35 did not clear the 4.03 selection hurdle (best-of-N luck cannot be ruled out). Buy-and-hold benchmark: -5.3% CAGR over 2.9 years (-1.0% CAGR in the out-of-sample window).
Educational research from historical backtests — not investment advice. Past performance does not predict future results.
20Y Treasuries (TLT): Stochastic Momentum Index Beat the Index, But Couldn't Beat the Odds
Broad index funds like 20Y Treasuries are diversified by construction, which makes them stubbornly hard to beat — most of what looks like edge in a backtest is just the market's own return, rearranged. Here the picture is genuinely mixed. Out of 1,507 indicator configurations we tested on TLT, the strongest was Stochastic Momentum Index on the 1-hour timeframe. It outperformed buy-and-hold in both the training and out-of-sample windows, with a profitable trade profile across 102 out-of-sample trades. That is more than most setups on this asset manage, and less than what we would call validated.
Read the figures with the selection problem in mind. An out-of-sample Sharpe of 1.35 sounds fine until you remember it was chosen as the best of hundreds of attempts; our hurdle of 4.03 exists precisely to discount that luck, and this setup did not clear it. The out-of-sample alpha of +7.5% over 0.9 years, a 61.8% win rate, and a -17.3% drawdown describe one historical path, nothing more. Only 29.7% of setups beat buy-and-hold at all here, while buy-and-hold itself compounded at -5.3%. Market regimes shift; past performance carries no promise about 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.
Stochastic Momentum Index
Mechanical rule (exactly as backtested): Buy oversold SMI crosses up below -40, exit above +40 — a refined stochastic. 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 1.35 · alpha +7.5% · 34 trades over 0.9 yrs.
Stochastic
Mechanical rule (exactly as backtested): Buy oversold %K/%D crosses under 20, exit above 80 (14,3,3). 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 1.34 · alpha +8.5% · 37 trades over 0.9 yrs.
Bollinger Mean-Reversion
Mechanical rule (exactly as backtested): Buy the lower band, sell the middle — fade stretches below the 20/2 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 0.96 · alpha +5.6% · 31 trades over 0.9 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 1,507 setups (indicator × parameters × timeframe) on 20Y Treasuries (TLT). Only setups with ≥30 trades qualify (1,318 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 1,507 tries mines even the holdout, the VALIDATED verdict additionally requires the top setup’s OOS Sharpe to clear a selection hurdle of 4.03 (√(2 ln N)/√T) AND positive alpha in both windows. Of the eligible setups, 29.7% had positive out-of-sample alpha (median OOS Sharpe -0.59) — 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 1,318 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 | Stochastic Momentum Index | 1-Hour | -10.3% | -0.46 | -17.3% | 61.8% | 102 | +1.6% | 1.35 | +7.5% | 34 |
| 2 | Stochastic | 1-Hour | -12.0% | -0.44 | -21.5% | 57.4% | 108 | +1.0% | 1.34 | +8.5% | 37 |
| 3 | Bollinger Mean-Reversion | 1-Hour | -12.5% | -0.65 | -19.6% | 53.0% | 83 | +0.8% | 0.96 | +5.6% | 31 |
| 4 | Fibonacci Bands | 1-Hour | -12.5% | -0.65 | -19.6% | 53.0% | 83 | +0.8% | 0.96 | +5.6% | 31 |
| 5 | RSI Mean-Reversion | 1-Hour | +11.2% | 0.58 | -7.8% | 70.6% | 51 | +9.1% | 0.95 | +5.1% | 16 |
| 6 | Keltner Mean-Reversion | 4-Hour | -0.4% | 0.02 | -13.2% | 60.0% | 35 | +3.9% | 0.91 | +5.4% | 11 |
| 7 | VWAP Bands | 1-Hour | -11.3% | -0.59 | -16.2% | 49.3% | 75 | +1.2% | 0.79 | +4.4% | 26 |
| 8 | Order-Flow Reversion | 1-Hour | -5.0% | -0.25 | -14.9% | 53.3% | 75 | +3.6% | 0.76 | +4.3% | 25 |
| 9 | Connors RSI | 4-Hour | -2.4% | -0.06 | -10.4% | 55.4% | 101 | +3.4% | 0.64 | +4.6% | 33 |
| 10 | Projection Bands | 4-Hour | -16.8% | -0.59 | -21.3% | 49.2% | 61 | -0.6% | 0.54 | +4.4% | 21 |
| 11 | Stochastic RSI | 4-Hour | -5.6% | -0.21 | -11.1% | 62.7% | 51 | +2.5% | 0.49 | +3.4% | 17 |
| 12 | Hull MA Trend | Weekly | +160.5% | 0.45 | -21.5% | 61.4% | 57 | +0.5% | 0.48 | +6.9% | 14 |
| 13 | Bandpass Oscillator | Weekly | +115.6% | 0.37 | -25.3% | 57.9% | 57 | -0.3% | 0.45 | +6.7% | 14 |
| 14 | Schaff Trend Cycle | Weekly | +22.6% | 0.35 | -3.9% | 57.6% | 33 | -2.7% | 0.45 | +3.2% | 11 |
| 15 | TEMA 10/30 Cross | Weekly | +184.4% | 0.51 | -20.1% | 54.8% | 42 | +0.9% | 0.42 | +6.2% | 14 |
| 16 | SMA 20/50 Cross | Daily | +109.0% | 0.34 | -24.3% | 44.3% | 61 | -0.5% | 0.41 | +6.5% | 17 |
| 17 | Inside-Bar Breakout | Daily | -20.3% | -0.17 | -31.4% | 46.7% | 152 | -4.6% | 0.41 | +4.4% | 37 |
| 18 | Derivative Oscillator | Weekly | +44.1% | 0.24 | -14.2% | 54.0% | 87 | -2.0% | 0.38 | +5.0% | 21 |
| 19 | Bollinger Mean-Reversion | 4-Hour | +2.1% | 0.11 | -11.9% | 52.9% | 34 | +4.6% | 0.38 | +3.2% | 10 |
| 20 | Fibonacci Bands | 4-Hour | +2.1% | 0.11 | -11.9% | 52.9% | 34 | +4.6% | 0.38 | +3.2% | 10 |
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.