Does anything beat buy & hold on JPMorgan Chase (JPM)?
Every setup we tested on JPMorgan Chase (JPM) — 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: +30.1% CAGR over 2.9 years (+5.7% CAGR in the out-of-sample window).
Educational research from historical backtests — not investment advice. Past performance does not predict future results.
JPM: we tested 1,516 setups and none beat simply holding JPMorgan Chase
For JPMorgan Chase (JPM), we ran 1,516 indicator configurations against a plain buy-and-hold benchmark, and none of them earned the right to replace it. Individual stocks carry idiosyncratic risk — earnings surprises, guidance changes, sector rotation — that indicators built on price history cannot see coming. The best-looking setup, DeMarker on the 1-hour timeframe, ranked at the top in-sample but delivered an out-of-sample Sharpe of 2.8, short of our hurdle of 4.03. Buy-and-hold returned +30.1% annualized over the test period; the top strategies mostly captured pieces of that same move while adding trading friction.
How to read this honestly: when you test 1,516 setups and keep the best one, something will always look impressive by chance alone. That is why we require the survivor to clear 4.03 out of sample. DeMarker managed 2.8, with out-of-sample alpha of +35.3% across 0.9 years and 70 trades, and only 3.0% of everything we tested beat the benchmark on unseen data — roughly what luck would produce. None of this predicts anything: the regime that generated these numbers can shift, and single stocks shift faster than most. Treat this page as a record of what failed under honest testing, not a forecast.
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.
DeMarker
Mechanical rule (exactly as backtested): Buy DeMarker up through 0.3, exit above 0.7 — a demand/supply gauge. 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 2.8 · alpha +35.3% · 25 trades over 0.9 yrs.
Intraday Momentum Index
Mechanical rule (exactly as backtested): Open-to-close RSI — buy up through 30, exit above 70. 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 2.51 · alpha +34.7% · 21 trades over 0.9 yrs.
Order-Flow Reversion
Mechanical rule (exactly as backtested): Net-liquidity reversion — fades a 2-sigma price stretch only when signed-volume order-flow shows sellers are exhausted (imbalance z-score), exits on the mean. Needs real volume, so it abstains on feeds that don't report it. 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 2.19 · alpha +18.1% · 25 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 FLAT.
We tested 1,516 setups (indicator × parameters × timeframe) on JPMorgan Chase (JPM). Only setups with ≥30 trades qualify (1,379 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 1,516 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, 3.0% had positive out-of-sample alpha (median OOS Sharpe 0.23) — 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 1,379 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 | DeMarker | 1-Hour | +48.0% | 0.93 | -21.5% | 68.6% | 70 | -15.5% | 2.8 | +35.3% | 25 |
| 2 | Intraday Momentum Index | 1-Hour | +85.7% | 1.61 | -12.3% | 73.0% | 63 | -6.1% | 2.51 | +34.7% | 21 |
| 3 | Order-Flow Reversion | 1-Hour | +28.3% | 0.8 | -21.3% | 72.2% | 72 | -21.1% | 2.19 | +18.1% | 25 |
| 4 | VWAP Bands | 1-Hour | +28.9% | 0.72 | -23.3% | 72.0% | 75 | -20.9% | 1.98 | +18.4% | 27 |
| 5 | CCI | 1-Hour | +59.2% | 1.15 | -13.3% | 75.8% | 99 | -12.6% | 1.97 | +24.5% | 37 |
| 6 | Nadaraya-Watson Envelope | 1-Hour | +12.2% | 0.5 | -12.6% | 71.8% | 39 | -26.0% | 1.93 | +9.1% | 13 |
| 7 | RSI Mean-Reversion | 1-Hour | -1.1% | 0.03 | -25.5% | 67.6% | 34 | -30.5% | 1.68 | +11.9% | 13 |
| 8 | Stochastic RSI | 1-Hour | +33.4% | 0.78 | -17.3% | 58.6% | 152 | -19.6% | 1.67 | +18.6% | 50 |
| 9 | Keltner Mean-Reversion | 1-Hour | +27.1% | 0.7 | -20.1% | 66.7% | 63 | -21.4% | 1.67 | +14.2% | 24 |
| 10 | Williams %R | 4-Hour | +47.0% | 0.82 | -11.8% | 69.2% | 39 | -11.0% | 1.62 | +16.3% | 16 |
| 11 | Holy Grail Confluence | 1-Hour | +21.0% | 0.51 | -22.3% | 69.7% | 33 | -23.3% | 1.57 | +15.1% | 14 |
| 12 | Connors RSI | 1-Hour | +37.4% | 0.89 | -19.6% | 60.8% | 240 | -18.4% | 1.49 | +15.2% | 77 |
| 13 | Stochastic | 1-Hour | +20.4% | 0.48 | -19.5% | 69.5% | 82 | -23.4% | 1.41 | +18.7% | 30 |
| 14 | Connors RSI | 4-Hour | +14.3% | 0.34 | -22.3% | 60.6% | 66 | -17.9% | 1.34 | +11.2% | 28 |
| 15 | Stochastic Momentum Index | 1-Hour | +18.3% | 0.5 | -15.9% | 61.0% | 77 | -24.1% | 1.21 | +11.2% | 29 |
| 16 | Bollinger Mean-Reversion | 1-Hour | +12.9% | 0.41 | -22.3% | 67.5% | 77 | -25.8% | 1.17 | +6.8% | 24 |
| 17 | Fibonacci Bands | 1-Hour | +12.9% | 0.41 | -22.3% | 67.5% | 77 | -25.8% | 1.17 | +6.8% | 24 |
| 18 | Zero-Lag LSMA | Daily | +991.9% | 0.33 | -92.5% | 48.4% | 833 | -7.8% | 1.06 | -1.5% | 227 |
| 19 | Projection Bands | 4-Hour | +16.4% | 0.36 | -16.4% | 65.2% | 46 | -17.5% | 0.98 | +7.0% | 18 |
| 20 | Connors RSI-2 | 4-Hour | +18.2% | 0.46 | -16.8% | 61.8% | 68 | -17.1% | 0.97 | +6.2% | 26 |
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.