Gold Reaper-style swing-level breakout (gold, 1-hour)
A trader reverse-engineered a gold EA that places stop orders past unbroken swing highs and lows. We tested the core idea as one 1-hour engine on five years of gold with real costs. With guessed settings it lost 16%; tuned, it made a thin +3.7% on six months held back from the tuning, while a daily trend filter that looked great in tuning lost on those months.
From r/Trading. The idea, in our words: Trade a break of an unbroken 1-hour swing high or low on gold: buy above the high, sell below the low, with a small stop and a much bigger target.
held upAs posted · the rules as described (see what we assumed below)
Optimized: best settings Quant Core found · last months held back as unseen data
Improved: only trade with the trend (price above/below the 1D 200 EMA)
We then tried 10 changes a quant would test first (800 settings in all) and kept the best on past data only. The held-back months played no part in choosing it. It did not hold up on the unseen months.
Ideas tried for this post
Each idea was tuned on past data only, then run on the held-back months (£ won per £1 lost).
| Idea | Tuning months | Unseen months |
| let winners run: a 3 ATR trailing stop instead of the fixed target | +36% · £1.10 | +4.5% · £1.14 · 110 trades |
| only trade breaks during London and New York hours (7:00 to 20:00 UTC) | +13% · £1.05 | +4.1% · £1.22 · 79 trades |
What we had to assume
- One engine, not nine: the 1-hour swing level is the highest high / lowest low of the last 30 bars (a level nothing has broken since)
- Enters on a 1-hour close beyond the level rather than a resting stop order placed a few dollars past it
- Stop 1 ATR, target 3x the risk (the post's 'small SL, much bigger TP'); up to 2 trades a day
A different reading of the rules could give a different result.
Backtested with the StratOS optimizer on historical data, with spread and costs charged, and the most recent months held back as unseen data. A backtest is not a forecast, and nothing here is investment advice.