Previous-week high breakout
NAS100, XAUUSD · 4H, 1H chart · strategy s16_prev_week_breakout · 4 variations
Buys when price breaks above last week's high, with a very tight stop and a target ten times the amount risked.
What it is about
Last week's high is a price level many traders watch. When price closes above it, the move often keeps going: sellers who bet on that level holding are forced out, and breakout buyers jump in.
Many breakouts fail, though. So the stop is very tight, only a quarter of an average day's range, which keeps each failure cheap. The target is 10R (R = the amount risked), so one winner pays for nine losers. Wins are rare (15-20% of trades) but large.
This is the project's 'tight-stop module': research found that a tight stop with a 10R target works for breakout and trend-day setups, and fails for pullback or buy-the-dip setups.
How it trades
Gold: 4-hour candles, about 2 trades a month (1 for the wider-stop version). NAS100: 1-hour candles, about 3 to 4 trades a month. These are swing trades: they can stay open for hours or up to three weeks, including overnight and weekends.
- Only 15-20% of trades win on the 10R versions: strings of 15 to 25 losses in a row are normal and do not mean the strategy is broken.
- Gold default, backtest 2018-today at zero cost: 206 trades, +229.6R, largest drawdown about 15.8R, every calendar year positive.
- NAS100 10R version, backtest 2019-today: 259 trades, +162.7R, largest drawdown about 19.5R, every year positive.
- Results are lumpy: a year depends on a handful of 10R winners (gold 2025 +79.7R, but 2021 +6.1R).
- A long run of losses is hard to sit through, and the real temptation is to stop just before the next big winner.
- Buys only, with a daily uptrend filter, tested during a long gold bull market. A falling market gives fewer signals and more failed breakouts.
- Overnight financing (swap) costs are not included; they matter more for trades held for days or weeks.
- Tight stops are sensitive to spread and slippage, which these zero-cost results leave out: check your broker's spread in the calculator.
- Few trades per year on gold (13 to 38), so any one year can look much better or worse than average.
Where it comes from
A classic breakout level (last week's high) combined with Victor's tight-stop module, researched in October 2026. Train = 2018-2023 (NAS100 from 2019), test = 2024 to Oct 2026.
Main variation: s16-A ★ Main
The variations
Each variation is the same strategy with a different approach, trading its own paper account. Click one to see exactly what it changes.
The default: gold on 4-hour candles with a quarter-ATR stop and a 10R target. It accepts many small losses to catch the breakouts that turn into big trends.
s16-BGold 4H, full-ATR stopSame signals, but the stop is four times wider (one full daily ATR) with the same 10R target. Fewer trades get knocked out by normal noise, so the win rate is higher, and because trades last longer, fewer new ones are taken.
s16-CNAS100 1H, quarter stop, 10RThe same tight-stop, 10R idea on NAS100, using 1-hour candles so the breakout is spotted sooner on a faster-moving index.
s16-DNAS100 1H, 2R targetNAS100 again, but with a much closer 2R target instead of 10R. Wins come about three times as often, trading the big home runs for a smoother ride.
Variations side by side
Live result of all 4 together: 0.0R (zero cost).
| Live paper account (R, zero cost) | Backtest 2018 – today (R, zero cost) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| s16-A★ MainPrevious-week high breakout | XAUUSD | 4H | – | 0 | – | – | +229.6R | +1.11R | 9/9 | 1.9 | 20% | −15.8R |
| s16-BPrevious-week high breakout | XAUUSD | 4H | – | 0 | – | – | +67.3R | +0.69R | 8/9 | 0.9 | 36% | −10.9R |
| s16-CPrevious-week high breakout | NAS100 | 1H | – | 0 | – | – | +162.7R | +0.63R | 8/8 | 2.8 | 15% | −19.5R |
| s16-DPrevious-week high breakout | NAS100 | 1H | – | 0 | – | – | +129.2R | +0.34R | 8/8 | 4.1 | 45% | −10.5R |
Main = the variation with the most backtest R per R of max drawdown (2018 – today, zero cost). It is picked on that history, so the live accounts are the real test.