01Your rules
We write your rules back so a computer can follow them, and you confirm them before anything runs.
| Market and timeframe | USDJPY, 4H chart, all sessions |
| Bias | Trendlines from the Monthly down to the 4H, drawn from xxxxxxxxxx swings |
| Setup | A 4H candle closes past a live trendline. Two zones open: the broken line and the xxxxxx gap |
| Entry trigger | 4 triggers at a zone (xxxxxxxx, xxxxxx, xxxxxxx, xxxxxxxxx). Entry at the trigger candle's close |
| Stop | Behind the most recent xxxxxxx swing. Skip the trade if under 20 or over 120 pips |
| Exit | Trailing stop, xxxxxxxxx after +3R. No fixed target |
| Limits | Max 1 buy and 1 sell open. Line zone 3 trades, gap zone 2 trades, 60 candles each |
| Risk | 1% of the account per trade |
02Test setup
Every 4H candle is processed in order, using only closed candles. No peeking at the future.
03Costs
Costs are taken off every trade. Many strategies that "work" stop working here.
04Results
Equity curve, in R
At 1% risk per trade, $1,000 would have grown to about $5,620.
05Stress tests
| Scenario | Total profit |
|---|---|
| Costs 0 pips | +210.8R |
| Costs 1.5 pips (base) | +191.4R |
| Costs 3 pips | +172.0R |
| Costs 4 pips | +159.0R |
| Miss the 5 best trades | +101.6R |
It survives double and nearly triple the costs. Missing the 5 best trades halves the profit, so never skip setups.
06Consistency
| Year | Trades | Total | Win rate |
|---|---|---|---|
| 2020 | 79 | −0.2R | 30% |
| 2021 | 88 | +26.2R | 41% |
| 2022 | 78 | +55.0R | 31% |
| 2023 | 73 | +6.2R | 30% |
| 2024 | 63 | +52.7R | 33% |
| 2025 | 92 | +15.3R | 36% |
| 2026 (to Oct) | 82 | +36.2R | 32% |
6 of 7 years made money. 2020 was flat and 2023 was weak, so expect slow years.
07Luck check
We ran 500 versions with random entries but the same stops, trail and costs. Only 1 in 500 beat the real system. Random entries still made money because the trailing stop rides USDJPY's trends, so the real edge is the gap: about +135R.
08Other markets
| Market | Trades | Total | Verdict |
|---|---|---|---|
| USDJPY (TradingView) | 555 | +191.4R | Edge |
| USDJPY (broker candles) | 602 | +113.9R | Edge, weaker |
| EURUSD | 702 | +5.1R | Flat |
| EURJPY | 683 | −9.8R | Flat |
| GBPJPY | 665 | −38.3R | Loses |
| GBPUSD | 732 | −96.0R | Loses |
Same rules, same period. Only USDJPY had an edge, and it was about 40% weaker on a second data feed. Trade it on USDJPY only.
09Risk plan
| Risk per trade | $1,000 becomes | Worst drop in test | 1-in-20 worst drop |
|---|---|---|---|
| 1% (recommended) | $5,621 | 14% | 30% |
| 2% (max) | $22,752 | 26% | 52% |
| 3% | $69,879 | 36% | 67% |
The 1-in-20 drop comes from reshuffling the same trades 10,000 times. Above 2%, an unlucky run could take half the account.
10Honest verdict
Real edge on USDJPY. Fragile, so follow every rule.
- It survives higher costs, a luck check and a second data feed.
- 10 of 555 trades made 77% of the profit. Missing setups is the biggest risk.
- No edge on other pairs. Risk 1% per trade, 2% at most.
11Disclaimer
Backtest results, not a promise of future profit. Not financial advice. Not included: slippage beyond the cost above, news spikes, swap fees, and execution delays. Live results can be worse than a backtest.
What we need from you: the 11 core items
- Market and timeframe
- Trading hours or sessions
- Bias: how you decide buy or sell
- Setup: what must be in place
- Entry trigger and entry price
- Stop loss
- Exit or target
- Limits: max trades, max open
- Risk % per trade
- 2–3 example trades (screenshots help)
- Your broker's costs (spread, commission)
Plus, for this package
- A second data source: your broker's candles
- Other pairs you'd consider trading