Daybreak
XAUUSD · M15 · MetaTrader 5 Expert Advisor
Coming soon. This Expert Advisor is in final testing and is not for sale yet. Its description, manual and research are published so you can follow it before release.
Read the manualHow it was built
Headline backtest
Profit factor 2.37 · worst equity drawdown 6.4% · 159 trades · return x1.62 on 10,000
Backtest, real ticks, Fusion Markets, 2024-01-02 to 2026-09-26, default settings, version 1.01. The weakest of the live brokers tested, not the strongest.
Figures on this page come from the MetaTrader 5 Strategy Tester on real broker ticks. They are backtests, not live results.
Up before the market, Daybreak chalks yesterday's high and low on the wall and waits at the line. When price crosses it, he goes with the first burst and is out again in seconds - one level, one trade, and a quiet day if nothing breaks.
Daybreak Gold is a breakout system for gold. Every trading day it sets two prices from the previous day's high and low. A break through either level opens one position - never more than one at a time - with a stop loss and a take profit measured from the level itself, so every trade carries the same risk distance regardless of the fill. Once a position is far enough in profit, a trailing stop takes over and gives back only a fixed amount from the best price reached.
The strategy skips new entries on the day of a US employment report, a Federal Reserve rate decision, and the thinly-traded days around the year-end close. It also holds new entries on a day whose prior 24 hours moved too little, unless the break reaches into price the market has not traded in five days.
The results below come from gold in 2025 and 2026, a high-volatility period; the same rules run on gold from 2018 to 2024 were flat to slightly negative. This is a system for a moving gold market, not an all-weather one.
No grid. No martingale. No averaging into a losing position. One risk setting (Risk level: Low, Medium or High) controls position size as a percent of balance per stop; every other input is open for review.
Backtest, MetaTrader 5 Strategy Tester, real ticks, 100% history quality, $10,000 deposit, 2025-01-02 to 2026-09-25, fixed 250 ms execution delay, compounding:
| Risk level | Feed | Net growth | Profit factor | Max equity drawdown |
|---|---|---|---|---|
| Medium (default) | BlackBull | +52.9% | 2.49 | 4.23% |
| Medium (default) | Fusion | +70.2% | 2.77 | 6.21% |
| Medium (default) | Pepperstone | +285.3% | 5.96 | 6.76% |
| Low | BlackBull | +23.4% | 2.50 | 2.11% |
| Low | Fusion | +30.5% | 2.84 | 3.08% |
| Low | Pepperstone | +95.5% | 6.34 | 3.38% |
| High | BlackBull | +132.8% | 2.47 | 8.41% |
| High | Fusion | +185.3% | 2.65 | 12.23% |
| High | Pepperstone | +560.7% | 5.38 | 9.69% |
High risk is margin-fit adjusted for a 1:20 (ESMA retail) gold account - the shrink is what keeps these trades executable at all on such an account. Losses come in runs: in the same backtests, five to seven consecutive stop-outs occurred about once a year, costing roughly 3% at Low, 7-11% at Medium and 12-20% at High, depending on the broker's tick stream. These figures are backtests on historical data; they are not a promise, guarantee or projection of future results.
Frequently asked questions
Does it hedge, martingale or grid? No. One position at a time, always with a stop loss.
What symbol and timeframe? Gold only. Any chart timeframe works - the strategy reads the daily bar and a fixed M15 clock internally regardless of the chart it is attached to.
What controls risk? The Risk level input (Low / Medium / High), or a custom risk percent / fixed lot under Risk level = Custom.
Does it work on a prop-firm account? Account type includes Prop firm 2-step and Prop firm 1-step presets that add a daily-loss and maximum-loss guard sized to typical challenge rules; a Prop-firm rules section lets you enter your own firm's exact numbers.
Why doesn't it trade every day? Two levels a day, at most one trade per level, and several days are held entirely (thin prior-day range, a scheduled calendar event). This is by design.
Evidence
