Gold Level Breaker
- Experts
- Version: 1.15
- Activations: 5
Gold Level Breaker trades one simple, transparent idea on XAUUSD: the breakout of yesterday's high or low. No grid, no martingale, no averaging down. Every trade has a fixed stop loss and take profit from the moment it opens, and the default risk is 0.4% of balance per trade. Every rule is an input you can see and test yourself.
This is a trend-following breakout system. Expect a low win rate: most trades are small losses, and the profit comes from a few large winners. It earns in years when gold trends strongly and gives little or loses in quiet or falling years (see the backtest section).
How it trades (default rules)
- Daily breakout entries: buys when price breaks above yesterday's high, sells when it breaks below yesterday's low. At most one entry per direction per server day.
- Trend filter: trades only in the direction of the D1 EMA(200) and only when D1 ADX(14) is at least 20. Both can be adjusted or turned off.
- Fixed exits by default: every trade opens with a stop loss of 5 USD and a take profit of 50 USD (per 0.01 lot, scaled with the lot size). By default the stop and target are not moved afterwards. One optional setting, Trailing stop, offers two fixed presets (see "Trailing stop" below).
- Position sizing: 0.4% of balance at risk per trade by default (also available: fixed lot, or lot per balance step).
- One position at a time: maximum 1 open position by default. A signal that appears while a position is open is skipped.
- Equity-curve filter: risk is halved while the EA's closed-trade equity curve is below its 20-trade moving average, and returns to normal when it recovers.
- Time rules (server time): no entries before 01:05; no new entries on Friday after 22:30; all positions are closed on Friday at 22:57.
- Risk controls (off by default): spread limit, daily loss limit, maximum lot, and an equity drawdown pause with optional cooldown.
Trailing stop (one setting)
InpTrailMode ("Trailing stop") is one setting with three fixed choices. There are no further trailing inputs to tune. All amounts are in USD per 0.01 lot and scale with the lot size.
- None (TRAIL_NONE): Default. The stop loss (5 USD) and take profit (50 USD) stay exactly as set when the trade opens.
- Tight trail (TRAIL_FIXED_USD): Once the trade is in profit by +1.00 USD, the stop loss follows the best price reached, at a distance of 0.50 USD.
- Break-even + steps (TRAIL_BREAKEVEN_STEP): Once the trade is in profit by +1.50 USD, the stop loss moves to the entry price +0.15 USD. After that, for every further +1.50 USD of profit the stop loss moves up by another 1.50 USD.
Backtest comparison of the three choices (Vantage XAUUSD, "Every tick based on real ticks", 0.4% risk, estimated slippage deducted; each year run separately, 2021 covers August-December only; figures are % of balance):
| Trailing stop | Sum of yearly returns 2021-2026 | Worst-year drawdown |
|---|---|---|
| None (default) | +114% | 9.3% |
| Tight trail | +10% | 2.9% |
| Break-even + steps | +22% | 3.3% |
None gave the highest backtest return but the equity curve swings more. The two trailing choices lock in profit early, so the equity curve is smoother, but the return is much lower. They are also sensitive to slippage, because every winning trade is small. These are backtest figures only and do not promise future profits.
Main inputs and defaults
| Input | Default | Meaning |
|---|---|---|
| InpTrendEMA | 200 | D1 EMA period: buy above, sell below (0 = off) |
| InpMinADX_D1 | 20 | Minimum D1 ADX(14) to allow entries (0 = off) |
| InpStopUSD | 5.00 | Stop loss, USD per 0.01 lot |
| InpTargetUSD | 50.00 | Take profit, USD per 0.01 lot |
| InpTrailMode | TRAIL_NONE | Trailing stop: None (default), Tight trail, or Break-even + steps (see "Trailing stop" above) |
| InpSizingMode | SIZING_RISK_PCT | Position sizing mode |
| InpRiskPercent | 0.4 | % of balance lost if the stop is hit |
| InpMaxPositions | 1 | Maximum open positions at once (0 = off) |
| InpEqFilter | true | Halve the risk while the equity curve is below its average |
| InpEqFilterLen | 20 | Equity-curve moving-average length (closed trades) |
| InpEqFilterCut | 0.5 | Risk multiplier while below the average |
| InpStartHour / InpStartMinute | 1 / 5 | No entries before this server time |
| InpCloseFriday | true | Close all positions on Friday (InpFridayCloseHour / InpFridayCloseMin = 22 / 57) |
| InpMaxSpreadPoints, InpDailyLossPct, InpMaxDrawdownPct, InpMaxLots | 0 (off) | Optional risk limits |
Requirements
- Symbol: XAUUSD (your broker's gold symbol)
- Chart: any timeframe. The EA reads D1 levels; tests used H1
- Account: low-spread / ECN account recommended; hedging or netting
- Minimum balance: the smallest trade is 0.01 lot, which risks about 5 USD at the default stop. On balances below roughly 1,250 USD the real risk per trade is therefore higher than 0.4%
- VPS recommended
Strategy Tester backtest results
These are backtest results from the MetaTrader 5 Strategy Tester. They are not results of real trading. Past performance does not guarantee future results.
Test settings: Gold Level Breaker v1.15 with default inputs, XAUUSD (Vantage), H1 chart, "Every tick based on real ticks" model, 10,000 USD initial deposit, one continuous run from 2020.10.21 to 2026.09.24.
| Run | Trades | Net profit | Profit factor | Maximum drawdown |
|---|---|---|---|---|
| Strategy Tester, no slippage simulated | 468 | +14,640.93 USD (+146%) | 1.74 | 9.31% (equity) |
| Same period, estimated slippage (see below) | - | about +123% | - | 11.4% (closed-trade equity) |
The Strategy Tester does not simulate slippage. The second row is an estimate made outside the tester by deducting 0.10 USD at entry and 0.30 USD on stop-loss exits. It is an estimate, not a tester result, and real execution can differ.
Return by calendar year (estimated with slippage, compounded, as % of the balance at the start of each year):
| 2021 | 2022 | 2023 | 2024 | 2025 | 2026 (to 24 Sep) |
|---|---|---|---|---|---|
| -1.6% | -3.0% | +7.8% | +12.7% | +72.0% | +11.5% |
What these numbers do not show well:
- 2021 and 2022 were small losing years, and most of the profit comes from 2024-2026, when gold trended strongly. A year like 2025 should not be expected as normal.
- Long stretches below the equity peak happen: in the backtest the account once spent more than 2 years below a previous high.
- For comparison, simply buying and holding gold over the same period returned roughly +121%, with a maximum price decline of about 27%. The EA's backtest drawdown was much smaller, but its return was roughly the same as holding.
- The default rules were chosen using 2021-2026 data, so this is not an independent out-of-sample test. Results depend on broker data, spread, execution and account settings.
Before you start
Test with "Every tick based on real ticks". Check your broker's gold contract size, trading hours, spread and server timezone. The default entry and Friday times assume a GMT+2/GMT+3 server. The spread, daily loss, maximum lot and drawdown limits are off by default; set them for your own risk policy. If you enable the drawdown pause, set a cooldown above 0 unless you want the EA to stop permanently.
Trading gold carries a risk of loss. With a fixed stop and a far target, the win rate is low and losing streaks are normal, so size your risk accordingly. Start on demo and check that the EA's behaviour fits your risk tolerance.
