信号
添加小工具
规则
分配
| 交易品种 | 交易 | Sell | Buy | |
|---|---|---|---|---|
| GOLDm# | 2 | |||
|
1
2
|
1
2
|
1
2
|
| 交易品种 | 毛利, USD | 损失, USD | 利润, USD | |
|---|---|---|---|---|
| GOLDm# | -5 | |||
|
5
10
15
20
|
5
10
15
20
|
5
10
15
20
|
| 交易品种 | 毛利, pips | 损失, pips | 利润, pips | |
|---|---|---|---|---|
| GOLDm# | -522 | |||
|
500
1K
1.5K
2K
2.5K
3K
3.5K
4K
|
500
1K
1.5K
2K
2.5K
3K
3.5K
4K
|
500
1K
1.5K
2K
2.5K
3K
3.5K
4K
|
- 入金加载
- 提取
基于有关不同交易商真实账户的执行统计的平均滑移点按点数指定。它取决于 XMGlobal-MT5 19 提供商以及订阅者之间不同的报价，以及订单执行的延迟。值越低意味着复制的质量越高。
无数据
What this is
Live account track record of the Gold GodTier Dual Neural expert advisor. One instrument: gold. One timeframe: M15. Every trade shown here was placed by the same compiled EA published on the Market — nothing manual, no discretionary overrides.
How it trades
Two neural models run on every closed M15 bar: an LSTM reading 60 bars of normalized OHLC, volume and ATR, and a gradient-boosted tree reading the most recent bar. A position opens only when their combined confidence passes 0.59. Entry is at market. Stop loss and take profit are set immediately from current ATR (1.6x and 4.0x). The position is then managed in stages — partial close at 1R and 2R, break-even move, and an ATR trailing stop.
Risk control
- One position at a time. No grid, no martingale, no averaging down.
- Every position carries a hard stop loss from the moment it opens.
- Maximum loss per trade is capped in the code at 2% of equity.
- Position size halves automatically after a 10% drawdown and is cut to one quarter after 20%.
- Daily loss limit: 5% of equity.
- Average holding time is roughly 2 hours. Positions are rarely carried overnight.
Before you subscribe — read this part
Gold spread is the single biggest factor in whether this strategy works. It is a short-horizon system with a modest average edge per trade, so execution cost matters more here than in most strategies. I run it on a low-spread gold account where the measured 30-day average spread is about 30 points (0.30 USD). On an account with a 55+ point gold spread, this strategy does not survive its own costs.
Check your broker's gold spread before subscribing. If it is wide, your result will not resemble mine, and that is a cost problem, not a copier problem.
Honest notes
- This account started in September 2026. Judge it on live history, not on claims.
- Drawdown is part of the design, not a malfunction. Testing of this logic across past data shows drawdowns of 25-30% are possible.
- Past performance does not predict future results. Subscribe only with money you can afford to lose