Cable Hunter EA
- エキスパート
- バージョン: 100.1
- アクティベーション: 5
Turtle Soup Strategy — Counter-breakout + RSI Divergence | GBPUSD M15
| Profit Factor | Max Drawdown | Annual Return | Walk-Forward IS/OOS |
|---|---|---|---|
| 1.67 | 5% | ~15%/year | 1.51 / 1.53 |
Compatible: GBPUSD M15 | All MT5 brokers | FTMO / Prop firms | VPS
1. Strategy Description
Cable Hunter EA is based on the Turtle Soup strategy, a counter-breakout approach that exploits false breakouts of swing levels. Markets regularly trap traders who follow classic breakouts. Cable Hunter detects these traps and trades in the opposite direction.
The strategy targets a universal human behaviour — the stop-loss clusters of breakout traders. This structural behaviour does not disappear over time, which is why the edge remains valid across multiple market cycles.
4-Step Signal Logic
- Step 1 — Swing detection : identifies the last significant swing high or low over a configurable lookback window.
- Step 2 — Sweep detection : price briefly breaks the swing level (wick pierces), trapping breakout traders.
- Step 3 — RSI divergence : RSI at the sweep bar is less extreme than at the original swing bar — confirms momentum exhaustion.
- Step 4 — Rejection close : candle closes back on the other side of the level — signal validated. Entry at market on the next candle.
SL / TP
- Stop Loss : beyond the sweep wick + configurable buffer in pips
- Take Profit : opposite extreme of the consolidation range (natural R:R target)
2. Validated Results
| Period | Profit Factor | Max Drawdown | Trades | Status |
|---|---|---|---|---|
| 2023–2025 (3 years, reference) | 1.67 | 5.74% | 170 | Validated |
| Walk-forward IS 2024–2025 | 1.51 | 4.91% | 71 | Robust |
| Walk-forward OOS 2024–2025 | 1.53 | — | ~35 | Stable |
| Spread x2 (20 to 40 points) | 1.75 | — | — | Robust |
Walk-forward IS/OOS ratio : 1.013 — OOS performance slightly exceeds IS. Confirms the absence of curve-fitting. The strategy has a genuine statistical edge.
Annual Performance — $10,000 account
| Year | Estimated Profit | Return % |
|---|---|---|
| 2023 | ~$1,400 | ~14% |
| 2024 | ~$1,600 | ~16% |
| 2025 | ~$1,600 | ~16% |
| Average | ~$1,530/year | ~15%/year |
3. Parameters
In MT5, parameters are identified by their comment label shown in the interface.
Strategy Parameters
| MT5 Label | Default | Description |
|---|---|---|
| Swing lookback period (candles) | 10 | Candles to detect a significant swing |
| RSI period for divergence | 19 | RSI period for divergence confirmation |
| Minimum R:R ratio to accept trade | 2.2 | Minimum risk/reward — rejects setups that are too tight |
| M15 candles to block after SL hit | 16 | Cooldown after a loss — avoids revenge trading |
| Enable London / New York session filter | true | Restricts entries to liquid sessions |
Risk Management
| MT5 Label | Default | Description |
|---|---|---|
| Risk % of equity per trade | 0.5% | Fixed — do not modify |
| Maximum spread in points | 20 | Blocks entries if spread > 2 pips |
| Daily drawdown cut-off % | 2.5% | EA stops automatically if daily loss reaches 2.5% — FTMO criterion guaranteed by code |
| Total drawdown cut-off % | 8.0% | EA stops permanently if total DD reaches 8% |
Optional Filters
| MT5 Label | Default | Description |
|---|---|---|
| Enable ADX regime filter | false | ADX trend filter — see section 5 |
| ADX period | 14 | Used only if ADX filter enabled |
| ADX max level | 25 | Blocks entries if market too directional |
| Enable partial close at RR 1:1 | false | Partial TP — degrades PF, not recommended |
| Enable trailing stop | false | Trailing stop — not recommended, see section 6 |
4. Installation
- Copy the .ex5 file into MQL5/Experts/
- Restart MT5 or click Refresh in the Navigator
- Drag and drop Cable Hunter onto a GBPUSD M15 chart
- Enable Algo Trading in MT5 (top toolbar)
- Check Allow automated trading in EA properties
Recommended minimum capital : $1,000 (with RiskPercent=0.5%). Raw spread broker recommended (ECN/STP) — tested up to 40 points spread.
5. ADX Filter
The ADX filter is disabled by default as it slightly degrades performance in normal market conditions (PF 1.67 vs 1.34 with ADX on).
| Period | ADX=false | ADX=true (Period=19, Max=30) | Recommendation |
|---|---|---|---|
| 2023–2025 (normal market) | PF 1.67 | PF 1.34 | ADX=false |
| 2020–2022 (crisis) | PF 0.22 | PF 1.54 | ADX=true, Period=19, Max=30 |
If you experience a strongly trending market (COVID-type), activate the ADX filter with Period=19 and Max=30. Annual re-optimisation is recommended.
6. Partial TP and Trailing Stop
Both options are available but not recommended — tests consistently show they degrade Turtle Soup performance.
| Configuration | Profit Factor | Drawdown | Verdict |
|---|---|---|---|
| Fixed TP only (default) | 1.67 | 5.74% | Optimal |
| Partial close 50% at RR=1 | 1.42 | 4.50% | -15% performance |
| Trailing stop 20/5 pips | 1.42 | 7.06% | DD increases |
The Turtle Soup is a short, precise reversal. Price targets a specific level quickly then reverses. The fixed TP captures this move optimally.
7. Compatibility
| Criterion | Status |
|---|---|
| All MT5 brokers | Yes |
| Any symbol | Yes (optimised for GBPUSD) |
| Any timeframe | Yes (optimised for M15) |
| FTMO / prop firms | Yes — DD 5%, daily DD 2.5% guaranteed by code |
| VPS | Yes |
| Built-in daily DD protection | Yes — automatic cut-off at 2.5% |
| Built-in total DD protection | Yes — automatic cut-off at 8% |
8. Best Practices
- Do not over-optimise : default parameters come from a validated walk-forward. Changing them without reason degrades performance.
- Annual re-optimisation : a light re-optimisation on the last 12 months once a year is recommended.
- Monitor spread : during major news (NFP, Fed, BOE), Cable Hunter automatically blocks entries if spread exceeds MaxSpreadPoints.
- Demo first : always test 4–6 weeks on a demo account before going live.
- Adequate capital : with RiskPercent=0.5%, a 20-pip SL = ~$10 risk per trade on a $1,000 account.
Risk Disclaimer
Trading foreign exchange on margin carries a high level of risk and may not be suitable for all investors. Past performance is not indicative of future results. Backtested results do not guarantee future performance. Walk-forward validation reduces but does not eliminate the risk of over-optimisation. Never invest money you cannot afford to lose. The author is not responsible for any losses incurred as a result of using this software.
Cable Hunter EA v1.0 — GBPUSD M15 — 2026
