{
  "id": 10406,
  "rank": 1341,
  "name": "Twin Reaper FX",
  "aka": [
    "Twin Reaper"
  ],
  "family": "B",
  "subtype": "B1",
  "one_liner": "Two grid engines, Phantom for USD-strength voting and Wraith for Bollinger bands, each with its own parameters and stop-losses.",
  "market_regime": "any",
  "timeframes": [
    "H1"
  ],
  "symbols": [
    "GBPUSD"
  ],
  "sessions": [
    "null"
  ],
  "indicators": [
    {
      "name": "Bollinger Bands",
      "params": {
        "period": null,
        "stddev": null
      },
      "applied_to": "close"
    }
  ],
  "entry_long": [
    "close above the upper Bollinger band"
  ],
  "entry_short": [
    "close below the lower Bollinger band"
  ],
  "exit_rules": [
    "give-back trail"
  ],
  "filters": [
    "null"
  ],
  "stop_loss": "Basket SL (points) — close the basket if it runs this many points against the average entry; Equity-% SL — close the basket when its floating loss reaches this percentage of account equity.",
  "take_profit": "Basket TP 500 pts for Phantom",
  "position_size": "0.01 start lot • add every 500 pts • max 15 orders • martingale ×2 for Phantom; 0.01 start lot • martingale ×1.25 • max 20 orders/side for Wraith",
  "evidence_type": "backtest_shown",
  "completeness": 50.0,
  "signal_warnings": null,
  "claim_warnings": [
    "RECOVERY SIZING: position size increases after losses (martingale/grid/averaging). Equity curves look smooth until the run that does not recover - judge it on worst-case exposure, not win rate."
  ],
  "opacity_warnings": null,
  "build_warnings": null,
  "canonical": "https://prismaticalgos.com/compendium/algo/10406-twin-reaper-fx",
  "snapshot": "2026-10-31 15:46 UTC",
  "source_url": "https://www.mql5.com/en/market/product/201925",
  "limitations": "These are machine-extracted research records, not verified trading systems. Catalog rank measures specification detail and claimed evidence, not profitability. The catalog extraction pipeline did not independently backtest these records. Its separate selected testing ledger does not validate the wider corpus. Rules, labels and parameters can be incomplete or wrong; consult the original source and its reuse terms."
}