Pacifica Carry Cycle
- エキスパート
-
Donizeti Aparecido Alves Ferreira Junior
Explore gold, currencies, US stocks and an index through a single portfolio instance in one MT5 hedging account, or choose individual EAs by market. Each EA page presents historical IS/OOS evidence in its research context. Try the free EA demos in the MT5 Strategy Tester.
Portfolio: D7 EA Portfolio. - バージョン: 1.4
- アクティベーション: 5
A carry approach to AUDJPY with a slow, regular rhythm. Pacifica Carry Cycle is a MetaTrader 5 Expert Advisor that takes long positions in AUDJPY in a recurring calendar cycle and holds them for several days, aiming at the risk premium of this pair. It suits traders who accept a long-only exposure, about two trades a month and flat weeks between them.
One position at a time, with protective stop and target on every trade and a time limit on each position. No grid, martingale, averaging or loss recovery. There is one setting: the risk per trade. The EA finds the AUDJPY symbol at your broker by itself (with or without suffix) and works from any chart. Entry rules stay proprietary; the published tests and the Market demo allow evaluation.
Historical test evidence
These are MetaTrader 5 Strategy Tester simulations, not live account results. Each test started with virtual USD 10,000, leverage 1:500 and 1.00% risk per trade, with fixed settings and no optimization.
Vantage demo data / Model 1 (1-minute OHLC); AUDJPY H1. In the in-sample test, JPY results in 2021 and part of 2022 were converted to USD with daily USDJPY data, as complete intraday conversion history was not available; the out-of-sample test is not affected.
| Window | Period | Net result (USD) | Trades | Profit factor | Max balance DD |
|---|---|---|---|---|---|
| In-sample | 2021-01-01 / 2024-12-31 | +347.87 | 103 | 1.07 | 7.23% |
| Out-of-sample | 2025-01-01 / 2026-10-20 | +935.67 | 47 | 1.54 | 3.22% |
The settings were fixed before the in-sample test, and the out-of-sample period was opened once, after it, without any change. The first in-sample run was stopped by the test machine before producing any result; the identical second run is the one shown. The same AUDJPY out-of-sample period had been read before for a different strategy; it was never used to choose these settings. The gallery joins the closed-trade balances of the two independent tests with a labelled cut; the out-of-sample test restarted at USD 10,000 and is only shifted for display. This is neither a continuous account test nor an equity curve.
Costs from the test ledgers (USD), included in the net result: in-sample commission −58.68, swap +152.17; out-of-sample commission −28.20, swap +71.44. Commission was 3 USD per lot per side. Swap was a credit for the long side in both windows; it uses current broker rates, depends on the interest difference between Australia and Japan, and can turn into a cost.
Material limits
The in-sample edge is small and uneven: profit factor 1.07, with 2022 and 2024 negative (2022 −175.41, 2024 −75.32 USD) and most of the result from 2021 (+595.89 USD). The out-of-sample window was better (2025 +720.25 USD, partial 2026 +215.42 USD); that does not promise the next period. All trades are long AUDJPY: the EA carries the risk of sharp yen rallies and risk-off moves, positions stay open over nights and weekends, and gaps can exceed the stop.
Risk and setup
Attach the EA to any chart. Keep AUDJPY available in Market Watch with synchronized H1 history and correct server time. Check contract size, tick value, volume step, margin, spread and trading sessions at your broker.
The tested risk was 1.00% per trade. The input accepts 0.10% to 5.00%; values above 1.00% were not measured in the published history and increase loss and drawdown. A small risk on a small account can fall below the broker's minimum lot; the EA then skips the trade instead of forcing a larger size. Protective orders do not guarantee the execution price.
Before renting
Test the demo in the Strategy Tester on your broker's AUDJPY data with the same risk and compare it with the published windows. With about two trades a month, a short test can show few or no trades.
