1. Important Notes
Generally, for any new or old EA, we always prioritize LIVE SIGNAL, REAL ACCOUNT, NOT DEMO to verify the EA's performance.
To understand whether the EA performed well in the past, and if the LIVE SIGNAL matches the past backtest (BT) results for personal estimation, we need to learn how to backtest correctly. This blog will guide you through backtesting my new EA, Diamond Ultimate.
First, you must accept that the BACKTEST results are only for reference, as they are in a testing environment, not a REAL one. It cannot replicate the real environment 100%, so an expected result of 80-85% is already very good.
2. Test Modes
There are three main test modes: Every Tick, Based on Real Tick, and OHLC 1 Minute.
People often mistakenly believe that Based on Real Tick is the most accurate mode. This is not entirely true; it also depends on the strategy of each EA.
To determine which of the three modes gives results closest to real live performance, we must have LIVE SIGNAL data for at least one month. We will then run all three modes and compare the actual results.
The mode that produces results closest to the real live signal will be the one we use for backtesting the strategy. This is straightforward and completely transparent.
3. BACKTEST
Download the EA here:
- MT4: https://www.mql5.com/en/market/product/122947?source=Site+Market+My+Products+Page
- MT5: https://www.mql5.com/en/market/product/122943?source=Site+Market+My+Products+Page
3.1. Actual Results After 5 Weeks of Running the EA
- Live signal start from 2024/08/21
- Start: 480 USD
- Profit: 63% - 302 USD
- DD: 11%
- Win-rate: 87.7%
3.2. BT Results = Based on Real Tick
- Live signal start from 2024/08/21
- Start: 480 USD
- Profit: 48 USD
- DD: 20%
- Win-rate: 82%
3.3. BT Results = Every Tick
- Live signal start from 2024/08/21
- Start: 480 USD
- Profit: 274 USD
- DD: 12.49%
- Win-rate: 91%
3.4. BT Results = OHLC 1 Minute
- Live signal start from 2024/08/21
- Start: 480 USD
- Profit: 253USD
- DD: 10.58%
- Win-rate: 91%
4. Conclusion
As you can see, the results of all three modes are inferior to the actual results (surprising, right?). The actual results are much better than the backtest results.
Among them, Based on Real Tick produced the least accurate results compared to the real live signal (contrary to what many might think).
The OHLC 1 Minute and Every Tick results are much closer to reality.
Therefore, the conclusion is that with EA Diamond Ultimate, you cannot use the Based on Real Tick mode for backtesting and obtaining estimates.
You should use the Every Tick mode or OHLC 1 Minute, and an estimate of 80% is sufficient.
I hope this article helps you understand how to backtest each EA. It’s possible that for different currency pairs, the Based on Real Tick mode may yield closer results, depending on each EA’s trading strategy.
