The criteria for selecting EA is the "Recovery Factor"
When choosing an EA, the most important metric to focus on is the recovery factor (risk-reward rate) calculated from backtesting.
The recovery factor is
"Net profit ÷ Maximum drawdown"
derived.
The ideal benchmark is 10 or higher over 10 years.
However, the following conditions must be met.
・Backtesting conducted with sufficiently wide spread settings
・Trading frequency ensures a sufficient number of trades
■ Why is the recovery factor important?
Because, unlike the profit factor, the recovery factor is important: even if short-term forward-test performance is good, a low recovery factor means there is a possibility of large losses in the future.
If the recovery factor is low, even with good forward results in the short term, you may face significant losses later.
A recovery factor of 10 over 10 years implies, by simple calculation, an annual figure of 1.
In other words, if you suffer the maximum drawdown, it takes 1 year to recover.
If it is below 10, it will take even longer to recover.
■ Is the EA you are operating okay?
When you calculate your own EA or a commercial EA, you may find that many do not meet the standard.
The recovery factor is an indispensable metric for judging whether an EA can profitably accumulate profits over the long term.
Furthermore, if the backtest's profit/loss graph shows a consistently upward slope at all times, that becomes a major strength in your selection.
■ Differentiating between short-term EA and stable EA
The above discussion is a criterion for long-term operation.
A stable EA can be operated in parallel and left running for the long term,
while, meanwhile, you could experiment with a high-potential, small-lot EA for short-term bursts—this could be considered an acceptable strategy.
■ Checklist for backtesting
For reference, here are key points to check when evaluating an EA.
・Is the recovery factor sufficiently high?
・Is the backtest performed with a wide spread?
(For example, set 1.5 pips = 15 points for USD/JPY or EUR/USD, 3.0 pips = 30 points for GBP/JPY, etc.)
・Is the expected profit per trade not overwhelmed by the spread in pips?
・Is the profit/loss graph consistently rising to the right?
・Is it a long-term backtest of over 10 years?
・Is there a sufficient number of trades?
? An EA that clears this checklist can be said to be a genuine EA capable of long-term operation.
The recovery factor is
"Net profit ÷ Maximum drawdown"
derived.
The ideal benchmark is 10 or higher over 10 years.
However, the following conditions must be met.
・Backtesting conducted with sufficiently wide spread settings
・Trading frequency ensures a sufficient number of trades
■ Why is the recovery factor important?
Because, unlike the profit factor, the recovery factor is important: even if short-term forward-test performance is good, a low recovery factor means there is a possibility of large losses in the future.
If the recovery factor is low, even with good forward results in the short term, you may face significant losses later.
A recovery factor of 10 over 10 years implies, by simple calculation, an annual figure of 1.
In other words, if you suffer the maximum drawdown, it takes 1 year to recover.
If it is below 10, it will take even longer to recover.
■ Is the EA you are operating okay?
When you calculate your own EA or a commercial EA, you may find that many do not meet the standard.
The recovery factor is an indispensable metric for judging whether an EA can profitably accumulate profits over the long term.
Furthermore, if the backtest's profit/loss graph shows a consistently upward slope at all times, that becomes a major strength in your selection.
■ Differentiating between short-term EA and stable EA
The above discussion is a criterion for long-term operation.
A stable EA can be operated in parallel and left running for the long term,
while, meanwhile, you could experiment with a high-potential, small-lot EA for short-term bursts—this could be considered an acceptable strategy.
■ Checklist for backtesting
For reference, here are key points to check when evaluating an EA.
・Is the recovery factor sufficiently high?
・Is the backtest performed with a wide spread?
(For example, set 1.5 pips = 15 points for USD/JPY or EUR/USD, 3.0 pips = 30 points for GBP/JPY, etc.)
・Is the expected profit per trade not overwhelmed by the spread in pips?
・Is the profit/loss graph consistently rising to the right?
・Is it a long-term backtest of over 10 years?
・Is there a sufficient number of trades?
? An EA that clears this checklist can be said to be a genuine EA capable of long-term operation.
× ![]()
