Financial Instruments Business Operator Kanto Finance Bureau Director (Kinsho) No. 1960 / Membership Association Investment Trusts Association, Japan Membership No. 012-02323 Go to GogoJungle Home
icon List
うさぎ先生
10/03/2026 18:30
Release： 10/03/2026 18:30
Update： 10/03/2026 18:30
うさぎ先生
Investment style : Trading System
Investment career : More than 10 years
Investment amount : 100,000 - 300,000 JPY
This is Usagi-sensei. Focusing on FX, I develop and validate indicators, trade assist tools, and automated trading EAs for MT5. Currently, I publish MT5 tools that make “market condition checks” and “multi-chart workflows” a bit easier in discretionary trading. When listing products, I strive to verify them in as close to real operating environments as possible, and clearly explain: - what can be done - what cannot be done - in what environments it was verified - what could not be verified I aim to provide tools that are easy to understand and use, not guarantees of profit or future price movements, but rather aids for checks and tasks during trading. If you have any questions about the products, please feel free to contact me.
Article cover image

When looking at EA, the first thing I don’t check is my win rate

When you open the introduction page for the EA, your eye is inevitably drawn to the “win rate.” If it says 80%, it somehow feels reassuring.

However, during research and testing of EAs, there are many things that cannot be understood from the win rate alone. This time, I will整理 the reasons for not looking at the win rate first and the items we look at together instead.

Win rate is only the ratio of winning trades

Win rate is the percentage of all trades that ended in profit. It does not include how much was won or lost per trade.

For example, compare the following two EAs.

  • EA-A: win rate 80%, average profit 1,000 yen, average loss 5,000 yen
  • EA-B: win rate 40%, average profit 3,000 yen, average loss 1,000 yen

If you make 10 trades, EA-A yields a total of 8,000 yen profit and 10,000 yen loss, for a net of -2,000 yen. EA-B yields 12,000 yen profit and 6,000 yen loss, for a net of +6,000 yen.

Note: The numbers are fictional for explanation and do not reflect real EA performance.

Even with a high win rate, a single large loss can make the overall result negative.

Items we look at together

When evaluating the win rate, we look at it in combination with the following items.

  1. Risk-reward ratio (average profit ÷ average loss): The balance between the size of wins and losses per trade
  2. Maximum drawdown: How much capital has been reduced at its worst point. Is this within what you can endure?
  3. Number of trades: If the count is small, the performance may have been a fluke
  4. Validation period: Does it include periods with different market characteristics?
  5. Profit and loss trend: Is the growth not just concentrated in a single period?

What to check when the win rate is high

Even with a high win rate, if the loss per trade is large, a few losses can significantly affect overall performance. The EA-A in the above example is of this type.

When looking at the win rate, it helps to also confirm the size of losses per losing trade to grasp the profit-and-loss pattern more clearly.

Note

This does not mean the win rate is meaningless. The win rate is an important indicator, but it cannot by itself determine risk or ease of operation. What matters is not to judge solely by the win rate.

What we value at the Research Institute

At Usagi-sensei MT5 Laboratory, we continue to develop and test EAs. In validation, we check not only the win rate but also the pattern of losses and drawdowns.

If we publish an EA, we want to ensure it has been thoroughly validated from these perspectives.