[Episode 12] I look at how I lose rather than my "win rate 90%".
When searching for an EA or indicator, you may come across a figure that says “90% win rate.”
A high win rate can look very appealing.
“This seems like I can use it with confidence.”
It's natural to feel that way.
However, after testing 22 years’ worth of USD/JPY data, I rarely look at the win rate first.
What I check first is,
how that logic can lose.
That’s where the critical insight lies.
Win rate tells you the number of losses.
But that’s all it tells you.
For example, even with a 90% win rate,
if the remaining 10% of losses would wipe out profits in a single loss, you cannot say it can be run safely.
Conversely, even if the win rate isn’t that high,
as long as the losses stay within a certain range and balance with profits, there is a possibility it can be run for a long time.
In other words, win rate alone cannot judge the strength of the logic.
What I want to know is “what happens after a loss.”
How long do drawdowns continue?
How does it recover from drawdown?
Is the pace of profit buildup stable?
Or does it rely on a single big win?
By checking these aspects, the character of the logic becomes gradually clearer.
I place more emphasis on “how the logic behaves after losses” than on a single number like win rate.