The figure of "30% monthly interest" hides the drawdown

“Achieving 30% monthly return.” I will explain how to read this display. If 30% monthly return is real, compounding for one year would multiply the principal by about 23 times, and in two years by 540 times. 1 million yen would become 500 million yen in two years. This is a performance that World-wide institutional investors have not achieved, so at this point it’s clear that this is a non-sustainable figure. In other words, the monthly return display isinterpreted as the best month extractedfrom the period selection I described in a previous article—the monthly version.

And what isn’t written next to the monthly return is the maximum drawdown (how much funds fell in the worst moment). Operating to target a 30% monthly return inevitably involves a significant drawdown somewhere. Showing only returns and not drawdown is like a method that shows only the accelerator’s performance and hides the brake’s failure.

What you should check are these two things: what is the average number of months for that monthly return (is it a single-month extraction?). What is the maximum drawdown in percentage? A monthly return that has no answer to these two questions is not a numerical truth but a performance.

The validation results for ‘GENSEN-AI’ are not presented as monthly returns. They are expressed as win rate with the number of decisive trades as the denominator and total pips. The reason for using pips is that profit or loss can be manipulated by lot size and money management. Pips are not easily manipulated. I deliberately chose a unit that is less prone to manipulation and makes the numbers less inflated.