Who is Risa Asahina?
An EA developer writes everything about herself

Nice to meet you. I’m Risa Asahina.

I develop and sell two gold (XAUUSD)–only EAs on Gogojan. A scalping-type “Shoukinryu” and a day-trading type “CHAOS_GOLD_LISKOFF_MT5.” Both are fully automated trading systems specialized only for gold.

In this article I will write everything about why I started EA development, what failures I’ve had, and how Shoukinryu was born. I believe understanding “the way this person who made this EA thinks” is more important than the specs on a product page.

I started FX with the simple motive of wanting to increase money. At first, discretionary trading. I looked at charts and traded based on my own judgment.

The results were terrible.

When I had a floating loss, I prayed it would come back. When I made a profit, I thought I could push it further and it reversed. Even with rules for stop-loss, I froze in the moment. Repeating trades ruled by emotions,I incurred losses of over 1,000,000 yen

What I learned then was that “my emotions are the biggest enemy in trading.” No matter how much knowledge I have or how well I know the rules, emotions break them.

So I decided to eliminate emotions. Have the program follow the rules instead of humans. That’s how I entered EA development.

I learned MQL4 and MQL5 on my own. Starting with a simple EA that enters on moving-average crossovers. I still remember the thrill of making something that actually moved.

But “moving” and “winning” are entirely different. Even if backtests show profit, in real trading I don’t win much. I repeatedly created and destroyed EAs that were over-optimized to past data and couldn’t adapt to future markets.

Then I realized“Only logic that remains stable over long backtests works in real life.”An EA optimized over three years can break in the fourth year. An algorithm that doesn’t break in five years can break in seven.

So I set a rule to build “EAs that can endure backtests for at least ten years.” I pursued long-term, steady stability rather than short-term flashy numbers. This principle is the origin of Shoukinryu’s design philosophy.

I have developed EAs across many currency pairs such as USDJPY, EURUSD, GBPJPY, etc., but the most stable results came from gold (XAUUSD).

Gold has structural characteristics not found in other currency pairs.

“Buying-focused strategies work, RSI contrarian strategies work, liquidity is high.” These three characteristics are found only in gold. For that reason I decided not to create EAs for anything other than gold.

“Specializing in one currency pair” might seem risky from a diversification standpoint. But I believe the opposite. A single asset studied thoroughly and designed with a complete understanding of its characteristics is far more trustworthy than one that superficially covers multiple pairs. Shoukinryu’s 11 years of stability prove the correctness of this idea.

The development of Shoukinryu involved a long process of trial and error. It didn’t start with RSI2 + ATR dynamic averaging in the first place.

At first I built it with RSI14 (a common setting). It was mediocre. Reactions were too slow to catch short-term pullbacks in gold.

Then I shortened the RSI period. 10, 8, 5, 3… and finally “2.” RSI2 is often criticized as “unconventional” and “noisy.” But testing on the gold M5 chart showed that RSI2 most accurately captures the moment when prices are oversold.

I rewrote the averaging logic many times. Fixed-interval averaging could blow up the account in sudden volatility. If intervals are too wide, entries become too few. Finally I arrived at a method that measures real-time volatility with ATR and automatically adjusts the averaging interval according to market roughness.

Then backtesting. From 2015 to 2026, 11 years and 22,189 trades. Through the Covid crash, rapid rate hikes by the Fed, and geopolitical risks. Through it all, the results were win rate 84.28% • PF 1.93 • RF 9.95 • max drawdown 12%. When these numbers appeared, I was finally confident: this is something that can be released to the world.

Run it on my real account.This is a minimum condition. Selling to others without using it yourself is like serving customers food you won’t eat. Shoukinryu and CHAOS_GOLD are running on my real accounts every day, and I stream their activity 24 hours a day on YouTube Live.

Over ten years of backtesting.It’s easy to show a clean upward curve over three years. But that might just mean it’s optimized for three years of past data. Whether the logic is genuine can only be judged after going through more than ten years of diverse market environments.

Publish all numbers.Backtest results, live trading performance, and drawdown status. Good days and bad days are posted daily in the Investment Navigator daily reports. Even days with drawdowns over 50,000 yen are reported openly. Transparency is the sole source of trust.

Respond honestly to buyers.If asked to increase entry frequency, I say honestly that you could change parameters to increase, but I don’t recommend it. The default settings have been the most stable for 11 years. If asked whether profits can be guaranteed, the answer is no—FX carries risk. Not lying to buyers is the only way to build long-term trust.

Shoukinryu is a buy-focused M5 scalper. It targets only the rebounds after gold’s oversold moves, a very sharp design in a sense.

Meanwhile, CHAOS_GOLD_LISKOFF_MT5 is a H1 day-trading type. It supports long and short positions and takes advantage of the gold market’s time-based anomalies (patterns tend to appear around 21:00, 18:00, and 04:00 Japan time).

Because the design philosophy is entirely different, the two EAs target gold without overlapping positions. Running both in parallel allows risk diversification across timeframes and strategies.

CHAOS_GOLD’s backtest results are PF 2.96 • RF 11.49 • win rate 72.07%. It’s an EA with strengths different from Shoukinryu.

I will continue to write daily reports for Investment Navigator.I will publish daily results of both EAs in the Daily Reports and share the EA design philosophy and FX education content in serialized articles and columns. I will record everything, good and bad. When I look back in a year or two, these accumulations will be the strongest evidence of trust.

I will continue daily YouTube live streaming.I will continue to stream Shoukinryu and CHAOS_GOLD running 24/7. I will maintain an environment where anyone can verify whether the backtest numbers are real with their own eyes.

I also provide EA development services.MT4⇔MT5 source code conversion service (10,000 yen), EA/indicator creation service (from 15,000 yen), base code templates for EA development. I also support people who want to build their own EAs.

EA is not “buy and end”; it’s “buy and start.” I respond to buyers’ questions and concerns via Gogojungle’s messaging feature within 24 hours. As a single EA developer, my job is to keep creating an environment where buyers can operate with confidence.

I am not a genius programmer, nor a legendary trader. I was a typical FX trader who once lost over 1,000,000 yen for not being able to cut losses.

However, I reached a rule-based system that eliminates emotions, thoroughly studied the characteristics of gold, and built an EA that endures ten-plus years of backtesting.

Shoukinryu’s numbers—win rate 84% • PF 1.93 • RF 9.95 • max DD 12%—are the product of my failures and trial-and-error. If reading this makes you feel “this is an EA I can trust,” that would mean the most to me.

This article is for information purposes and is not investment solicitation. The performance results listed are past results and do not guarantee future profits. FX/CFD trading involves risk. Please make investment decisions at your own risk.