Alchemy Duo, the true form of even-handed Nanpin — current unrealized losses and notice of free trial version
Serial: Yoko and Kenji’s GOLD EA Mastery Notebook Episode 4
Hello, this is Yoko.
We took a short break from the series. Regarding the official forward for Alchemy Rising Tide on September 28,here is our reportwe have summarized the facts. We apologize again for causing concern.
Because of that day, Kenji and I reconsidered how to present our EA going forward. From now on,we will focus on “not increasing lots with a grid (Nanpin)” and Alchemy Duoto talk about. This time, we will tell you the “true nature” of that Duo, including its good points and painful aspects.
Notice: We have公開 a free trial version of Alchemy Duo
From October 1,for 7 days, with no functional restrictionsyou can use a free trial version. The logic and safety systems are exactly the same as the product version.
Rather than immediately entrusting your own funds,first run it in your own environment and decide after you’re satisfied. That is why we prepared it. We'll discuss usage cautions in the latter part of the article.
“Standard lot size” Nanpin has the slowest drawdown growth
As explained in Episode 2, Nanpin uses a multiplier—the number by which to increase the lot each time you add to a position.
| Rising Tide | 1.1x |
| Alchemy Freedom (Stable) | 1.06x |
| Duo | No change in lot size (1.0x) |
No matter how many levels of Nanpin Duo stacks,the lot stays the same as at the start. Even if the market keeps moving in an unexpected direction, since the lot does not grow, the way the unrealized loss increases among the three is the most gradual.
However, it takes time to recover
That said, there is a drawback.
EA that applies a multiplier tends to hold larger lots in deeper positions, so even a small market retracement can make the average entry price much more favorable, allowing profits to turn quickly.
Duo is not like that. Because the lot does not increase,it takes longer to return to the entry price and turn profitable after a reversal. The waiting time is longer than the other two.
In other words, Duo has a clear character:
・The unrealized loss grows at the slowest pace
・Consequently, the recovery speed is the slowest
This is the true nature of Duo: “Tenacious against reversals, but you must be patient.”
And now, the official forward is in the very phase of waiting
To be honest, as of October 2, when I’m writing this article, the official forward for Duo carriesan unrealized loss of about 250,000 yendue to gold’s decline since late September, stacking up Nanpin.
This is precisely the scenario of “time to recover” mentioned above. Because the lot does not increase, even with more than 20 levels of Nanpin, unrealized loss remains around 250,000 yen. There is still a distance to the stop-loss brake (80万 yen in M5 mode).
However, if gold continues to fall, the brake may activate. If that happens, we will report the outcome in this series without hiding it.
Using a Nanpin EA means dealing with this kind of time. Therefore, choosing the next mode and understanding the brake amount is essential.
Which to choose, M5 or M15
Duo has two modes: M5 and M15. The settings you decide are only two things: this “mode” and the amount of the stop-loss brake.
| M15 mode | M5 mode (default) | |
|---|---|---|
| How price movement is perceived | Slow | Frequent |
| Trading frequency | About 1.6 times per day | About 5 times per day |
| Default brake amount | 650,000 | 800,000 |
| PF (12 years) | 1.99 | 1.60 |
| Brake activation over 12 years | 0 times | 1 time |
| Leverage 1000x Maintenance margin |
9,711% | 4,375% |
| Leverage 20x Maintenance margin |
196.33% | 38.95%※ |
Backtest from July 2014 to July 2026 with initial capital 1,000,000 and fixed lot 0.01.※The 20x leverage figures for M5 are based on increasing the brake amount from 800,000 to 950,000 yen.
For beginners, and for those with domestic accounts (about 20x leverage)
→M15 modeis recommended. Trading is leisurely, but there was not a single brake over 12 years, and even with 20x leverage, the margin maintenance rate remained at 196% for comfortable operation.
For those who want many trading opportunities or have high-leverage accounts
→M5 mode. More trading opportunities mean faster profit accumulation, but also more sensitivity to rapid market changes.
With a high-leverage account, there is plenty of margin
Duo was designed with the assumption of high-leverage accounts like 1000x. In this environment, the minimum maintenance margin remains4,000%–9,000% range, providing substantial leeway across 12 years.
However, one more thing I’d like to clarify to avoid confusion.
Having margin leeway does not mean your account funds won’t decrease.The “stop-loss brake” of Duo operates by the amount of unrealized loss.If you increase the lot size, the unrealized loss grows accordingly, and the brake will reach sooner.
Therefore, even on a high-leverage account,the basic rule is 0.01 lots relative to the initial capital of 1,000,000. Maintaining this ratio is the premise to reproduce the 12-year results on your own account.
Important note: If using M5 mode on a low-leverage account
As mentioned in the third episode’s addendum, here I’ll elaborate again.
Duo has a built-in mechanism, when unrealized loss reaches a certain amount,“stop-loss brake”that activates and closes all positions. It’s a safety feature to protect the account, not a forced loss cut by the broker.
When we re-ran backtests with leverage 20x,M5 mode, with the default brake amount (800,000), brake activates deeper than expected twice (Nov 2014 and Mar 2026). If we raise the brake amount to950,000, the brake activates only once in Mar 2026, and the results become comparable to high-leverage accounts.
However, in that case, the minimum maintenance margin could drop to38.95%. Some domestic brokers may force liquidation at maintenance margins of 50% or 100%. Please confirm your broker’s liquidation threshold. If you’re unsure,M15 mode is the safest choice. We will also update the manual and product page soon.
Please first observe Duo’s behavior with the free trial版
If you’ve read this far and feel curious, pleasetry the free trial版.
| Period | 7 days from purchase |
| Functions | No restrictions. Logic, safety devices, and dashboard are the same as the product version |
| Recommended account | Demo account |
| Recommended mode | M5 mode (default). Even in 7 days you can observe trading adequately |
In 7 days you cannot determine profitability. Duo is an EA that takes time to recover. Instead, with 7 days you can confirm the following:
・Your broker’s spread and execution frequency for entries
・What times or filters halt new entries (displayed on the dashboard)
・How Nanpin accumulates and intervals when the market moves against you
Especially the third point, in a volatile market like now,it’s a good opportunity to see “the tough moments” with real money, with your own eyesand if you feel you can handle it, consider the product version.
Important notice about the trial period end
After 7 days, the EA will automatically stop. It will not only stop new orders,but also positions you hold will not be settled automatically; positions remain. Please exercise caution when using on a real account. As the period ends, please remove the EA or manually settle positions.
Yoko’s Word
There are no flashy multipliers or promises of huge profits. But,Duo can fight most quietly and most stubbornly, I believe.
And now, that persistence is being tested. Therefore, instead of showcasing strong numbers for good times, I want you to see this current姿 and decide after trying the trial版 with your own eyes. That’s what I really wanted to convey today.
Next Episode Preview
Next time, we’ll discuss another weapon hiding in Duo,the “Profit-Seeking Mode”What changes when you just hold off on settlement timing a little? We’ll also share how to verify this with the trial版.
Thank you for reading to the end.
Yoko
※This article presents past verification results and forward performance and does not guarantee future profits. Investment decisions are your responsibility. The forward unrealized loss is as of October 2, 2026.