The true value of a semi-discretionary EA: It closed the position without dragging to the TP, and immediately afterward a decline followed

In practice, I’ve been running the semi-discretionary EA “KURAMA GOLD SWING AUTO,” and there was a moment when I truly felt, “This is the beauty of semi-discretion.” I’m sharing it as is.
From the outset, the conclusion is—

I closed a position on my own judgment, earlier than the TP set by the EA.
And immediately afterward, the price dropped.

Entry on two timeframes, ended up with two positions

First, the situation.

This EA operates independently on M30, H1, and H4 timeframes.In this case, because conditions aligned on two timeframes, I ended up with two positions.

Here is one key point: each timeframe makes its own independent judgment, so in favorable moments multiple positions can be on.
It’s not averaging up. Each is an independent 1-signal-1-position.

Why I didn’t trail to the TP

There was already a substantial unrealized profit.

But looking at the chart, the situation was this.

・Approaching the previous high
・In a position where the higher timeframe MA acted as resistance

In other words, “From here, it’s more likely to be held back than to rise straight up.”

The EA’s TP was still a bit higher.
But I chose not to pull it that far and closed the position.

The reason is simple.
There was ample profit and a clear resistance right in front of me.
In that case, securing the profit is more favorable in terms of expected value than trying to push for more.

As a result, a drop came immediately afterward

And right after I closed.
Gold then declined from there.

If I had held to the TP, the unrealized profit would have been eroded.
There was a possibility that it could have retraced to the entry price.

In the end, early closure was the correct move.

Of course, it won’t always go so smoothly.
Sometimes cutting early leads to further extension afterward.
Still, the idea of “secure when resistance is near” makes sense in the long run, in my view.

This is the beauty of “semi-discretionary”

This is what I want to convey most today.

With a fully automated EA, you keep holding mechanically until the TP is reached.
That can be good, but in situations like this, where resistance is clearly near, you end up holding on as-is.

KURAMA GOLD SWING AUTO allows human judgment to intervene at that point.

・Full close
・Partial close (half-close)
・Cut at entry price
・Stop new entries

All of these can be done with one touch from the chart panel.

This time I did full close, but you could also choose “take partial profit and aim for the TP with the remainder.”
Having partial closes creates a middle ground option of “secure while letting it run.”

Part of it is automated, part of it you control.
Having both in play is the strength of a semi-discretionary EA.

Entrypoint and lot size are delegated, exit can be chosen

To summarize, this EA has the following role split.

Parts delegated to the EA
・Signal determination (grading S/A/B/★)
・Entry timing
・Lot sizing (auto size by risk %)
・TP/SL settings

Parts you can choose
・Whether to let it run or secure it
・Close all or half
・Escape at entry price

The trickiest and most error-prone part, “lot calculation,” is automated.
And the exit, which you judge by looking at the market, remains in your hands.

Because of this design, you can implement moves like in this case.

Summary

・Conditions aligned on two timeframes, resulting in two positions
・Sufficient unrealized profit plus resistance near the previous high and MA
・Did not pull to the EA’s TP, closed early
・As a result, a drop occurred immediately afterward. Early securing was the right move
・Full close, partial close, and exit at cost can be chosen with one touch, which is the strength of semi-discretionary trading
・Entrypoint and lot size are delegated, exit is chosen by you

Not fully automatic, not fully discretionary either.
Somewhere in between lies the most practical approach, in my opinion.

Purchase here (GogoJungle exclusive)

Due to verification, this EA is sold exclusively on GogoJungle.

https://www.gogojungle.co.jp/tools/indicators/82015