10/3【EA】Why do I develop multiple EAs? — An EA portfolio considered by the double-EA of Reiwa
Hello.
This is the era of Double-Ee in Reiwa.
This time, I would like to write about the concept of a “portfolio” that I value when developing EAs.
First of all, this is not saying “this is how you should operate.”
It is, above all,
my viewpoint on EA development and operation, based on the idea that I develop EAs with this way of thinking and run them by combining multiple EAs
.This is my own approach to EA development and operation.
I used to be a discretionary trader
The origin of my EA development lies in discretionary trading.
While looking at the chart,
“Won’t there be characteristic price movements like this?”
“In such markets, would this entry method work?”
I think about these things and discover trading methods from chart patterns and price action.
Then I quantify that thinking and implement it into EA logic.
This is the basic style of my EA development.
Therefore, I did not start with the idea of making many EAs from the beginning.
By observing markets, identifying different price movements and characteristics, and creating logics for each, various EAs have emerged.
There are good times and bad times for a single trading method
This is something I have felt since my discretionary trading days.
That is,
“Relying on only one trading method cannot adapt to all market conditions.”
Some methods work well in trending markets but struggle in ranging markets.
Conversely, approaches that perform well in ranges may not function well in strong trends.
Also, for some logics, there may be few market conditions that meet the entry criteria, resulting in few entry opportunities.
I believe this remains true even after becoming an EA.
That’s why I don’t think,
“Let’s capture everything with a single EA in any market.”
I don’t think that way.
Each EA should have its own role.
That’s how I’ve come to think.
Targeting different market characteristics with different logics
For that reason, I develop EAs with various approaches.
For example,
- Scalping
- Trend following
- Swing
- Anomalies
and so on.
Even within the same scalping category, there are logics that differ in the technicals used, filters, entry criteria, and exit conditions.
This is not the same as the idea of
“increasing the number of products to increase the number of EAs.”
In my case,
Find different characteristics and advantages in the market
↓
Consider them as trading methods
↓
Quantify and turn them into logic
↓
Validate as EAs
This is the development flow.
As a result, the number of EAs increases.
Even with different logics, you may enter at the same timing
This is surprisingly important when thinking about a portfolio.
For example, suppose there are two scalping EAs using completely different technical indicators or conditions.
Even if the logics differ, the market being watched is the same.
Therefore, when a characteristic price movement occurs,the entry conditions of both may become true around the same timingin some cases.
I think of this as somewhat similar to a baseball pitcher.
Even with the same “fastball” category,
there are four-seam, two-seam, and one-seam variants.
Even if it looks like the same fastball, grips, movement, and intended effect differ.
In EA, even within the same category of “scalping,” the underlying logics may not be the same.
And in some situations, multiple logics may react at similar timings.
On the other hand,price movements that can only be captured by one logicalso exist.
So I don’t think it’s enough to say,
However, “just different logic” isn’t enough
This is where, in portfolio thinking, I want to place importance.
Even if internal logics differ, if multiple EAs hold positions in the same direction at the same time in actual operation, risk may concentrate at the same timing.
There can be times when that pays off, and obviously times when it hurts at the same timing.
Therefore, I think not that
“diversification is achieved because the logics are different.”
but
we need to consider the portfolio while watching actual results.
For example, I consider from multiple perspectives, such as
- Profit and loss movements
- Simultaneous holdings
- Bias toward positions in the same direction
- Drawdown overlap scenarios
- Entry frequency
- Holding time
- Market environments each is good or weak at
and so on.
Aim for mutual complement rather than “one EA does everything”
My ideal is not to have one EA capture all price movements.
EA-A captures what it’s good at.
EA-B captures what it doesn’t fulfill in A.
In other market conditions, EA-C may work.
Sometimes A and B both meet their conditions at the same time.
But there are scenes where only A can capture, and others where only B can.
By combining EAs with different characteristics, I aim for operations that do not rely too much on a single logic.
This is the core of my EA portfolio concept.
The risk of simultaneous entries can also be improved by design
And now, I’m thinking of something else.
In particular, scalping is very popular, and I have developed multiple scalping EAs with different approaches.
However, even if different technicals and logics are used, if market conditions that confer an edge overlap, multiple EAs may hold positions at the same timing.
Therefore, for example, after loading multiple logics into one EA,
limit the maximum number of positions to one
to manage risk.
There are different logics A, B, C,
If only A’s conditions are met, enter with A.
If only B, use B.
If only C, use C.
If several are met simultaneously, pick only one based on a pre-set priority.
That way, the opportunities perceived by each logic can remain, while preventing positions from piling up in the same scenario.
Of course, this does not guarantee better performance.
Profit may decrease, or conversely, drawdown may be reduced.
Thus,
Form a hypothesis → Test → Look at results → Improve
This is the EA development cycle I want to reaffirm.
I don’t think there is a fixed portfolio
I have developed EAs for many years, but
I don’t believe “this is the absolutely correct portfolio.”
Markets change.
New logics appear.
Through actual operation, there are features I didn’t notice during development.
So I continue to develop my own EAs and manage them,
“Is there a better combination?”
“Is risk not concentrated somewhere?”
“Are there aspects one EA cannot capture that another EA can?”
and keep thinking.
This is the way of thinking I have arrived at and currently practice as I develop and operate EAs.
It does not mean it is suitable for everyone, nor that profits are guaranteed by this method.
However,
instead of entrusting everything to one logic, combine logics with different characteristics and operate while complementing their strengths and weaknesses.
This is the principle my current self, “Reiwa no Daru-Ee,” values when developing EAs and operating them.
From now on too, I intend to look at actual results and continue verification and improvement.
Reiwa no Daru-Ee
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
【Flow from starting EA to intermediate level】
■First, mastering “EA installation” is important.
--------------------------------------------------------------------
● MT4 EA (Expert Advisor) and Indicator installation steps
https://www.gogojungle.co.jp/post/1/8212
--------------------------------------------------------------------
● GoGoJungle video version*
--------------------------------------------------------------------
At first, many people cannot understand everything at once and may stumble.
I myself stumbled many times at first, so take your time and you’ll be okay!
If you have questions, there is a “Contact Form,” so please use it!
----------------------------------
■ GoGoJungle“Contact Form”
https://www.gogojungle.co.jp/inquiry
【How to choose an EA】
Now that you think you can operate an EA,
“I don’t know which EA will win… I don’t want to lose…”
Therefore, Reiwa no Daru-Ee has serialized articles to help
to determine quality via “backtests.”
I published the following articles.
--------------------------------
■ How to find winning EAs from backtests
https://www.gogojungle.co.jp/finance/navi/series/1714
----------------------------------
Forward results are the real performance, but the backtest is what supports them.Forward performance is important, but backtests back it up.
Backtests can largely indicate whether forward performance is plausible.
Usually, the backtest period is longer than the forward period,
and whether you are winning in forward just by chance can be inferred by rigorous backtesting.
【How should EA risk management be done!?】
And no matter how excellent an EA is,
neglecting risk management leads to ruin for EAs.
So for EA,how should risk management be done?
The article answering this is below.
----------------------------------
■ On calculating EA risk management
https://www.gogojungle.co.jp/finance/navi/articles/64703
----------------------------------
【EA operation mindset is also important】
And for sustaining EA operation for a long time, an “EA operation mindset” is also important.
Those articles are
----------------------------------
12/30★ Year-end and New Year SP!! All-EA Overview Course Part 3
https://www.gogojungle.co.jp/finance/navi/articles/80551
8/12 [Obon holiday special] Now you too can master EAs! “EA operation mindset”
https://www.gogojungle.co.jp/finance/navi/articles/72829
8/17 [Obon holiday special] Now you too can master EAs! “Notes during EA operation”
https://www.gogojungle.co.jp/finance/navi/articles/73797
----------------------------------
the content is substantial and quite important, so
even if you can’t understand it immediately, please make sure to read it!
Then!
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
【Reiwa no Daru-Ee EA Listing】
https://www.gogojungle.co.jp/users/112481/products
Specifically, categorized listing of Reiwa no Daru-Ee EAs
https://www.gogojungle.co.jp/finance/navi/articles/67162
(Genre diversification is essential for building a portfolio)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
【Reiwa no Daru-Ee serialized articles】
■ EA Beginner Course
https://www.gogojungle.co.jp/finance/navi/series/1700
--------------------------------
https://www.gogojungle.co.jp/finance/navi/articles/64738
https://www.gogojungle.co.jp/finance/navi/articles/65619
--------------------------------
https://www.gogojungle.co.jp/finance/navi/articles/64723
https://www.gogojungle.co.jp/finance/navi/articles/64703
--------------------------------
--------------------------------
Develop EAs with solid methods,
Gain the trust of more people,
Maintain a long-lasting WIN-WIN relationship,
Aim to revitalize the EA market and raise literacy.
https://youtu.be/JVdq3Hih3vg?si=zvGKUD0II0P_iv8z
