Help make the robot profitable - page 7

 
Yuriy Asaulenko:
This is the very question. No one needs these trend lines in history any more. You can't fill a glass, you can't make a fur coat. For some reason I associate it with collecting empty bottles - ah, what a cognac was there.

))) well said.
 

Friends, thank you very much for your comments.

From your point of view, what is the right thing for me to do to make my system profitable?

What do you think I need to do to make it profitable?

 
Евгений Панин:

Friends, thank you very much for your comments.

From your point of view, what is the right thing for me to do to make my system profitable?

What do you think I should do to make it profitable?

Two variants:
  1. Maybe I should change the indicator? I should leave the only one - the main one and use it as the basis for development of my system.
  2. Or even take two indicators - oscillator and trend indicator. I will develop two systems on the basis of these indicators in parallel and then combine them.
 
Евгений Панин:

Friends, thank you very much for your comments.

From your point of view, what is the right thing for me to do to make my system profitable?

What do you think I need to do to do this?


You will not have to think.

And to understand that all indicators work, every indicator has its own role and at the same time no indicator has predictive properties and that is solely you or your system.

So, it doesn't really matter which indicators to use as a base. What is important is the idea of the system, and there will be an indicator for it.) You may even find the right one for yourself.

 
Yuriy Asaulenko:


Thinking.

You should understand that all indicators work in their own way and at the same time, none of them has prognosticating properties, and it is you or your system that does it.

So, it doesn't really matter which indicators to use as a base. What is important is the idea of the system, and there will be an indicator for it.) After that, you may create the indicator you need.

I cleared my throat and barely stopped.

)

 
Yuriy Asaulenko:


Think about it.

You should understand that all indicators work in their own way and at the same time, none of them has prognosticating properties, and it is only you or your system that does it.

So, it doesn't really matter which indicators to use as a base. What is important is the idea of the system, and there will be an indicator for it.) After that, you may create the one that suits your needs.

Well said, dammit!

But in general, it is strange that the author started such a thread. SeemsEugene Panin, according to his profile, is a rather reasonable person who started reasonable topics and even completed one job. It is strange to hear such a question as "Why I have a loss, although the MA shows the trend and I entered it following the trend - what should I do to make the MA always show the trend accurately?

 
Евгений Панин:


OK, what exactly should I post?

I make the robot in Wizard MQL5.

The main signal is Larry Williams volatility breakout, when the distance from High to Close changes sharply up or down.

The problem is sideways. I try to sift them out, they do not work.

I added filter by averages - to buy when fast is higher than slow and vice versa. It did not help.

I added an ADX filter - to buy/sell when ADX is rising - did not help.

I still get long losing series.

I am sure there is a way to fix this - I would like to know it.


Afternoon! Specifically to "The problem is sideways. No matter how I try to weed them out, nothing works." - When I wrote the robot, it was trading on the plus side only when I opened with a trend, but signals against the trend and sideways more often gave me losing series. I tried to sift out these signals, but nothing worked, and I was working on it for weeks... Finally I made my trading robot semi-automatic - i.e. I introduced extern variable Trend (although for each currency pair I had to create my own robot that works specifically with this pair) and at the beginning of each day I simply draw all the possible support lines/supers/levels, etc. for each currency pair and if the trend changes, or a sideways channel, or there is news and it is better not to trade on specific pairs, then I change the value of variable Trend for each pair to the one that makes the robot not trade, or only trade in the direction of the trend.

 
Евгений Панин:


OK, what exactly should I post?

I make the robot in Wizard MQL5.

The main signal is Larry Williams volatility breakout, when the distance from High to Close changes sharply up or down.

The problem is sideways. I try to sift them out, they do not work.

I added filter by averages - to buy when fast is higher than slow and vice versa. It did not help.

I added an ADX filter - to buy/sell when ADX is rising - did not help.

I still get long losing series.

I am sure there is a way to fix this - I would like to know it.


With sideways you need to define the trend on the long term and sift out sell or bay, if you define the trend on a small timeframe then nothing will work from H1 trend look at the indicator i-trend
Files:
i_trend.mq5  10 kb
 
George Merts:

I don't quite understand why it's impossible to programmatically define support-resistance levels or a trend channel.

Of course, everything someone does with their hands cannot be programmed - because hands sometimes make completely unpredictable, in fact, random actions.

But, if the actions are clearly defined - why can't they be programmed?

Take support and resistance levels - what is so "unprogrammable" there? You take a fractal extremum and that is the level.

Or a channel - the same thing, take the maximum and the minimum for a certain period - here you have a channel. What cannot be programmed?

The only thing that cannot be programmed is "according to the left heel". When the trader takes a number from the ceiling and says "this is the level". Everything else is easier or more complicated, but programmable.

Do you really think so? It's probably a joke.
 

Made a filter on the MACD. I have noticed that during a sideways move the MACD fluctuates in a corridor. I consider it a trend when the MACD moves out of this corridor. In principle, it works as a theme.

Right now I am testing the variant when the short term average goes outside the Bollinger channel.

Then I want to make a channel for a period (as advisedby George Merts) and look for an exit from it.

These are the immediate plans

Reason: