The principles of non-syndicator trading systems. - page 3

 
Georgiy Merts:

That's right. Instead of showing "unsyndicated trading" - you're talking about some "rights".

An unsyndicatorless expert is an expert who does not use indicators. That is, knows nothing about the price chart and its derivatives. For example, a news expert. We read the news release moment from the calendar and place a pending position at the current price with fixed TP-SL before the news. That's all. This is a candlestick-free bot - I don't know any others (well, if we exclude random opening and closing of trades).

And if this is your no-indicator bot, or other rendering bot, based on plum martins or turbulence, run through the optimizer, will they immediately turn into an indicator, because in fact the indicator of historical prices will already be used?)
 
Vitalii Ananev:

Have you tried impulse trading?

You don't need any indicators other than the price itself. We wait for a strong impulse (rapid change of the price in one direction. The size of the bar is times larger than the average value). Trades are opened in the direction of the momentum.

But there are difficulties:

Impulses occur at moments of false breakouts of some key level, and then, as a rule, the movement moves in the opposite direction from the impulse.

The price movement after the impulse can develop in different ways: immediately after the impulse the correction or the price moves immediately in the direction of the impulse and the correction occurs the next day or 2-3 days later.

You can find many examples on the historical chart. In the picture you can find one of the "murky" situations: after the momentum, the next day the correction almost absorbs the momentum, then the continuation of the flat movement 2 days and again the continuation of the movement.


I agree about the impulses. I also noticed that sometimes the robot slows a bit when the momentum is gone and then begins to consume. I made the robot detect the borders of the price corridor and find its quiet pace somewhere in the middle of the corridor.

I also noticed that the calmer the price is in the corridor, the stronger the spike or impulse will be.

 
Vitalii Ananev:

Have you tried impulse trading?

You don't need any indicators other than the price itself. We wait for a strong impulse (rapid change of the price in one direction. The size of the bar is times larger than the average value). Trades are opened in the direction of the momentum.

But there are difficulties:

Impulses occur at moments of false breakouts of some key level, and then, as a rule, the movement develops in the opposite direction from the impulse.

The price movement after the impulse can develop in different ways: immediately after the impulse the correction or the price moves immediately in the direction of the impulse and the correction occurs the next day or 2-3 days later.

You can find many examples on the historical chart. In the picture you can find one of the "murky" situations: after the momentum, the next day the correction almost absorbs the momentum, then the continuation of the flop 2 days and again the continuation of the movement.


Where the momentum continues, it could be its end... it often happens too...

 
Vitalii Ananev:

Have you tried impulse trading?

You don't need any indicators other than the price itself. We wait for a strong impulse (rapid change of the price in one direction. The size of the bar is times larger than the average value). Trades are opened in the direction of the momentum.

But there are difficulties:

Impulses occur at moments of false breakouts of some key level, and then, as a rule, the movement moves in the opposite direction from the impulse.

The price movement after the impulse can develop in different ways: immediately after the impulse the correction or the price moves immediately in the direction of the impulse and the correction occurs the next day or 2-3 days later.

You can find many examples on the historical chart. In the picture you can see one of the "murky" situations: after the momentum, the next day correction almost absorbing the momentum, then the continuation of the flat movement 2 days and again continuation of the movement.


It is impossible to catch the impulse during its formation.

You should open orders in advance.

for example, at the corridor boundaries.

But there are a lot of other factors to consider...

especially it concerns the cumulative loss of an average series of losing trades.

As for the trading robot, I don't really have enough nerves to watch it work, because it works properly, and even with all my experience I am often mistaken. To be honest I don't have enough nerves to watch my robot work, because it works correctly and even with all my experience I´m often wrong.

 
Martin Cheguevara:

where the pulse continues could be the end of it... that happens a lot too...

Yes, I agree. The end of a move may be accompanied by a strong momentum. Followed by a reversal, maybe immediately or after a flat. I think it has to do with closing positions. For example a move down. Then an impulse downwards, provoking the "meat" to sell, themselves buying at the same time closing short positions.

 
Vitalii Ananev:

Yes, I agree. The end of a move can be accompanied by a strong momentum. Followed by a reversal, maybe immediately or after a flat. I think it has to do with closing positions. For example a move downwards. Then an impulse downwards, provoking the "meat" to sell, themselves buying at the same time closing short positions.

honestly, we will never know) our job is to react to it properly and catch it in time.

My problem is that my robot often works at flat, i.e. the price doesn't move anywhere for a long time and jumps from one corridor border to another like mad...

except for the variation coefficient [https://studfiles.net/preview/5316293/page:3/] nothing helps so far...

I don't want my risk module to stop my trading and I need this coefficient to earn profit depending on the situation.
 
Martin Cheguevara:

We'll never know, to be honest) it's up to us to react correctly and catch it in time.

My scourge is that my robot often works in a flat, i.e. when the price doesn't move at all for a long time.

I don't know how to use this coefficient of variation [https://studfiles.net/preview/5316293/page:3/], but I've got no luck so far.

And even this ratio is needed to keep my risk module from stopping the trade and to earn profit depending on the situation within the risk limits.

I don't know how to use this coefficient. I am still tracking the impulses manually, using an indicator that analyses the size of the candle body, its shadows, and the direction of the candle relative to the ATR. I have an Expert Advisor for this strategy but it needs to be finalized. I stopped working on it at the moment. It seems to have good results but it needs periodic optimization due to constant market volatility. And the price does not always cover the same distance after the impulse. Therefore we need to come up with some kind of dynamic take profit. Trall is not suitable because then the stop to take profit ratio of 1 to 3 or higher cannot be achieved. If the ratio is 1 to 1 or less the Expert Advisor is losing in the long run.


 
Vitalii Ananev:

I don't know how to use this coefficient. I am still tracking the impulses manually, using an indicator that analyses the size of the candle body, its shadows, and the direction of the candle relative to the ATR. I have an Expert Advisor for this strategy but it needs to be finalized. I stopped working on it at the moment. It seems to have good results but it needs periodic optimization due to constant market volatility. And the price does not always cover the same distance after the impulse. Therefore we need to come up with some kind of dynamic take profit. Trall is not suitable because then the stop to take profit ratio of 1 to 3 or higher cannot be achieved. If the ratio is 1 to 1 or less, the expert will fail in the long run.


The risks are big and depending on the drawdown I see that your scheme is based on principle - a grid is opened along the trend... As I already wrote in other threads, the probability of a common event is accumulating, that is, a number of orders
But it will be good for deposit acceleration if initial deposit is about $150-200.
 
Martin Cheguevara:
The risks are big and the drawdown shows that your scheme is principled - a grid is opened along the trend... As I've already written in other threads, the probability of a common event, i.e., several orders are accumulated.
But for the acceleration of the deposit it is good if we deposit about $150-200.

In this picture the risk is 1% of the deposit per trade. That is, in the case of a loss, the loss of no more than 1% of the deposit, regardless of the size of the stop loss. That is, it does not matter how many points stop loss is. The amount of loss is regulated by the volume. If a stop loss reaches 100 points and gets triggered 1% of deposit, 50 points stop loss is also 1%. And as take profit is 3-4 times bigger, due to this the system does not lose money. Trades are opened immediately after the impulse, even if the previous trade is not closed. There may be several one-way deals as well as multi-directional deals.

But this is all test data the system is not perfect, it needs more work.

 
Vitalii Ananev:

In this picture the risk is 1% of the deposit per trade. That is, in the event of a loss, the loss of no more than 1% of the deposit, regardless of the size of the stop loss. That is, it does not matter how many points the stop loss is. The amount of loss is regulated by the volume. If a stop loss reaches 100 points and triggered 1% of deposit, 50 points stop loss is also 1%. And as take profit is 3-4 times bigger, due to this the system does not lose money. Trades are opened immediately after the impulse, even if the previous trade is not closed. There may be several one-way deals as well as multi-directional deals.

But this is all test data the system is not perfect and we should still work on it.

Ahh I see) it's a good topic for overclocking. And if you have more than one trade, how do you calculate 1%? Or for each order in the grid you have 1% then 1%*kol.orders?
I should try it with my system, I haven't set it that way yet... maybe it will work even better)
And how do you determine that 4 trades is the optimum value?
Reason: