Once again, it's about the eternal: trend/flat. - page 6
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
The picture does not "open" incomprehensible what t\f, V2-N2 impulse, which was almost immediately "cancelled" by counter-movement (if the market has a good buy volume, it is interested to sell at a higher price - buy at a lower price, ie.After the break and movement downwards again (probably they have grabbed the orders and took them downwards - they are clearing the way) - variations on the theme of "gut feeling" )))) .
At lower prices where the sideways movement is 5-6 pips long, it may be advisable to "bind" to the level.
Let me clarify things a bit. Timeframe is implied in TF, and t/f implies trend or flat.
I saw a 5-minute chart in this picture. I apologize. The date is September 15.
In the market, "gut feeling" is our understanding of the market. Developed through years of practice.
Well, don't grieve....
On the merits of the topic - the problem of defining t/f on history is not a problem at all.
It is the problem of predicting t/f for the future.
And if you have solved the problem of predicting - there is no special problem in t/f.
P.S. In short, "...it's a no-brainer" (c).
It's never a shame for me to adopt or share the best practices of others. Even if I have accumulated experience in a field, I don't mind even the tiniest of new techniques.
It's a pity that no one on this forum takes it seriously, except for a few.
Arrogant ones like you Dimitri are in the majority, and they destroy any desire to have a discussion. The forum visitors prefer to read in silence and keep their heads down so as not to be spat upon.
My horoscope is Scorpio and I am not afraid of nobody's venom, I myself can sting very painfully. But where is the point?
the horoscope is starting to come on, yo, yo, yo, .....
The BB boundaries will show a flat with a lag of course.
In order to speak seriously on the topic the top starter would have to
1. Formalise the hypothesis of a flat / trend. I.e. to give a clear definition with the possibility of clear writing in the code and subsequent testing by anyone.
2. Show that the described phenomenon is statistically significant on a sufficiently long history. Say, a hundred percent one, a hundred percent another and a hundred percent intermediate forms.
And the MOST IMPORTANT thing is to prove that the given pattern indeed has predictive properties. For example. "We define this pattern as 50% (5%, 85%) enough time (bars, fractals, events) of its occurrence on a chart for the remaining 50% to work out with such a probability.
I'm sure that if the top-starter had done this work, a lot of questions would have fallen away on their own.
The human eye is so constructed that it instantly recognizes on any picture flats, trends, zodiac signs, the devil and smileys, without even noticing it. In order to remove the human factor one should just check his ideas in practice. I have personally checked for myself and seen that market fluctuations rarely produce recognisable events and are very often hybrid.In order to talk seriously about the topic, the topicstarter would need to
1. Formalise the hypothesis of a flat trend. I.e. give a clear definition with the possibility of clear writing in the code and subsequent verification by anyone.
2. Show that the described phenomenon is statistically significant over a long enough history. Say a hundred percent one, a hundred percent another and a hundred percent intermediate forms. 3.
And the MOST IMPORTANT thing is to prove that the pattern in question really has predictive properties. For example. "We define this pattern as 50% (5%, 85%) enough time (bars, fractals, events) of its occurrence on a chart for the remaining 50% to work out with such a probability.
I'm sure that if the topicstarter had done this work, many questions would have fallen away on their own.
And if you had switched your logic to ON mode, you would not have written this post because it would become clear that why would I create a topic if I know and know how to define t/f numerically?
What kind of people.... just to blurt something out...
The human eye is designed in such a way, that it instantly singles out flops, trends, zodiac signs, the devil and smiley faces in any picture, without even noticing it. In order to remove the human factor one should just check his ideas in practice. I personally checked and saw that market fluctuations rarely produce recognizable events and very often they are hybrid.
I don't know about your 'human view', but I don't apply it to automatic trading, and have written before:
...
At the moment I do it elementary - timeframes, say, 00:00-8:00 - flat movements, 08:00-22:00 - trend movements (one continuous directed movement or with change of direction several times), 22:00-00:00 - flat. But this simplified method is very approximate though it gives better results for TS, but does not allow to use it on TF older than H1.
Do you see? What "anthropogenic factor" is there, if I've clearly written that even a simple split of day and night already shows an improvement in indicators, it means that there is a difference in the behavior of price and not only in your "human eye".
I don't know about your "human view", but I don't apply it to automatic trading, and have written before:
See? What "human factor" is there yet, if I have clearly written that even a simple division of the day into parts already gives an improvement in performance, it means there is a difference in price behaviour and not only to your "human view".
A normal discussion should be like this.
Hypothesis - trending during the day, flat at night.
Check.
I do this and that during the day (code sample) andthis and that at night (code sample).
The results are as follows - here is a table of forwards on non-optimized areas for comparison.
Conclusion - it's really better at night this way, not that way. (Variant. I was wrong, it is the same at night). Any ideas for solutions?
Then there will be a subject for discussion. For now it's all just lyrics. And for the time being I do not see that flat and trend do not exist, especially as a prognosticating tool.
A normal discussion should be like this.
Hypothesis - trend by day, flat by night.
Check.
I do this and that by day (code example), andthis and that by night (code example).
The results are as follows - here is a table of forwards on non-optimized areas for comparison.
Conclusion - it's really better at night this way, not that way. (Variant. I was wrong, it is the same at night). Any ideas for solutions?
Then there will be a subject for discussion. For now it's all just lyrics. I do not see yet that flat and trend do not exist, especially as a prognostic tool.
If you have an Expert Advisor, you should put a time filter on it and see for yourself. You cannot be more specific than that, allow trading from Hstart to Hend. This is a very specific one, I don't know any other way, that's why I've created this topic.
But I have not seen any specifics from you so far, unfortunately.
A normal discussion should be like this.
....
And while I do not see that a flat and a trend exist at all, much less as a predictive tool
Everything will fall into place if we understand the terms trend and flat.
A trend is an alternating sequence of waves.
A trend up and down is tops to this or that.
A flat is a sequence of tops not fitting the term trend. Somewhere so short.
It is visually easier for us to see the flat stage in the lower TF from the higher one. I hope I explained it clearly on my fingers.