Once again, it's about the eternal: trend/flat. - page 10
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
I didn't ask for a definition from the Wiki, I asked for your definition.
And I drew above what I want. Basically, now I have almost a calculation algorithm in my head.
I wanted to show that there are only abstract definitions and we need specific ones.
There are many ways of defining it. I suggested a universal, easy to code, maybe someone will find it useful.
P.S.
I should add that this way defines a flat at an early stage.
I would add that this method detects a flat at an early stage.
On ZZ at an early stage? I doubt it, and I doubt it fiercely.
I've spelt it out a little, but you still don't get it, so what can you do, I guess the approaches to understanding are different.
It is a pity that I have not mastered the creation of videos on YouTube, would show you what you doubt.
We have a price range, with specific price minima and maxima. To find them it is convenient to apply a zig-zag, for example.
We do not need anything else! We obtain the prices of the first candle (maximum/minimum) - calculate the distance between them. It satisfies - we look at the second candle. If its maximum/minimum breaks the first one's extrema - we renew the extrema. And so on, until we find the required figure of the required size (in terms of length/height). If the size is exceeded, we reset the current settings and wait for a new candle.
If a flat has already been formed - check for an exit outside the flat.
That's all we need! We got the price of the first candle (maximum/minimum) - calculate the distance between them. It is satisfying - we look at the second candlestick. If its maximum/minimum breaks through the first one - we renew the extrema. And so on, until we find the required figure of the required size (in terms of length/height). If the size is exceeded, we reset the current settings and wait for a new candle.
If a flat has already been formed - check for an exit outside the flat.
You can programmatically search for a narrow range, but that doesn't identify the so-called "box" when prices hit the upper and lower boundary several times.
I mean if you kind of "stupidly" search by absolute maximum and minimum in the history, for example, as in the picture. The price will fall within the range and the software will consider it a box, but it's not.
But, it needs to be:
I hope you understand now, that illusion with "simply" finding a narrow range.
One of the options for identifying exactly the range is applied here.https://www.mql5.com/ru/forum/65427
You can programmatically search for a narrow range, but it doesn't identify the so-called "box" when prices hit the upper and lower boundary several times.
I mean if you kind of "stupidly" search by absolute maximum and minimum in the history, for example, as in the picture. The price will fall within the range and the software will consider it a box, but it's not.
But, it needs to be:
I hope you understand now, that illusion with "simply" finding a narrow range.
One of the options for identifying exactly the range applied here.https://www.mql5.com/ru/forum/65427
Here, and this is the next part - the filters.
Although, in the above screenshot what you've described is artificially limited by the number of candlesticks (big enough) and low height (small enough). Both are picked up by eye.
Here, and this is the next part - the filters.
Although, in the screenshot I gave, what you describe is artificially limited by the number of candles (quite a large number) and the low height (quite small size). Both are picked up by eye.
After the "movement" the market forms the range it is interested in, at extremums (flat) there is a high percentage of false-breaks and a "box" chart built on the extremums
It's better to look at the internal content of the chart.
Looking for a flat at an early stage ? first 2 calculation points - then 3. i.e. we need history, and when there is history, the return of the price to the "box" can be regarded as a signal
There are trading systems where the signal is a stochastic "exit" from 80%, but here the price acts as an indicator and these levels are "masked" by the flat.
Working in a "box" is a separate topic )))