Good! Let it be non-random, I have renamed the branch. I should also remind you that, unlike Alexander, I work with minute charts.
Will it be continuous? The general notion of continuity: small changes in the argument correspond to small changes in the function. And in your case the changes of the function cannot be smaller than a point...
p.s. Dear local physicists, mathematicians and so on! Since I am not a physicist or mathematician, there is no need to push your intelligence. Better yet, give links to your scientific articles, publications where they have been published.
The idea is that the channel should not be built from price increments or anything else. More precisely, a channel with an average is already there by itself, it just needs to be brought to a certain value.
On the other pairs, however, the picture is not as pretty. It's on the channels from the picture above and experimenting with a slightly different approach.
So far it's all about statistical rubbish in the style of Bryusov:
Maybe these candlelit bars
Worlds with five continents,
The arts, knowledge, wars, thrones
And the memory of forty centuries!
So far, it's all just circa-statistical Bruce-esque rubbish:
What's it to you?
Good! Let it be non-random, I have renamed the branch. I should also remind you that, unlike Alexander, I work with minute charts.
Only here I have taken a different input data.
p.s. Dear local physicists, mathematicians and so on! Since I am not a physicist or mathematician, there is no need to push your intelligence. Better yet, give links to your scientific articles, publications where they have been published.
Price after returning inside the channel:

The method here is a little different. Not all the bars in a row are taken for the calculations.On the other pairs, however, the picture is not as pretty. It's on the channels from the picture above and experimenting with a slightly different approach.
In general, I don't think Alexander can be helped by increasing the observation window, reading ticks after a certain time and so on.
I think it is necessary to work not with prices, but with probabilities of events (not a fact, but maybe time plays some role).
Let local mathematicians and physicists correct me if this is wrong.