Is any prediction doomed? - page 7

 
OnGoing:
...

Question. Where's the way out?

The way out is to trade backwards!!! But how???


Seems to have poured well or what?

If not, then where do you get such a bland approach ...:-))

 
Roman.:


Seems to have poured well or what?

If not, then where did you get such a bland approach... :-)))

Nope, crystalline as a baby's kiss) Just thinking out loud, maybe someone will understand, if I can make it clear...
 
OnGoing:
Nope, crystalline as a baby's kiss) Just thinking out loud, maybe someone will understand if I can make it clear...


I see... :-)))

On the reverse trading principle, it's the spread that drains... :-)))

If the pure TS gives average results, then its inverse version will also gradually drain... - see here and from here (to the end of the thread), for comparison.

Each TC needs an individual approach and sharpening for certain tools with a better approach, IMHO.

 
OnGoing:
Nah, crystalline as a baby's kiss) Just a thought out loud, maybe someone will understand if I can make it clear...
trade without thinking.
 
OnGoing:
My hand on the mouse is stiff...

I don't like mice... Piping with a mouse isn't bad, I guess, but it's not efficient. Why is my hand shaking to close? Yes, because the processes of "open-close" are physically difficult! You've said a lot and you haven't understood it. To win back the spread at a loss is a piece of cake if you do not use the mouse. It takes a long time to open a position by sending a check through the "existence of a file" - is it long? And it is not long to close the window confirming closure of a position and open the window "open a new order"? The "standard" position opening is a perversion in my opinion. After making a simple external "opener" - it's like being born again! Have you ever played Tetris? Do you think it's possible to "hustle" with a mouse in Tetris? ;)

Sure. This opener is no panacea. But it does make life easier. At least it does not create a little trouble and allows you to focus solely on solving the "when" "close-open" issues. On the subject, I don't know. Whether or not humanity is doomed... life will judge. But the fact is that it is necessary not to predict, but to make calculations. If you have made a calculation, check it with the market. To paraphrase the topic, one might ask "is it worth trying at all?" Why not try? You can always die in time. But "living" is trying.

 
lasso:


Friends! Some of you are wrong...

Who?

Wallpaper.
 
avtomat:

I am talking about something else.

For example, it is sufficient to determine - i.e. predict - the direction of further development of the process in the near future. But you can set the task of predicting a particular price at a certain point in time, and it would be a completely different task. And the technique used in these approaches is different. And the results are different too.

I don't understand the difference, can you be more specific?
 
LeoV:


I agree.

Just what was I going to say? What I meant to say was that no one would think of inventing a new way of making a hammer, because in their opinion this way of making a hammer is not quite right. It's like reinventing the bicycle.

As for predicting the colour of the next candle, you can logically conclude that this idea was the first at the beginning of price charts in the form of candlesticks. And only the "lazy" didn't try it. But where are the results? Well, there are no results. Or rather, they may have been once - "Who goes first, who goes first" - but now, you are not the first, and the market has long outgrown all available and cherished ways of predicting the colour of the next candle. The market is smarter than everyone....))))


I'm actually pessimistic about price forecasting, but at the same time I don't deny that such forecasting is possible in short periods of time given a certain set of circumstances. I liked your phrase about forecasting being meaningless and looking for patterns... "Market pattern" is an oxymoron. A pattern implies a constantly recurring correlation of events. Nothing in the market is constant. Regularities emerge and turn into coincidences. That is why it is very difficult to detect them and use them for profitable trading. By the time a neuronet has learned to detect some regularity, it has already turned into a randomness.

By the way, I don't see anything wrong with trying to "reinvent the hammer". Einstein explained that he created the theory of relativity only because all men of science decided that it makes no sense to question Newton's classical mechanics, and he, Einstein, decided to question the validity of the "axiomatic theory".

 
I think the concepts of "prediction" and "pattern" need to be redefined each time to represent what we are talking about. If by "pattern" we mean the alternation of "white-black" candlesticks, that is one thing. If by "pattern" we mean the way of crossing (or parallel) lines is another. After some experience, the candlesticks as the first object for investigation become secondary. The lines and the way they are broken (or not broken) by the price come out in the foreground. A "channel" and a "parallel" is a simple case of the "price movement". Moreover, the calculation of the channel should be done on its FIRST point, not on the "third" point. "The third from one match does not light" - protection from a night sniper. From the second to the third point - you must take a profit (taking into account the insurance calculation at least to return to breakeven). Why third "does not light"? The first one notices. The second - takes aim. The third one shoots.
 
LeoV:

Generally speaking, forecasting in the financial markets is a useless business. Only the search for patterns.....

I would like to join in.


I'll try to explain for those who don't understand the difference. Using a simple example.


Suppose someone is standing at a crossroads and looking at a car that has approached, stopped at a traffic light and must move. Standing there, someone is looking at the car and deciding in their mind where the car will go. So, if you transfer the current argument to this simple life situation, those who talk about prognostication most likely try to calculate the exact trajectory of the car (to within centimeters, sometimes ;) just by the way it approaches the crossroads. Sometimes it's possible, even in life, but more often it's 50/50. Or for example, determining where a car will move by its colour is a good task :) But searching for patterns is something else. For instance, you look - turn right, and it's all the same how exactly you turn right, along what trajectory, or even along curbs. Of course, this example is exaggerated and, in principle, the knowledge of turn signals is given by traffic regulations, but just imagine a group of monkeys on the street. Obviously, a group that will guess to look at turn signals and not the colour of the car, they are not afraid of crossings. And the others will go for cutlets. It's the same in trading.