What is a trend? - page 20

 
Andrew Petras:

Does it exist?

Andrey, do you mean the trend has to go through a flat phase before it turns? Look at the last three weeks, the USDJPY has been going through a flat phase:

 
Victor Ziborov:

You mean the trend has to go through a flat phase before it can reverse? Look at the last three weeks, the USDJPY has reversed past the flat phase:

Exactly. No so called phases are absolutely unnecessary. And every situation, a posteriori of course, has an explanation in the TA.
 
Serqey Nikitin:

Yes, about the picture: https://www.mql5.com/ru/forum/214374/page17#comment_5664189

I don't doubt that all this analysis works...

But for myself, I decided to design a system that would meet the parameters:

1. Easy to understand.

2. At any point in time there should be SIGNIFICANT clarity about the current situation in the market.

3. and, of course, profitable.

The analysis of the same pair:


The current state of the market - the trend is up.

Prospect - continuation of the upward trend.

Trade orders:

Opening - both indicators are in the same direction, but only after the end of the pullback.

Closing - a signal for reversal of the fastest indicator.

P.S. The Zig-Zag is only for setting the other trend indicators (not involved in the strategy).

In my opinion, the trend indicators give a simple and clear picture of the market.

Sergey, your trading system looks very convincing. These two indicators below, are they available to the public, or they are purely yours and for your own personal use? If they are publicly available, could you provide the link, please? I would like to try it.

 
Victor Ziborov:

These two indicators below - link please ? I would like to try them.

One of them seems to be here...

 

Thank you very much. I see it's paid for, but there's a Trial. The author is that Sergei Nikitin. If I like it very much, I'll write one myself. Or buy one, we'll see.

 
Ivan Butko:
Oooh...

What a lot of vague thoughts. You should get together with them and formulate formal criteria/rules/instructions whereby a strict point of trend start, its end and transition of directional trend phase into a flat is determined.

So far, the only formal clue is a binding to a scale (a wave level, if we speak for neo-wavers, and fractals, if we speak for conventional ones).

I.e., coming from the obvious one, namely - a flat is a trend on the mlTF, and a trend is a flat on the stTF (not always, except for impulses from some Brexits), in the sense that everyone can clearly see it (associative perception), but cannot formalize (algorithmize), we can determine whether a trend belongs to a strictly defined TF.

Then, the trend definition will include the TF:

A trend is a directed, non-horizontal movement of price in a certain TF, in which the reversal movement in the opposite direction does not exceed (tra-ta-ta-ta)... what next? Size? Does it not break the previous fractal? What do you think.
I agree, we know about the qualitative side of trend detection. We need to work out quantitative measures of trend recognition, like if the relative price increment for a certain period exceeds some value, we need to recognize the presence of a trend. Otherwise it is a flat. I ask participants to develop this idea and suggest their own variants of quantification, such as criteria. For me personally, a trend is the movement of the current price from one breakeven point (level) to the next breakeven point until the market price intervenes. A run-up and subsequent bump in the market price at the current price means the end of the trend. To many, this seems like nonsense, but, the concept of market price derives from real analysis of the market for goods and services and it is time to recognise it as a real participant in market battles. I am not imposing the concept of market price on anyone, but I am sure it cannot be done without recognising it and studying it comprehensively.
 

My view is that a line chart of a trading instrument is the main price trend.

Trends based on opening or closing prices are secondary trends, they provide more information for the analyst in the time section, but for the same reason, such secondary trends are more noisy in comparison to the linear chart of the trading instrument price. Maybe that is why Expert Advisors based on open/close price trends are losing money?

 
Yousufkhodja Sultonov:
I agree, we know about the qualitative side of trend detection. It is necessary to work out quantitative measures of trend detection, such as if the relative price increment for a certain period exceeds some value, we should recognize the presence of a trend. Otherwise it is a flat. I ask participants to develop this idea and suggest their own variants of quantification, such as criteria. For me personally, a trend is the movement of the current price from one break-even point (level) to the next break-even point until the market price intervenes. A run-up and subsequent bump in the market price at the current price means the end of the trend. To many, this seems like nonsense, but, the concept of market price derives from real market analysis of goods and services and it is time to recognise it as a real participant in market battles. I am not imposing the concept of market price on anyone, but I am sure it cannot be done without recognising it and studying it comprehensively.

A trend is like a consecutive increment in the next time interval. And the extent of the increment is probably not the point. The increment has stopped (equal to zero) - time X for trader and Expert Advisor, as the drawdown test is ahead)))

 
Uladzimir Izerski:

Not bad. But there could be a slope in the sideways trend. How do we put up with that?


If the slope angle is different from zero, it is a trend anyway. If BID draws a perfectly straight line within time, the trend is not present, while if there is even the slightest deviation, the trend is present. A trend is a direction of change along the Y-axis and the change along the X-axis is time. The trend is not considered to be IMHO. There is no trend if our line is parallel to the X axis

 
Ivan Butko:
Can you give a comprehensive definition?

How do you personally determine when it started and when it went flat (if it is not a trade secret D)

I'll give you a comprehensive definition: a trend is when two cars don't overlap :=))

Reason: