Finding local highs and lows in price - page 5

 
sever29 >>:

Вероно. Вы тогда заработали?

Я ловко рисую на истории пробои трендовых на 1,2,3.


I 'jumped' right after testing the neckline from the bottom. Instead of the test, the price knocked out the short stop loss and went upwards.

....

But after the two-day close I was already sure that it was a false positive. I did not act classically though (as shown in the picture).

Instead I went long twice from the support and closed after the return of the "first strike" + small profit. Although I could have set the TP on the classical level. (I did not regret it too much: the main thing is that I obtained the profit.)

I agree that "everybody is smart in hindsight".:)))))

 
ULAD писал(а) >>

Only with years of strenuous observation of price charts comes an understanding of the market and its nature.

agree

 
ULAD писал(а) >>

Have we crossed paths in poker, by any chance?

I don't think so, I only play backgammon for fun.

 
coaster писал(а) >>

I 'jumped' right after the neck test from the bottom. Instead of a test, the price hit a stop and went upwards.

....

But after the two-day close, I was already sure that it was a false positive. Although, I did not act classically (as shown in the picture).

Instead I went long twice from the support and closed after the return of the "first strike" + small profit. Although I could have set the TP on the classical level. (I did not regret it too much: the main thing is that I obtained the profit.)

I agree that "everybody is smart in hindsight".:)))))

:)

Good luck to you

 
sever29 >>:

:)

Удачи Вам


And more profits for you too! ;)
 
Mathemat:

Richie, and PapaYozh is correct. The best way to formulate a definition of a local extremum is to discard the right side of the chart and work with the left side only. So try to formulate such a definition that you will use in these conditions. Don't talk about guarantees. They are on the right, and you don't have them.


It occurred to me that the task of finding local extrema with practical targets in the "backtest" is limited "from below" by the spread and then also "from below" by the ability of the signal to "detect" that extremum. Why the hell do we need, for example, tick intra-minute extrema or minute extrema of the H1 TF?

Then the choice of a local range is limited by SL+TP because for some values of SL and TP orders will be triggered at these extrema but not at others.


 
Mathemat:

Richie, а PapaYozh правильно говорит. Самый лучший вариант формирования определения локального экстремума - отбросить правую часть графика и работать только с левой. Вот и попробуй сформулировать такое определение, которым ты будешь пользоваться в этих условиях. Только не надо о гарантиях. Они справа, а у тебя их нет

.


For me the thread was interesting, but IMHO a clarification of the problem statement is needed.

Take ZZ. We have two tops: #0 followed by the dangling end of ZZ, and #1, which has formed. We shall argue that turning ZZ #1 has 100% probability of being an extremum. We will consider the probability of the extremum at the turning point 0 at the moment when ZZZ was drawn to be 0% - we know for certain that it cannot be traded. As we move away from extremum #0 the probability of that extremum increasing and at the moment of a new reversal ZZ equals 100%. Intuitively, it is clear to me that 100% for pivot ZZ #0 is reached before a new pivot ZZ appears, i.e. we can take anything other than a new pivot ZZ(-1) as a confirmation of the pivot. Question:

1) what to take for a confirmation of pivot ZZ (just not a new pivot ZZ)?

2) what is the probability of confirming a ZZ pivot?

 
faa1947:


To me it is intuitive that 100% for a ZZ #0 pivot is achieved before a new ZZ pivot appears, i.e. something other than a new, #(-1) ZZ pivot can be taken as confirmation of the pivot.

Hmm, there is a condition for a ZZ reversal and its reversal occurs if and only if this condition is met.

To give a 100% guarantee of ZZ reversing before this condition occurs is to give a 100% prediction that it will occur after some time. Is that what you meant by this?

 
Candid:

Hmm, there is a condition for a ZZ to reverse and it will reverse if and only if that condition is met.

To give a 100% guarantee that ZZ will turn before that condition is fulfilled is to give a 100% prediction that it will be fulfilled after some time. Is that what you meant by that?


ZZ refers to redrawing indicators, i.e., pivot #0 may or may not remain marked by ZZ. What is the probability that it may or may not remain, which is the same thing? By the way, this is a problem with all redrawing indicators.
 

coaster:

I agree that "in hindsight everyone is smart".:)))))

ULAD:


Only with years of hard observation of price charts comes an understanding of the market and its nature.

To summarise:

Only with years of intense market observation does one come to realise that in hindsight, one can always find the "obvious" reasons why a pattern has worked exactly that way, and it has worked - TUTTLE TO TUTTLE!