Fibonacci levels: myth or reality? - page 8
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
Again, when someone trading on a semi-automated machine claims that their system works, it is a lie because the system has no clear rules and is highly subjective. And therefore is not even a system.
I don't understand that last line.
What is THIS then?
Or does the presence of a fraction of a subjective deprive it of the title of a system ?
I don't understand that last line.
What is THIS then?
Or does the presence of a fraction of a subjecƟve deprive it of the title of a system ?
For such cases, the system title can only be retained if the trader is awarded the same title.
Let's then look at the hypothesis of Khrenovsky levels working: 20% and 40%, and smash it to smithereens. After all, these are not holy fibo levels, just Khrenovsky levels. How shall we break them? By the simple method of the author of this topic.
It's not about Fibo, but in general about how people take for granted everything they write in books. For some reason it does not occur to me to check it.
This simple method of the author of the topic does not prove or disprove anything, because it is too primitive. Similarly, one could prove that the spread on the oira is always 18 pips by choosing a time frame when it is so and ignoring the other. And I've already written about the difficulty of testing the Fibs hypothesis.
Rich Swannel did the same primitive thing in the appendix to the book about Elliott Principle. And he obtained similar smooth curves without special outliers on Fibs. But what an extensive statistic! And all wasted - because it's too primitive.
This simple method of the author of the topic does not prove or disprove anything as it is too primitive. In the same way, one could prove that the spread on the oire is always 18 pips by choosing a time frame when it is so and ignoring the other.
Dear Sir, there is no averaging in the method of the author of the thread. You just had the number of different levels counted on your fingers and shown in the form of a chart. You can make a few more graphs for different intervals, or in 3D for clarity. The result will be the same: all intervals will be the same.
You judge the method to be primitive only because it is simple. Testing the cluster fibo hypothesis requires a much more sophisticated method. That's the beauty of complex hypotheses, you can't overturn them in practical terms, because it takes a lot of time and effort. And so complex hypotheses will live forever in the minds of people like you.
3 years ago I was trading with MMK NK.
The broker sent me the analytics. It had fibs and trend levels + fundamentals :)
I missed my targets :)
I think it looks so stupid. We can yell about everything - it's not working! All is rubbish ! Blah-blah-blah and so on.
Phoebe, like any other tool, is just a tool. But how you use it is up to the owner.
P.s. you can screw around with ADX, but people will never stop using it and wybka.
3 years ago I was trading with MMK NK.
The broker sent me the analytics. It had fibs and trend levels + fundamentals :)
I missed my targets :)
I think it looks so stupid. We can yell about everything - it's not working! All is rubbish ! Blah-blah-blah and so on.
Phoebe, like any other tool, is just a tool. How you use it is up to you.
P.s. you can screw around with ADX, but people will still use it.
With up to a hundred or two trades, even flipping a coin does not rule out profitability. But it does not define a pattern in any way. Simple patterns have been used publicly by the above people, and the statistics of their systems speak for themselves.
Dear Sir, there is no averaging in the method of the author of this thread. You have simply counted the number of different levels on your fingers and shown them as a graph. You can make several more graphs for different intervals, or in 3D for clarity. The result will be the same: the same on all intervals.
You judge the method to be primitive only because it is simple. Verification of the cluster fibo-hypothesis requires a much more complex method. That's the beauty of complex hypotheses, you can't overturn them in practical terms, as it takes a lot of time and effort to do so. So complex hypotheses will live forever in the minds of people like you.
Did I say anything about averaging? What averaging as applied to the Fibs?!
I propose to end the religious argument. I have no arguments as I don't trade on Fibs and I don't even know how to validate the cluster hypothesis yet. Right now my interests are focused on other hypotheses.
When the number of deals reaches a hundred or two, even flipping a coin does not rule out profitability. But it does not determine a pattern. Simple regularities were used publicly by the above-mentioned people. The statistics of their systems speaks for itself.
The ZZ and the faba were looked at, found nothing and verified everything: "Fibka in a batter, fibka in batter, fibka in sauce, fibka.... "
that doesn't make any sense. You can look for these patterns either way.
I suggest we end the religious argument. I have no arguments as I don't trade on Fibs and I don't even know how to validate the cluster hypothesis yet. Right now my interests are focused on other hypotheses.
Definitely don't give a damn about the "argument"...
come on, don't hypnotize, finish Shrek 4 :)
I think I've made my evidence pretty clear. The probability charts at different levels were shown. I expected sharp peaks at 23.6, 38.2, 50, 61.8... or in their vicinity. No peaks, no spikes, no bumps to be seen. Only measurement noise superimposed on the smooth spreading curve. As a result, I concluded that Fibo levels do not differ from others in terms of probability. Even if I take +/-4% windows near these levels, the probability is still the same as outside these windows. The opposite question. Let's unwind Fibo levels 38.2 and 50 to 36.6...39.7 and 48...52 (-4%...+4%). If these levels were more frequent than 39.7...48, wouldn't that mean that the histogram would show a much lower probability at these non-fibo levels?
If we unwind the levels wider, all the values would become Fibo.
You didn't investigate the manifestation of Fibo in the market, you investigated the ZIGZAG. And that says it all. That's why I say you haven't proved anything with your research.
Can you distinguish a study of ZIGZAG properties, from a study of Fibo manifestation in the market? If you think they are one and the same, you are deeply mistaken.
Don't substitute one concept for the other.
===========================
Generally speaking - such arguments are unproductive. To change the minds of those who have successfully applied phoebes that phoebes are nothing. All the more so with such crude methods.
When applying graphical methods, a combination of different analysis tools is always used.
Fibs give reference points where to pay attention to what is happening on the market. In other words, they show a potential reversal point (to put it simply). And in this place we look at the dynamics of movements. At these levels in addition (yes, additionally) we can perform the analysis by candlestick (Steve Neeson). Candlestick analysis is also a graphical analysis.
When using fibs, it is important! to be able to find anchor points. Yes, these points are extrema. These points can also be called the fractals. The essence thereof does not change. This is not a random process - finding anchor points. It obeys the rules. Some of these rules have been announced in this thread.
It is not very easy to automate this process.
===========================
The public examples of successful automata trading given just above are not simple either. They all require optimization after a while. Read - tweaking.