![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
The other day I decided to check how often Fibonacci levels occur in the market. I took EURUSD, built a zig-zag and calculated the percentage ratio of the height of each leg relative to the height of the previous leg:
100% * MathAbs((zz[i]-zz[i-1])/(zz[i-1]-zz[i-2]))
Plotted a histogram expecting peaks of 23.6%, 38.2%, 50%, 61.8%, 76.4%, 100%, 127.2% etc. I got this
Fibonacci in forex is a myth, guys. There are pronounced sharp peaks at 100% and 200% muves and a smoother peak at 84%. The histogram itself has the appearance of a Poisson distribution.
The expiriency is incorrect due to looking ahead when plotting the ZZ. Therefore, a spike of 100% is probably the value of or related to the SP threshold. And vice versa, corrections less than that threshold are ignored. Generally speaking, it is not the ratio of swings that matters in Fibo, but correction levels in swing coordinates.
Therefore for a correct expression it is necessary to
1. formalize the generated swing without looking ahead. we pull a fib from it: look at the levels of corrections developed in its coordinates
2. to determine the levels of existing corrections
For the 2nd one, the ZZ would be appropriate, but its threshold should be less than the length of the main swing. And you should not take the length of correction, but coordinate of the top relative to the swing in %. If the swing is down, then take ZZ's highs and translate them to the Fibo level relative to the swing, and vice versa for the swing up. And so on, until a new swing is formed
To identify the swing from which the fib is pulled, you can use confirmed extrema. A confirmed high is the maximum X bars before and Y bars after it, a confirmed low is the minimum. For simplicity X=Y. The swing will be from the confirmed high to the low, etc. as in WP. The swing is there too, but not by the value of future movements and correction levels do not depend on the swing parameters, though sometimes it will be calculated only after the fact (when the correction will be formed before the confirmation of one of the swing extrema)
The other day I decided to check how often Fibonacci levels occur in the market. I took EURUSD, built a zig-zag and calculated the percentage ratio of the height of each leg relative to the height of the previous leg:
100% * MathAbs((zz[i]-zz[i-1])/(zz[i-1]-zz[i-2]))
Plotted a histogram expecting peaks of 23.6%, 38.2%, 50%, 61.8%, 76.4%, 100%, 127.2% etc. I got this
Fibonacci in forex is a myth, guys. There are pronounced sharp peaks at 100% and 200% muves and a smoother peak at 84%. The histogram itself has the appearance of Poisson's Rapture.
The mistake of all such reasoning is to build a zigzag and try to find manifestations of fib. In this case you investigate properties of the ZIGZAG, not properties of fibs that appear in the market.
You may as well attach any other instrument, not a zigzag, to a chart and try to find the phibs manifestation on its highs and lows.
This is a striking example of sophistical logic. A subtle substitution of one concept for another... The result is proof that white is black and vice versa that black is white.
One must be aware and not substitute one phenomenon for another.
The zigzag is not a tool to detect the appearance of phibs in the market. Not if applied head-on. We construct a zigzag with parameters "from a light bulb" and try to draw far-reaching conclusions.
The results will be somewhat better if we try to adjust the zigzag to the market rhythm. But it is not an easy task. Not an elementary one.
Here's a little on the subject touched upon in this forum thread: http://www.onix-trade.net/forum/index.php?showtopic=30638&setlanguage=1&langid=ru
and here http://www.onix-trade.net/forum/index.php?showtopic=84054&setlanguage=1&langid=ru
But not everything is there. Some of the comments on the topic have been left in the Asics...
===========================================
If there is a need to identify fibs, then the research needs to be done correctly... consistent with the nature of the fib. A zigzag with parameters from a light bulb will reveal extrema belonging to different cycles, different waves... in general, random extrema. (The zigzag lives its life and the market lives its life.) And fibs are built from non-random extrema.
The topic is certainly sensitive. But thank you all for the answers, criticism and recommendations. I'll work on my program a bit longer. So far this is what I get with the corrected report of Fibonacci levels. My conclusion is still the same: these levels are in our heads and our eyes see them because we want to see them.
The price chart has two zigzags, a fast one (dark blue) and a slow one (light blue). Fibonacci levels report from the slow zigzag forward, into the future. Statistics are collected from the ends of the fast zigzag, for which levels are drawn.
There is a peak at 37%, which is close to 38.2%. But in general Fibonacci levels work as well as randomly chosen levels.
Of course these results depend on how the zigzags are built. And probably someone will argue that if you select a slow zigzag in a different way and look not at the ends of a fast zigzag, but at some prices in the middle, everything seems to work out.
It is possible that the levels are in our heads, but this does not diminish their potential. On the other hand their application is not confined to point-to-point determination. As usual you have to make allowances for quotes and specifics of interpretation. I am attaching an example. Of course, you can argue as much as you want, but it seems to me that you don't need a trained eye to see at least half of triggered levels. In real trading not all of them are applicable, at least at this stage, but still:
Read Kiyanitsa A.S., Bratukhin L.V. - "Fibonacci Levels: Where the Money Lies" - will bring you closer to understanding
To be honest, I haven't gotten into it myself, and I can't give an answer.
- I don't like cats.
- You just don't know how to cook them.
P.S. FOREXMASTER, is this a libation for newbie?
If Fibo levels worked in Forex and other markets, there would not be so many forums, unsatisfied traders, a lot of TSs and so on. Everyone would be rich and happy owners of the "grail" or the "printing press" ....
If most people didn't reason and wait for someone to chew it all up and put it in their mouth, but rolled up their sleeves and worked, more people would have a "printing press".
On the other hand, now, a man came to the Forex market. He just threw a few indicators on a chart. He does not think about it. And he wants that manna from heaven will appear to him.