The Fibonacci and Golden ratio myth

[motorway]

http://www.constructal.org/en/art/DN 1067.pdf

Motorway

 

[Garpal Gumnut]

Ah yes , the son of Bonacci was a genius.

gg

 

[brty]

From the OP,

I can't take them seriously as a trading tool.

I do take them seriously, because others use them. I don't, the % of wins is way to small. However if one of my signals happens to be on some fib number/ ratio, then all the better (yet it doesn't change what I was going to do anyway )

brty

 

[SmellyTerror]

...the 61.8% relationship (plus or minus 5% - it'll hardly ever be exact) occurs regularly enough to warrant caution...

I admit I'm new to TA in general, but this is what gets me about Fibonacci. With such a range of uncertainty, how is it telling you anything at all? I mean, you've got Fibonacci numbers at 23.2%, 38.2%, 50%, 61.8%, and 76.8%. If you extend the range around each by about 5% each way, you've got 50% of the range covered, and discounting the fringe areas (non-trending?) from 0-18% and 82-100%, then your 5% allowance means those 5 Fibonaccis are covering 78% of the central range! Well no wonder stuff happens at those numbers rather a lot!

Eg: the 50% range could go up to 55%, and the 62.8% range comes down to 57.8%. So there's only a 2.8% gap for "non-Fibonacci" numbers to fall into.

Now I'm not saying that everyone uses a +/- 5% range - but not everyone is getting a 78% hit rate, either. For a lay person who has just started looking into this, it just seems to me that, if there's any edge here at all, it's pretty tiny compared with MM and unconscious experience.

That is, from where I am (in the Land of the Ignorant, looking down the Path of Learing Something I'm Not Sure I Want To Bother With), it looks to me like Fibonacci is more about getting the confidence to get into a trade, and that it's money management and the experience you have looking at charts that makes the success. There's such a lot of room to move with this stuff that I think it's just your "eye-balling" and all the unconscious calculation that goes along with it that's really putting you onto a trade. The Fibonacci stuff is just there to make you feel better about it.

Instead of saying "a 62.8% Fibonacci retracement" I could just say "comes back a fair bit", couldn't I? What you're talking about with a certain number is a rough degree that the price has moved, not a mystical rule of the universe. "If it goes down about that far, it'll often come back up".

That's what I think. I'll come back in a year and tell you how stupid I feel for having been so ignorant...

 

[Wysiwyg]

It's a bit like the dude who has 10 support levels over a 50 point range. Gotta get one right. Not having a go at trading approaches but there is wheat, there is chaff and there is stuff that requires some more imagination.

Two levels ... support and resistance. Simple.

 

[Timmy]

I mean, you've got Fibonacci numbers at 23.2%, 38.2%, 50%, 61.8%, and 76.8%. If you extend the range around each by about 5% each way, you've got 50% of the range covered, and discounting the fringe areas (non-trending?) from 0-18% and 82-100%, then your 5% allowance means those 5 Fibonaccis are covering 78% of the central range! Well no wonder stuff happens at those numbers rather a lot!

Eg: the 50% range could go up to 55%, and the 62.8% range comes down to 57.8%. So there's only a 2.8% gap for "non-Fibonacci" numbers to fall into.

Re your words I quoted above SmellyT - can't see any problems with them. If new traders use their intelligence in this way they will have a huge head start.

Fib numbers can be helpful to be aware of only in that, as other posters have already stated, others might be using them and their use of them may have some effect.

 

[wayneL]

I still reckon people have TA all wrong.

There is this notion that TA is used to predict future moves. I don't believe it should be used that way for a millisecond (although I do indulge in that sometimes for pure fun). I think TA should be used to set a trading hypothesis.

Radge uses the terminology of "creating boundaries", which I like and would like to steal from him and using these boundaries as parameters of proof and disproof of that hypothesis. We should know this, it was repeated, parrot-fashion, ad nauseum here by a certain ex-member.

Despite it being being preached to the point of making us all puke, it is essentially true.

Does hitting a fib level predict a pivot point and reversion to the dominant trend? No way! That's toss of the coin territory or even worse.

But a fib level can be used as a boundary. eg "If the price drops below the x% level, my hypothesis is incorrect and I will take such action as determined beforehand should such boundary be breeched."

You can create a hypothesis that says the trend I want to ride does not retrace past the x% level. While above that, I'm in... below that I'm out. (Or something like that)

FWIW

 

[Timmy]

But a fib level can be used as a boundary. eg "If the price drops below the x% level, my hypothesis is incorrect and I will take such action as determined beforehand should such boundary be breeched."

You can create a hypothesis that says the trend I want to ride does not retrace past the x% level. While above that, I'm in... below that I'm out. (Or something like that)

FWIW

I like this idea, but ....

Using a Fib number is essentially using a random number - the hypothesis is therefore based on "if the price falls below some random number" I am incorrect. (Wayne, I have simplified, apologies, but hopefully the gist is clear). It is the 'no better than random' aspect of Fib that makes them useless for forming hypotheses, in my book.

(ps. There may be thinly-traded markets where the Fib numbers can have an impact - i.e. pays to be aware of them because others trading in the thin market may be using/aware of them. In thick markets the impact of Fib-using traders will be negligible).

(pps. I find the main use of TA is in assessing responses. Not an original idea, thanks mainly to Motorway)

 

[wayneL]

I like this idea, but ....

Using a Fib number is essentially using a random number -

Yes, agree. But many of what we traders use is essentially random numbers.

Moving averages, although they may be used with an overriding logic, are essential random. An arbitrary mathematical construct using past data, just like fib nos. or chandelier stops or bollie bands.

Let's say we use the 100day moving average as a trailing stop. If it touches, we exit. There may be reasons for selecting that MA, but it is still essentially a random number we have selected using some elaborate justification.

But it is a defined boundary and precipitates a decision based on the hypothesis we are using.

Fib numbers can have an extra layer of nebulousness do to the arbitrary selection of swing highs/lows.

Fibonacci is great for impressing people at parties and trading noooobs willing to part with $6k for a course, but otherwise totally agree.

 

[Frank D]

Randomness comes from the association of trading rules that are
purely directed to the number in question or pattern in question.

That’s the only way you can validate randomness, the association
of generic rules, 'the number', and then the end result: - Profit or loss.

A pivot point or a Fib number is useless unless it’s proven to work using
a generic set of rules.

Is the fib number the ‘boundary’, or the rules associated with the fib
number the boundary?

How many times does a level fail only to see it swing back late in the day
or the next day and move in the original direction it was intended?

You don’t know the answer because you haven’t formulated a generic set
of rules to test it.

Labeling something just for the sake of labeling something, is typical
what you read in 99% of trading literature. All looks good in hindsight.

 

[brty]

Wayne,

Does hitting a fib level predict a pivot point and reversion to the dominant trend? No way!

Agree 100%, yet how the price action hits a level can be far more important.

"If the price drops below the x% level, my hypothesis is incorrect and I will take such action as determined beforehand should such boundary be breeched."

I have a belief, based on a lot of research, that the bit of a spike through some of these common numbers/ easily identifiable levels is deliberate by some larger players to set off stops.

Frank,

How many times does a level fail only to see it swing back late in the day or the next day and move in the original direction it was intended?

That is my point above.

Randomness comes from......

Let's not go into randomness, or my belief of lack there-of, we will get wayyy off topic.

brty

 

[Timmy]

I have a belief, based on a lot of research, that the bit of a spike through some of these common numbers/ easily identifiable levels is deliberate by some larger players to set off stops.

A useful application of Fib numbers, brty.

 

[motorway]

Remember it is about RATIOS not so much the NUMBERS or THE STEPS

change the STEPS
eg
1 2 3 5 8 13

5 10 15 25 40

and you are defining different rates of COMPOUND INTEREST


http://www.pricetime.net/F_INTRO_expcurves_2004.html

those steps can be any interval

Nature isn't trying to use the Fibonacci numbers: they are appearing as a by-product of a deeper physical process. That is why the spirals are imperfect.

The race to derive Euclid’s φ value from principle is justified, but misdirected. It is justified
because the proportions that resemble φ occur around us in very large numbers. This means that the
emergence of designs with φ-like proportions is a natural phenomenon. A natural phenomenon
obeys the laws of nature, i.e. the laws of physics. Faced with an unexplained phenomenon, the
scientist
strives to explain the phenomenon based on known principles.
The race ‘to discover φ’ is misdirected because the physics phenomenon is not φ itself. No one has
found and measured φ on an object in nature (φ is not like π, which is measurable by dividing the
measured circumference by the measured diameter). The physics phenomenon is the emergence of
shapes that resemble φ.

OK ___>

"For a finite-size (flow) system to persist in time (to live), its configuration must evolve such that it provides easier access to the imposed currents that flow through it."

A finite flow could be a trend in a STOCK PRICE. For that TREND to

persist in time (to live), its configuration must evolve such that it provides easier access to the imposed currents that flow through it.

easier access is about PARTICIPATION..

a Fibonacci series is an exponential growth function

TRENDS are a growth function

Now not too HOT not too COLD
BUT JUST RIGHT--> = GOLDEN

It is about PROBABILITIES unfolding in a way that attracts PARTICIPATION

Not about MAGIC NUMBERS




Motorway

 

[Garpal Gumnut]

The son of Bonacci's numbers are indeed ratios, which apply in nature as we know it.

This is the big difference between fib ratios and other thought numbers.

Fib numbers can be accurately and simply measured objectively.

A good link for all.

http://goldennumber.net/hand.htm

Your hand shows Phi and the Fibonacci Series

There are many examples of the Divine Proportion found throughout the design of the universe and everything in it, but let's take a look at one of the most important things first: You! We'll need a few Golden Section / Fibonacci building blocks:

Successive Golden Sections of a Line

Each line is 1.61804... times longer than the one before it. (Conversely, a section drawn at 0.61804 (or 61.8%) of each line equals the length of the one before it.)
Let's start with something simple. Take your hand off your keyboard or mouse and look at the proportions of your index finger.

Your finger demonstrates the Fibonacci series of 2, 3, 5, 8

Hold your hand up to the screen. Don't be shy!
Each section of your index finger, from the tip to the base of the wrist, is larger than the preceding one by about the Fibonacci ratio of 1.618, also fitting the Fibonacci numbers 2, 3, 5 and 8.

gg

 

[SmellyTerror]

Not accurate or objective.

Take the finger example right there. Natural variation between people already blows the exact proportion out of the water. I just measured my index finger digits: 2.2cm, 2.4cm, 3.8cm. So that's a ratio of 1.09 and 1.58. I'm afraid the Smelly Terrible ratio is going to be different to the fibonacci one. I wonder if I can trade it? (*)

And I'm not freakish-lookin'. My fingers aren't all mishappen and deformed. I suggest the folk at home measure, too, and see what they come up with.

Regarding further "fibonacci in nature" stuff, please see here: http://www.lhup.edu/~dsimanek/pseudo/fibonacc.htm

So it's not in nature at all, really. No more than any other arbitrary ratio you care to name.

(*) Thing is, I probably could. As I said, IMO it's money management (exit rules and position sizing) that makes money, and entries informed by a lot of processing going on in the back of the head that no-one can teach. Most people's entry systems are just for scraps and comfort as far as I can see. Better than random, maybe, but not by a lot.

Dumbo feathers.

 

[Garpal Gumnut]

Not accurate or objective.

Take the finger example right there. Natural variation between people already blows the exact proportion out of the water. I just measured my index finger digits: 2.2cm, 2.4cm, 3.8cm. So that's a ratio of 1.09 and 1.58. I'm afraid the Smelly Terrible ratio is going to be different to the fibonacci one. I wonder if I can trade it? (*)

And I'm not freakish-lookin'. My fingers aren't all mishappen and deformed. I suggest the folk at home measure, too, and see what they come up with.

Regarding further "fibonacci in nature" stuff, please see here: http://www.lhup.edu/~dsimanek/pseudo/fibonacc.htm

So it's not in nature at all, really. No more than any other arbitrary ratio you care to name.

(*) Thing is, I probably could. As I said, IMO it's money management (exit rules and position sizing) that makes money, and entries informed by a lot of processing going on in the back of the head that no-one can teach. Most people's entry systems are just for scraps and comfort as far as I can see. Better than random, maybe, but not by a lot.

Dumbo feathers.

I doubt if you have measured properly.

Get an xray done and measure it again.

If not I'd get dna testing , just in case, or a check up.

gg

 

[SmellyTerror]

1. First, obviously, even if the finger thing above was right, it's still wrong. See, from the proportions you're giving: 3/2 = 1.5. 5/3 = 1.66. 8/5 = 1.6.

None of them are the golden ratio, and they're ALL DIFFERENT.

2. Second, honestly, people are different. We're not all built to the same spec. Here, have a squiz: http://images.google.com.au/images?hl=en&q=hand x-ray&um=1&ie=UTF-8&sa=N&tab=wi&safe=active

Get out your ruler, bring up some images, and see if the 2, 3, 5, 8 ratio holds.

It doesn't.

Here's a nice one from the first page of results (hope this isn't too big, here's the link if I need to delete it: http://www.newswise.com/images/uploads/2008/03/28/fullsize/hand_xray.jpg):



By my ruler and screeen resolution, the index finger digit proprtions are: 1.6 : 2.2 : 3.7 : 6.9. Which equates to proportions:
1.375
1.687
1.865

See any golden ratios in there?

If you think you hold a mystical key to the universe, you're probably wrong. It's not even hard to check, if you could be bothered. :

 

[Timmy]

SmellyT, something I just noticed from that X-ray.

Each finger appears to have 4 bones.
Yet the thumb, the essential, opposable, thumb, has 3
... a Fibonacci number.
Spooky.

 

[Timmy]

And in The Simpsons,
each character has four fingers.

But those with 2 hands (2, get it, Fib again) the total is 8 ...
and there is Fibonacci again.

It is time for my lie down now, so will have to go. :

 

[Wysiwyg]

SmellyT, something I just noticed from that X-ray.

Each finger appears to have 4 bones.
Yet the thumb, the essential, opposable, thumb, has 3
... a Fibonacci number.
Spooky.

And in The Simpsons,
each character has four fingers.

But those with 2 hands (2, get it, Fib again) the total is 8 ...
and there is Fibonacci again.

It is time for my lie down now, so will have to go. :

Aye, there be wisdom.