Best Thread Debunking Fibonacci's Code...

[snapper]

Fibs are as random as anything else in technical analysis.

Anyone who comes back and refutes this should do so by way of a forward example which we can see develop in real time rather than the more typical approach which is to 'prove' it by reference to a retrospective example where it worked perfectly aka cherry pick.

I agree - & the E-wave technique has always seemed to me to be too subjective. Trade what is, not what you think it's going to be! After all, nothing is predictable to that extent. I know it's a probabalistic theory but I like to establish changes of direction before entering rather than hope I'm at a significant Fib level that will decide that for me.

 

[PieterSteidelmayer]

I agree - & the E-wave technique has always seemed to me to be too subjective. Trade what is, not what you think it's going to be! After all, nothing is predictable to that extent. I know it's a probabalistic theory but I like to establish changes of direction before entering rather than hope I'm at a significant Fib level that will decide that for me.

I've got no problem with levels for entry/exit/target/stop, but they need to be based on something a little more empirical than the distribution of petals on a flower.

 

[andrewdb]

I've got no problem with levels for entry/exit/target/stop, but they need to be based on something a little more empirical than the distribution of petals on a flower.

If it works, it could be the distribution of empty coke cans at the local recycling centre - if it works.
My definition of 'works' would be something like to call the 3 highs and lows of the ftse100 over the next 3 months (april-june '14) in order within 1%.
Others may have a different definition.

 

[PieterSteidelmayer]

If it works, it could be the distribution of empty coke cans at the local recycling centre - if it works.
My definition of 'works' would be something like to call the 3 highs and lows of the ftse100 over the next 3 months (april-june '14) in order within 1%.
Others may have a different definition.

Precisely. I'd even go with the distribution of petals on a flower if it provided results better than random chance. So much mumbo-jumbo is borrowed from quite disparate fields and exalted by some for the very reason it is totally unrelated to trading. The more esoteric and occult the better - that way, fewer can see any immediate significance in it and the greater the opportunity for those who pretend they do, to hold others in their thrall.

There is an oft quoted argument that Fibs and such like are self-serving in that if enough people believe them it will be self perpetuating, but the fact is, show a chart to any two Fib-fans and they'll be drawing lines all over the place and are unlikely to have come up with the same results at all. Quite random.

If you take trading seriously, like a professional, you'll look for the simple, the obvious, the mundane, and use your commonsense. It's an apparent rarity in the retail world.

 

[adatherton]

404 link

I get a 404 on this link

Saw this research paper commented on in my newspaper's business pages today, which used it to stick the boot into TA of all descriptions. Basically concludes that Fibonnaci type retracements occur no more frequently than would be expected by chance.

(by the way, the thread title comes from the University's press release announcing the publication of the research)

http://www.cass.city.ac.uk/media/stories/resources/Magic_Numbers_in_the_Dow.pdf

 

[andrewdb]

In the spirit of being helpful ...

A link that works is http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.135.1144&rep=rep1&type=pdf (PDF).

You could also look at the perils of overfitting

http://www.ams.org/notices/201405/rnoti-p458.pdf (PDF)

Commonsense tells you that the simple mechanical application of fibonacci ratios cannot not work reliably - otherwise we would all be doing it.

Commonsense also tells you that it is so easy to overfit.

This leaves 2 likely possibilities for those who claim success

A- They are just lucky
B- They are doing something else as well that has not been captured by a mechanical trading technique.

I think there is some (B) in there somewhere - but have not found it.

I really do think there is worth in looking into simple 'reversion to mean' rules,
_but_ as an alphaville post points out in a slightly different context, what mean do you mean and when is it measured...

http://ftalphaville.ft.com/2014/04/17/1829302/bonds-for-the-long-run/

 

[andrewdb]

My reversion to mean picture

... and here is my 'reversion to mean' piccy which serves reasonably well

https://drive.google.com/file/d/0B2Ks53fVmykARDJpMHBQenZVX3c/edit?usp=sharing (Google Image)

 

[winner4x]

I for one use Fibs to find retracement but i dont trade all 50% or 61.8% retracements. I also use SNR/SNR to find significant areas to buy/sell, and also a little bit of EW knowledge. For me fibs is like the candlestick. A shooting star or a hammer doesnt always mean a reversal. U must have 2-3 confirmation that the market is going up/down