Practicality of Fibonacci Retracements

[LondonUSTrader]

Those that say fibs are great are right and those that say they are rubbish are also right.

My personal experience with fibs are as follows:

(1) At times they can be incredibly accurate

(2) They are one of the most useful technical tools available

(3) Like drawing trendlines there is a lot subjectivity involved as the trader needs to decide the significant levels to draw them from

(4) Any technical tool is merely an aid to understanding market conditions. Tested mechnically fibs would probably have no edge. As an aid to discertionary analysis they can be very valuable. However, you would never want to make any trading decision solely on the basis of any technical tool as market context is so critical.

So, those that argue that they don't work are right as are those that argue they do work.

 

[murray t turtle]

Quote from jem:

many people say support resistance numbers don't work. Yet the fed says they do "work" in the currency markets.

Before you can say they do not work you would have to do at least as through investigation as the fed did.

===============
Yes ,dont fight fed;
but you may ignore silly fed remarks like ''irrational exuberance',
especially when market ignores comments like that.

Combo maybe help,s& r like fed said;
HOV is still below 50% level, 50dma, below 50 day lo;

easily to find longs above most 50 day measures

 

[taowave]

Quote from rcanfiel:

Major studies have been done with Fib levels. They provided no value add when compared with random levels, such as 50% or 25%. I don't give a whoopee what anyone believes.

The reality is they don't work for anyone, your "knowing of several successful traders aside." Appealing to "several who did so" is called anecdotal, and has no statistical value. They could pick other near levels, and experience little change in their LONGterm success.

You bring up some very good points,but just because "they provided no value add when compared with random levels,such as 25% or 50%,does not mean they arent effective.And I am in the camp that the Golden ratio is complete BS...

I would say that Fib levels in the hands of a successful trader works due to the fact that the trader has a stastistically valid trading approach,not because of the magical Fib level.As a simple example,I was always under the impression that I should look for stocks where Stochastics were <25 and crossed to the upside.Being a software junkie,I was demoing a program called StratSearch and ran an AutoSearch with Stochastics.For the subset of stocks I trade ,I found the most succesful use of stochastics was to purchase stocks where the stochastic had crossed BELOW 70.In a nutshell,buy stocks that were toward the high of their selected range and were making pullbacks or consolidating.Not shocking,but an interesting use of Stochastics.


With that said,there is a fairly strong chance that if one purchase stocks that are near highs and making a pullback,with a sound trading plan,you will do well.For those who look thru Fibonacci colored glasses,they indeed can buy that particular setup anywhere in between a 38 and 68% retracement.In fact,more often than not,if one applies ATR stops,the differential between the 38% retracement and the 68% retracement will be less than the stop level,so one could enter long at any given Fib retracement between the levels stated and not be stopped out.A virtual fibonacci shmorgasboard setup!!

When I buy pullbacks,i have no idea what the given Fib level is,but as we all know,there are enough Fib retracements out there that I am buying one of them.A Fibber would look at my trades and pronounce that a validation of Fibbonacci,and maybe they are right.To me,its no more than %D crossing below 70..

 

[squall]

Quote from gnome:

Reminds me of the Howard Borden classic comment... when asked, "How DID you ever get to be a navigator, Howard?", and he replied, "Oh no you don't, you'll have to go flight school just like I did"....

I remind you of a famous Mr. Spock quote... "When all other possibilities have been eliminated, whatever remains (however implausible it seems) MUST be the truth"... trading is like that.

Oh I definitely agree....it's common sense. I use two in particular that are almost uncanny.....

I know that talk is cheap however, so I can't really say I "was just curious" which ones worked for you woithout sounding like I am full of it. ha ha

 

[rcanfiel]

Quote from taowave:

You bring up some very good points,but just because "they provided no value add when compared with random levels,such as 25% or 50%,does not mean they arent effective.And I am in the camp that the Golden ratio is complete BS...

I would say that Fib levels in the hands of a successful trader works due to the fact that the trader has a stastistically valid trading approach,not because of the magical Fib level.As a simple example,I was always under the impression that I should look for stocks where Stochastics were <25 and crossed to the upside.Being a software junkie,I was demoing a program called StratSearch and ran an AutoSearch with Stochastics.For the subset of stocks I trade ,I found the most succesful use of stochastics was to purchase stocks where the stochastic had crossed BELOW 70.In a nutshell,buy stocks that were toward the high of their selected range and were making pullbacks or consolidating.Not shocking,but an interesting use of Stochastics.


With that said,there is a fairly strong chance that if one purchase stocks that are near highs and making a pullback,with a sound trading plan,you will do well.For those who look thru Fibonacci colored glasses,they indeed can buy that particular setup anywhere in between a 38 and 68% retracement.In fact,more often than not,if one applies ATR stops,the differential between the 38% retracement and the 68% retracement will be less than the stop level,so one could enter long at any given Fib retracement between the levels stated and not be stopped out.A virtual fibonacci shmorgasboard setup!!

When I buy pullbacks,i have no idea what the given Fib level is,but as we all know,there are enough Fib retracements out there that I am buying one of them.A Fibber would look at my trades and pronounce that a validation of Fibbonacci,and maybe they are right.To me,its no more than %D crossing below 70..

u speak wisely. fib people are anecdotal. when they see one that works, they post. But are silent when they dont

 

[rcanfiel]

Quote from gnome:

For those who claim, "Fibs don't work"... I suggest you are trying to apply them too broadly.

There are some chart "situations"* where Fibs work very well. If you restrict your usage to those, you'll be very pleased.

* I COULD tell what some of those are, but I'm not. If you put on your thinking cap, you can find some..

"I suggest you are trying to apply them too broadly."

Yes. Apply them to anything at a .382 or .618 level, plus or minus 0.618 ...

 

[squall]

Quote from rcanfiel:

u speak wisely. fib people are anecdotal. when they see one that works, they post. But are silent when they dont

LOL you are absolutely right...however my stop takes care of that for me. Good discussion- as even though I use fibs, I don't "believe" in them the way that alot of people do.

 

[rcanfiel]

Quote from squall:

LOL you are absolutely right...however my stop takes care of that for me. Good discussion- as even though I use fibs, I don't "believe" in them the way that alot of people do.

This is one example study, by Cass Business School in London. that looked at the market from 1914 to 2002 at Fib levels (they also studied round numbers). That is 86 years of market performance. This is only one study among others:

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

Their conclusion:

"Our conclusion must be that there is no significant difference between the frequencies with which price and time ratios occur in cycles in the Dow Jones Industrial Average, and frequencies which we would expect to occur at random in such a time series. In our introduction, we noted that empirical evidence from academic studies suggests that not all of technical analysis can be dismissed prima facie. The evidence from this paper suggests that the idea that round fractions and Fibonacci ratios occur in the Dow can be dismissed." [/B][/QUOTE]

I will cross post this in the other fib discussions going on.

 

[gnome]

Quote from rcanfiel:

"I suggest you are trying to apply them too broadly."

Yes. Apply them to anything at a .382 or .618 level, plus or minus 0.618 ...

What DO you use to trade*? Your understanding of TA appears limited.

*rhetorical question.

 

[rcanfiel]

Quote from gnome:

What DO you use to trade*? Your understanding of TA appears limited.

*rhetorical question.

Don't get started with TA that is nearly as useless as Fib levels, but then again, the fact that most traders use TA and most traders lose money seems lost on the believers in it. (And yes, there are plenty of studies showing the near valueless of TA). There are a couple of books bouncing around Amazon.com to this effect. One tested over 60,000 TA indicator/interval combinations over a long time, and the results were bleak.

There have been plenty of posters who keep saying, "study price action, not the mass of indicators." These people are farther along than the TA aficionados.