Fibonacci Numbers, is it The Holy Grail or Holy Smoke Why Did I Believe That?
I haven’t written this article to prove or disprove the use of Fibonacci. Rather I have pointed out some of the curious aspects of this approach to market analysis.
However, I have to admit that after a decade of looking at technical analysis, I am beginning to feel a little nausea when I hear other TA guys and girls talk about the Fibonacci retracements.
At the same time I know that I am going against what some of my best friends and most respected mentors taught me years back. It feels like knocking at the very foundation of your skill set. I spent 10 years going on TV, making many amazing calls using Fibonacci analysis.
Still, I need to put this on paper, at least just so my view is known. I know that I at some point in the future will have to use Fibonacci again.
It all started 6 years ago. I was a devout technical analysis man, and I had invited Larry Williams over for a speech in London. He showed a slide which essentially poured scorn on Fibonacci ratios.
I dismissed him as a stupid American who didn’t know any better; very much like a born again Christian will ignore any scientific evidence for the true source of planet earth. However, I had a nagging feeling and it grew.
When you enter the trading arena, you find yourself looking for a constant. A constant is a rule that you know will support you in times of uncertainty. It should never let you down. Your action according to the rule facilitates a known reaction. Fibonacci like religion does that to you. It soothes your soul, even when it doesn’t fit circumstances.
For example, when you learn to drive a car (sorry America – but in Europe we actually have to learn to operate a manual gear shift before they let us loose on the public roads on our own), you quickly understand that successful driving requires a careful blend of footwork between the accelerator, the clutch and the break.
As you acquire the skills of the 3 functions, you appreciate it is a function of administrating the three factors in the proper sequence and in the right proportion. However, you were never in doubt what each function stood for. There never was a time when the break pedal turned into an accelerator. Nor would the clutch all of a sudden act as a break pedal. There was certainty.
The astute reader will at this point add that another function is crucial to successful driving, namely the steering wheel. If so, you share my vicarious appetite for cynicism and you should find this article appealing.
Now we enter the trading arena, and we begin to look for the same kind of certainties that we look for in our everyday life. On our path to trading wisdom and the search for certainty it is very difficult to avoid Fibonacci.
Fibonacci stands for a slightly mystical segmentation of the markets. It is a method which slices and dices the market into known quantities. It is said to foretell the end of bull and bear markets. It is hailed as the greatest forecaster by many. I am told that our very existence, our very sphere, our life, our solar system, follows the laws of Fibonacci (a big sorry goes out to the born-agains).
Pay close attention here. It is not even a rule. It is a law!!
This thing is so big that it is not labelled a mere rule or an axiom. It is a law, on par with gravity and the 24-hour day, i.e. simply beyond question or dispute.
Finally we have a constant. A beacon of light in the dark landscape of indicators, oscillators, incubators, neural networks, volume, time of day, day of the week, seasonality and the millions of tools designed to further our understanding of the markets and enrich our bank accounts.
We pick up a book on Fibonacci and we see one convincing example after the next. Our gurus and mentors rave about it. They also show us one appealing example after another. It appears justified that this has been given the seal of approval and labelled a law. It is hard to dismiss, easy to use, and we begin to experiment.
So did I.
I used Fibonacci extensively in my trading. I loved it. It was so great to know exactly where the market would stop. Or should I say, where it was supposed to stop.
It didn’t always stop where I wanted to, but like any new love relationship you focus only on the positive, and turn a blind eye to all the quirky sides. My friend and highly accomplished trader Dr David Paul calls it confirmation bias, the tendency for man and woman to seek out evidence which supports one’s own acquired belief system.
My love affair didn’t last long. I began to doubt my love. It wasn’t quite what it was hyped up to be. I thought it was my constant in an uncertain world. I would put up with the frequent hit-and-miss. I would tell myself that it was my lack of understanding of the laws of Fibonacci that caused my poor results. I would study more, study harder. I would pledge my faith and accept that it was me who was the villain, the one of little faith.
Then came a point where I thought it was pointless. It gave me as many bad signals as it gave me good signals. Did I not see what everyone including Fibonacci could see? What was I missing? What knowledge was lost on me that the old Fibonacci had passed on as a legacy?
Well, Mr. Fibonacci was not exactly a trader in our sense of the word. He was a mathematician. I suspect that Mr Fibonacci himself, the medieval number genius he was, would be quite astounded at the impact his discovery had on the financial markets, or at least traders, searching for something to believe in and hold on to.
Discussing Fibonacci theory is like discussing the bible with a born-again Christian (There we go again with the poor born agains’ – sorry, being agnostic is every bit a leap of faith as God is). There is nothing that works but their chosen belief.
Mr Fibonacci’s mathematical series was constructed from observations on the incestuous copulation patterns of rabbits. You start off with a Mr. Rabbit and Ms. Rabbit, and then you let nature do what it is supposed to do. It grows, it expands. The Fibonacci number series mathematically calculates at what rate Mr. and Ms. Bunny Rabbit and their off springs pro-create. Incest police, close your eyes please.
In the financial markets the ratios of the Fibonacci numbers are supposed to guide or tell traders where there might be important support and resistance in the market. I won’t bore you with the ratios. You may know them as well as I do. Otherwise, you should familiarize yourself with them.
I found this description on a website, when researching for this article. It is a typical description of the Fib numbers, as we now call them. I have put my own comments in brackets:
“These ratios, and several others derived from them, (apparently) appear in nature, and in the financial markets they often indicate levels at which strong resistance and support (apparently) will be found. They are (apparently) easily seen in nature (seashell spirals, flower petals, structure of tree branches, etc.), art, geometry, architecture and music.
Why are they (reportedly) so important to the financial markets? Because (apparently) the markets tend to reverse right at levels that coincide with the Fibonacci ratios. Whether you see this as cosmic order or coincidence makes little difference. It happens and tens of thousands of traders make decisions based on Fibonacci ratios, thus amplifying the results.”
And here I was, trading away, thinking that the majority of people trading lost money. I am a special breed because I am still here. I have succeeded (at least for now) where thousands have perished. Now I am being told that tens of thousands of traders make decisions based on Fibonacci ratios, thus amplifying the results.
So I will ask this question, to you, as I asked myself: if thousands are using Fib, and most people fail in trading, then does Fibonacci work? Or is the problem with the trader? If this is a law, then if I follow the law, should it not guarantee my success?
It is time to pull out the charts:
I have identified a great area of support in the SP500 futures at price level 1268. It is a 38% retracement of the recent range. The market is in an uptrend on the larger time frame, and I am looking to buy into the uptrend. I can chose between 3 levels, the 38.2% level, the 61.8% and maybe even the 78.6% (the square root of 61.8%). I hope that the 38% retracement will provide me with a low-risk entry. The chart below tells me to buy.
I convince myself it will be the last time I am using the laws of rabbit breeding patterns to make trading decisions, but like any addict I will just have one more go. I mean, what else have I got to lean on?
Unfortunately the position goes in my direction, and I make a spectacular amount of money. My pattern, right or wrong, has been enforced. The chart below shows so clearly the 38.2% level was hit and a strong bounce ensued. I am brilliant for trusting my instinct, and I am once again, for now a believer of the medieval mathematician.
I should make a statement in the interest of honesty. I teach people on a regular basis the “art” of trading. I love teaching, because I learn so much myself about the market through the teaching and more importantly through the eyes of the innocent and uncorrupted eye. They have no RSI luggage, no bad stochastic experience, and no Moving Average conceptions. They just watch and stare while I pollute their minds with pre-fabricated rubbish, intended to equip them to make the financial markets their playground.
The best part of teaching is when the pupils get to ask questions. I love the Q & A sessions, because it is in those moments that the adults turn into children and ask the obvious innocent questions such as:
What if the market doesn’t stop at 38%, then what? Do I then buy at the 61.8%? Where is my stop? Do I need a stop, if this is supposed to be a law? What about 78.6% percent retracements – are they any good?
How would I know? I am just the teacher who over the last 2 hours has provided you with 50 slides where all the charts worked perfectly. How dare you question Fibonacci? Did you not know that the great Leonardo da Vinci in his sculpture of man (see below) irrefutably proved that the law of Fibonacci works everywhere, in man, as in nature and in the markets?
Of course the student is right. Where do I place my buy order? Where is my stop? If I buy at 38%, and it fails, do I buy at 61.8%? If I wait to buy at the 61.8% retracement, has the trend not actually turned, and I am trading against the trend? Unfortunately there is no one to ask. The Fib man himself is long gone, and has been replaced by thousands of experts who all interpret his finding differently.
Of course Fibonacci works often enough to make us hungry for more as the example above shows, and we quickly forget about all the times it didn’t work. Expect I don’t. I keep a diary, and in there, in the bloody mess of my trading statements, I see the result of relying solely on the Fib numbers.
I will give you a simple example, displayed on the chart below. On the way up from the bear market lows in 2002, did we stop at the retracement numbers?
Well, the 38.2% level had absolutely no impact. I guess you could say the 61.8% had an impact, but then what should I say about the level just before the 61.8% line, where the market stopped for 4 months. Is that a magic ratio? Or what about the level above 61.8% – actually I measured it and it is the 73% level – that is not a Fib number.
Even the secret 78.6% retracement level meant nothing to the market, but it did stop at the 91% level. It is not a level I have ever heard about, but maybe Mr. Fib will be disappointed with me now because I didn’t know the special powers of the 73% and the 91% levels.
So what am I saying? Am I dismissing Fibonacci?
Before I draw conclusions, I like to take you with my on a short chart journey, this time the focus of attention is the Dax stock index in Germany. I have analysed the daily chart of the Dax from the beginning of the rally in September 2011 to consolidation in August 2012.
I have made a note of 23 retracements. Of those 23 retracements, there were 8 perfect Fibonacci retracements, either being 38.2%, 61.8% or 78.6%, or a percentage of 8/23= 34.8%.
I have 15 retracements that don’t fit the Fibonacci ratios. I could proceed with a complete analysis of the 23 numbers, with a bell curve and a histogram of observations. I am not going to do that. It doesn’t prove or disprove Fibonacci.
It was interesting to note though, with the benefit of hindsight, that Fib ratios work very well when the “cycle” is new, i.e. when a high or low of some significance has recently been formed. In other words I would expect a 38.2% or 61.8% retracement if the market has just recently made an important high or low.
Even there I am skating on thin ice from a trading perspective. Which one is it going to be? The 38.2% or the 61.8%?
So I will repeat myself once more: what am I saying? Am I dismissing Fibonacci?
No, I can’t. There are so many who watch it that I have to keep an eye on the key levels. However, I do think it is utter rubbish to rely on it solely. I don’t think Fibonacci traders will fare well if they rely on it solely.
My friend Larry Pesavento is considered a living legend in the field of pattern recognition and Fibonacci ratios. I wonder how much of his success is the result of the incredible mental database he has stored subconsciously by having looked at charts for nearly 50 years?
I think it is utter rubbish, if you are day-trading. I even think it is utter rubbish if you are swing trading the market over several days. You might as well put some arbitrary lines up, based on a throw of a dice and watch them. If my success rate is 34% on the biggest stock index in Europe, then what does that tell me?
Every night I do my homework. It involves going through the trading day with the benefit of hindsight. I am looking over the intra-day chart for the Dax today. The chart is not shown here. I will just give you the low-down of the numbers:
We made a great 69.4% retracement of the recent range, which was also an 81.5% retracement of the intra-day range. J
69.4% and 81.5%. Maybe one day they will be recognised as the great mystical numbers that turns the Dax back down for an 80 point trade, but I somehow doubt it.
It doesn’t quite add up, does it?
I will make a statement, which is difficult to prove, but I think I will take on the bet, if you want to challenge me. For every one of your perfectly chosen examples, where the market retraces perfectly into a Fib ratio, I can show you two examples on the same chart, where it didn’t work.
I can go through chart after chart after chart and for every single example you give me of a Fib set-up, I can show you an example where it didn’t work.
Think about if for a second: if a market retraces somewhere between 35% and 41%, traders will say it is a 38.2% retracement or near as possible. If it retraces somewhere between 47% and 53%, then it is called a 50% retracement, even though that is not even a Fib number. If we retrace somewhere between 59% and 65%, it is called a 61.8% retracement. In short order we are giving the Fibonacci army an extraordinarily amount of leeway.
Some will use slightly more sophisticated ratios such as 0.786, and even 0.886. Some extremists will find comfort and meaning in the 0.941 or even the 0.236. If they make money with it, good luck to them. I just suspect that they are making money due to their money management skills and chart reading skills and their ability to interpret price action, rather than actually the Fib numbers themselves.
Do I want to dismiss Fibonacci altogether?
No. I need to know the numbers and I use geometry which combines the extent of the price moves with the Fib retracements. When they join forces, the signals are worth their weight in gold. It is called confluence. If I had the time and the inclination, I suspect I could shoot that particular argument to the ground as well.
However, my initial passion has cooled and has turned into sort of a seedy love hate affair. I make note of the signals, and I use my judgement of price action to take signals accordingly.
I will always keep an eye on levels where there is supposed to be support or resistance in the market. I just don’t get all flustered about it any more Trends develop and if I have to run scared every time we are approaching a Fib level, I would never make a dime. All I would do is feed my broker and pay commissions through the nose.
I have had some heated debates with Fib lovers. It is like discussing the bible with a believer. It is NEVER wrong. If it doesn’t stop at 38%, then it will go to the next level. A little bit like when the bible says “an eye for an eye” vs. “turn the other cheek”. You are fully covered. You can’t go wrong.
I admit I have worn the religion/trading analogy quite thin in this article, but I can’t resist a last flirt with controversy, so here goes. The Fibonacci revelation in this article is very much akin to having been a believer in God and his almighty creation, only to find that Steven Hawkins and his theory of the creation of the universe resonates much better with you, but you still achingly hope that there is a God, and that this life is not the end of your existence.
Finally I will leave you with this thought. Using Fibonacci Time Analysis, I have been able to get in and out of the market at some crucial times. I have made some outrageously correct calls from bull markets and bear market on TV. Recently I got a signal to buy the close in the SP500, purely based on Fibonacci time analysis.
Over the next 4-5 trading session the signal rewarded me with 700 Dow points. This was a time based signal. If I had followed the market in Price, I would have had to buy the Dow at a 44% retracement. I have never heard about it.
But of course that is not the fault of Fibonacci, although I thought it was a law, but maybe 44% is a ratio yet to be discovered!!