Friday, September 25, 2015

Snake Oil or Not?

Many are aware of how the expression "snake oil" arose from certain products being sold in the American West of yore. If you are not familiar with this term, you can read about it here. Here it used as a title for this post because many believe or have read or have heard from others that all Elliott Wave analysis falls into this category : as a 'cure all' for market analysis that is nearly infallible.

Because many, many people constantly challenge me in what they think are cogent arguments about the usefulness of Elliott Wave theory, I have decided to write this short perspective so I do not repeatedly have to address this issue. It takes too much of my time, drains my energy, and generally puts me in a bad mood to constantly address it during the trading day, or in chat rooms. So, I will do it here, and simply refer people back to this original work as needed in the future.

Who makes these arguments? 

Usually the people who make these arguments are only vaguely familiar with Elliott Wave work. They may have read the Elliott Wave Principle (EWP) by Frost & Prechter, once. They may have understood some of it, but they haven't even committed the rules and guidelines to memory so they can see if they apply them real time. Then again, they may not have read it, the EWP, even once.

If they are familiar with the work, these individuals may have full time jobs, and simply may not have had the time to try to apply it "real time" in markets. This clearly is not as inexcusable as someone who has not read the work with full understanding.

Regardless of the knowledge level, these people are keenly aware of how one of the key developers and practitioners of wave theory has made some very bad calls. For this reason, they either see Elliott Wave Theory as 'useless' or, worse, actually 'detrimental' to market performance. And that is what those who argue against Elliott Wave theory usually use as the 'silver bullet' to end the Elliott Wave argument, once and for all, and for all time : they use someone's else performance.

But, beyond that, usually these people have not read the more modern works on Elliott Wave theory. These works include those by Steven Poser, Glenn Neely and Bill Williams. Clearly, when Robert Prechter had his forecasting difficulties, others seeing 'some' value in Elliott Wave analysis tried to advance the work. In fact, the modern state of Elliott wave analysis includes all of these works. Glenn Neely, in Mastering Elliott Wave Theory, sought to quantify typical wave patterns and provide better realistic representations of what the common wave look like, than the EWP did. Bill Williams, PhD., is a master trader who has sought to make wave theory more reliable, not by throwing away anything in wave theory, but by updating it for the modern science of chaos theory, and by adding a relevant series of indicators to guide wave formation.

More Reliable? - Not Good Enough

I know what you're thinking. Hey, bud, I'm a trader (or investor) and I have to make a trade or make a decision. I need to be sure that decision is correct because it's my money. Well, my first response is, that's a nice goal : to make a highly reliable decision. But, to never make a mistake is a bit unrealistic. In fact, let's call it a pipe dream. In no other endeavor do people not make mistakes. Mistakes happen in business, they happen in war, they happen in marriage, they happen in sports. Why would we expect no mistakes to happen in market prediction?

Yes, the fact that mistakes happen is why the issue of reliability in forecasting is an issue. Not all forecasts or predictions are 100% reliable : in fact, none are. To make this issue clear, I have taken them time to construct this graphic. Hopefully, it is a picture you can keep in mind.

Likelihood of a Correct Prediction


This chart shows how likely you are to be correct, if you make a certain prediction. For example, if you predict that the sun will rise tomorrow, you have greater than a 99% chance of being correct. That is because in more than the last 100 days, the number of times the sun has not risen has been exactly zero. But one day - for some reason - the sun will likely not rise. So, we can not say the chance is 'exactly' zero. But for all practical purposes it is 'near zero'.

In a similar vein, the likelihood that you will awake tomorrow is also greater than 99%. In the last 100 days, if you are reading this, you have woken up every day. But one day, it is certain you will not. So again, the probability of waking up is near 100% - just not exactly 100%.

I won't go through every example in excruciating detail. You know that local weather forecasts are now getting to be about 70 - 80% reliable for the near term. They 'can' tell you, for example, the chance of rain is 20%, tomorrow, for example, and be quite correct. The other examples are a coin toss - which you know is a 50 - 50 chance; predicting you will win at a game of solitaire about 17 - 25%, depending on skill; predicting a U.S. Citizen will never pay a dime in payroll tax  - about 14% of Americans never have. Yes, an asteroid will hit the earth again - one wiped out the dinosaurs, but it does not happen at a frequency of once in a hundred days, thankfully, and, yes, 'some' day the sun will become a red giant star, or explode, based on the history of other stars.

Elliott Wave Impulses

I won't go in to much detail now, but when an Elliott Wave impulse gets underway, it is the 'more' predictable of the patterns. There is 'better' than a 50:50 chance that it will have five waves, that the third wave will have the most momentum, that the third wave will have a price gap in it, and that the fifth wave will have less momentum than the third wave.

Once a true impulse starts, there are only three ways it can form : an extended first wave,  extended third wave, and extended fifth wave.

This is useful information. Is it infallible information? No, it is not. What you do with that information is up to you. You can use it effectively or you can misuse it.

Elliott Wave Corrections

By contrast, when an Elliott Wave correction begins it is the most notoriously difficult of the Elliott Wave patterns to predict. That's because the number of possible corrective patterns explodes enormously. There are sharps, which are zigzags, double zigzags, triple zigzags, and then there are triangles of several types (symmetrical & barrier contracting, and expanding triangles), and then there are flats of several types .. a few of which are regular flats, irregular flats, expanded flats, as well as double-three's and triple-three's.

Because there are so many types of corrections, your ability - even as a knowledgeable wave trader - to correctly predict which is going to form is highly limited.

Folks, let me be clear, this is just the nature of the game. It is a huge limitation. Yet, once a correction is over, just the fact that it is over and the type of correction that formed does, indeed, again provide you with more information. The information is there for your use, or again, your misuse.

Geez - yet more problems?

Readers knowledgeable of Elliott Wave theory will recognize that I did not even include the pattern termed the 'diagonal' in this discussion. This pattern both increases and decreases the useful of wave theory in certain situations. Sometimes it 'definitely' signals the end of a wave, but sometimes, in a fake-out or not, it is really signalling the beginning of the next wave.

And, then, there is the issue of "long range" forecasting versus short range forecasting. Just like the weather, it is much, much harder to accurately predict whether it will rain on June 21st in Miami three years from now, then it is to predict whether it will snow tomorrow.

So, the more wave sequences in the future one tries to predict, the inherently less reliable the Elliott Wave forecast is going to be. So, let me say it again, predictions of Primary sized - monthly sized - waves, for example, are much more inherently difficult than predictions of minor sized - daily sized - waves. This, too, is simply the nature of the beast.

Summary

From this discussion some readers can glean that if they truly understand wave theory - and not just some vague conception of wave theory - they might have the opportunity to make some more successful predictions than flipping a coin. To which some smart-alecks might just say, "too much work! I'll go with the coin!". And, the more short-term the prediction, the better chance it has or the fewer possibilities there are to contend with.

And, yet, others might recognize that maybe it wasn't wave theory, itself, that resulted in the poor but highly recognized calls of an individual or organization, but maybe the specific way they applied wave theory. Maybe it was his, her's, or their desire for fame and personal excellence or maybe it was an organization's need to sell newsletters of dramatic consequences that led to the incorrect predictions.

Which camp will you fall in? Will you recognize or not, that predicting waves is an exercise in reliability or probability, or will you just 'give up'? And will you recognize that modern wave theory is beyond where Prechter got it, or not? Most importantly, will you recognize that not only is it the theory, itself, but it is also the person using it, not using it, or misusing it that is also a factor in a successful prediction? That person is you!

Your call.

2 comments:

  1. Thanks E-T for this blog! I've been wanting you to do something like this for A LONG TIME!

    I've learned a lot from you guys over at EWT&C and I still have all my notes and I still read the Prechter, Neely, and Williams books that I have. I finally got smart and organized all my notes for easy reference and now when I need to check something, I just grab the label and look.

    These blogs of yours are gems! Absolute treasures in a small package. Thanks for what you are doing!

    Mark Alan

    ReplyDelete
  2. Well written article Joe and straight to the point. For those that have done the work, the research and studied the charts they know the usefulness and limitations of Elliott Wave theory. In my opinion it is not a theory but absolutely defines market behavior. However, with that comes the understanding of the fractal nature that makes it an exercise in probabilities of what may happen next. In the end it will never be the chart that is wrong only the observer. Studying your material over the past few years I have found your knowledge and application of Elliot Wave analysis to be unmatched. Your greatest strength is remaining objective. That is where most fail by fixating on what one believes will the the absolute outcome then expecting the market to fulfill their expectations. This is Elliott Wave misapplied. Keep up the great work and don't spend your time or energy debating detractors. The work is there for them to use or not use.

    ReplyDelete