Hurst Cycles and Elliott Wave

I have started writing this post several times over the past years, and have always given up after hours of work, with pages and pages of dense comparisons and similarities between Hurst Cycles and Elliott Wave. This time I am going to publish!

The problem is that the subject is such a big one, and I have a tendency to get carried away (in case you hadn’t noticed!). Before long the post is too long, and so I trash it and promise myself I’ll revisit the idea later. Well, later has arrived! A post by Sid about the AUDUSD has prompted me to publish this. My secret this time is that I am only going to scratch the surface of this vast subject by presenting a starting point for the discussion with a few simple points.

I should say at the outset that I am by no means an Elliott Wave expert. I have a good understanding of Elliott Wave, and traded on the basis of my amateur EW analysis rather unsuccessfully for several years before I discovered Hurst Cycles. And so please get involved in the discussion, point out any Elliott Wave mistakes I have made, or subtleties I have missed, and share your thoughts about how Hurst Cycles and Elliott Wave work together.

Two Cycle Theories

Of course at the most basic level there is an obvious link between Hurst Cycles and Elliott Wave Theory: they both describe cycles in financial markets. I was surprised to realize this at first because I had always focused on the concept of five moves up forming an impulsive move, and three moves down a correction, and had not considered Elliott Wave as a cycle theory per se. Of course the word “wave” is a dead giveaway.

I probably didn’t think of it as a cycle theory because it is not very common in Elliott Wave analysis to speak of particular cycle lengths, whereas in Hurst Cycles analysis you are constantly doing that (here’s a fun challenge: see if you can speak for an hour about 20-week, 40-week, 20-day and 40-day cycles without getting it wrong – I know I cannot!) Of course many EW analysts do mention cycle wavelengths, but as far as I know there is no official Elliott Wave “nominal model”.

And so at the most fundamental level the two approaches are both about identifying cycles in the price movement, and of course those cycles push price up, and then down again.

Cycle Shapes

The next interesting comparison between these two approaches has to do with what I call “cycle shapes”. Cycle shapes are not strictly speaking a part of either Hurst or Elliott, but they do form as a result of the underlying principles of both theories.

You will know that I go on a good deal about the M-shapes that cycles make in financial markets. They do this because of the way in which multiple cycles combine to influence price movement. And the way in which multiple cycles combine was defined by Hurst in his cyclic principles.

Elliott Wave also leads us to expect particular cycle shapes in the markets, because of the alternation of an impulse and corrective move. Here is a direct comparison of the basic cycle shapes each approach leads us to expect:

You will notice in that diagram some obvious similarities. There is a lot more to be said about this, such as the fact that a corrective wave in Elliott Wave consists of a combination of impulse and corrective waves itself, and of course I have drawn a bullish impulsive Elliott Wave whereas you get bearish impulsive waves as well. But I will have to come back to these discussions in future posts.

The number of waves

Perhaps the most fundamental concept of Elliott Wave theory is that price moves in impulsive waves which consist of five moves, and then corrective waves which consist of three moves. I know it can get more complicated, with extensions of waves, but let’s keep it simple for now.

Most people who have discussed with me the relationship between Hurst Cycles and Elliott Wave see this five and three wave combination as being a point of difference, but in fact it is a remarkable point of commonality between the two approaches.

Let me explain: Hurst Cycles lead us to expect M-shaped cycles, which consist of four waves (the basis of the “sentient trading methodology”). We also know that the ratio between the wavelengths of adjacent cycles in Hurst’s nominal model is mostly 2 to 1, and so for each longer cycle we have two iterations of the shorter cycle. (For instance the 80-day cycle encompasses two 40-day cycles). As each shorter cycle consists of four moves, the longer cycle consists of eight moves.

And so we have the commonality of 5 + 3 = 8, and 4 + 4 = 8.

There is an interesting detail here: in Elliott Wave there is a focus on the move from trough to peak, and then from the peak down to the trough, but in Hurst Cycles we are focused on the move from one trough to the next, and then to the next.

A question might be bothering you: what if the peak in the Hurst Cycle occurs after wave 3 in the Elliott Wave analysis? Because it cannot always occur in the second Hurst cycle.

There are two answers to this:

• Elliott Wave allows for what is called a truncated fifth wave, which does not rise to a new peak.
• It also depends on where one starts counting the waves. This is a big area to explore, but let me plant this seed of an idea: it is quite possible that the Hurst cycle starts at the position of a wave 2 in the Elliott Wave count, or the wave 4, or the wave b … you get the idea.

8 waves and 8 interactions

You probably know that I like to trade the interaction between price and the FLD (the FLD Trading Strategy). There are eight interactions between price and the 20-day FLD within each 80-day cycle.

There is that number 8 again. This is another common ground between the two approaches.

This subject also opens up another rabbit hole of comparisons and relationships, but I will leave you to consider the possibilities.

Alternation

Let me point out another interesting subtlety. Elliott Wave observes something called “alternation”, which leads us to expect that wave 2 and 4 in an impulsive move will take on alternating forms: if one is a sharp correction then the other is often (not always) a shallow sideways correction.

This alternation is also observed in the interactions between price and the FLD. Between interactions B & C we usually see price “track along” the FLD, whereas at interaction D it usually drops sharply through the FLD.

Again bear in mind that the B-C interactions are not necessarily wave 2 in the Elliott Wave count, nor is interaction D necessarily wave 4. But this idea of alternation suggests further possibilities: if we see an unusual B-C interaction which is a sharp move, should we expect a disappointing D-interaction?

Further areas of discussion

I have merely scratched the surface of this fascinating subject. Here are a few quick thoughts to stimulate further discussion:

• Price and Time: Elliott Wave focuses primarily on price moves, and uses Fibonacci ratios a good deal. On the other hand Hurst Cycles focus primarily on time. The combination of the two is very powerful – you benefit from the forecasting power of both price and time.
• Fibonacci ratios in time: Having said that Elliott Wave focuses primarily on price, Fibonacci ratios are often applied to time ratios in Elliott Wave (the 5 + 3 = 8 waves is pure Fibonacci after all), and there is a great deal of insight that can be gained from applying Fibonacci ratios to the time axis and see how they relate to Hurst’s nest-of-lows.
• The Fractal concept: Elliott Wave proposes a fractal model of price movement, and yet Hurst Cycles imply an imperfect (or nearly) fractal model because the harmonic ratio between cycles is not constant through all time frames. This results in a very interesting dynamic between the two analyses.

Please share your thoughts about how Hurst Cycles and Elliott Wave work together, what they have in common, and where they contradict one another. I look forward to some very fruitful discussion.