Elliott wave methods I

There are 2 manual methods and 2 computer-based systems for counting Elliott waves. I will use this Blog to illustrate 3 of them, and leave the fourth for those who are curious enough to buy the software and put it to good use.


The four methods are: 1. The manual channeling method. 2. The Neely Method. 3. The Advanced Get software. 4. The Refined Elliott Trader software.


The Manual channeling method is described in the basic book by Robert C. Beckman, "Elliott Wave Explained", and in lesser detail by Frost and Prechter, "Elliott wave Principle".


Let's take a look at a historical chart of the Nasdaq Composite (^IXIC in Yahoo financial, if you'd like to download the data). I am going to try to do the counts with each method, and see if we can arrive at the same conclusion.





Find a suitable turning point after the completion of a previous wave. For the Nasdaq composite, the lows of 1974 seem appropriate. Using a ruler or computer pencil tool, draw a line with an origin at the suitable turning point. By raising the angle of the ruler, find the first exposed contact point, which becomes the presumptive wave 2, until shown otherwise.



Next, draw a parallel line to the baseline that will be touched by the highest point preceding the presumptive wave 2. We can label this point as wave 1.



One begins to construct and imagine the sequence to follow in wave 3, but only if the impulse character of the sequence continues. We can only be sure that wave 3 is in progress, once the highest point of wave 1 is exceeded. Even then we cannot completely rule out a corrective series where A would substitute 1 and B would substitute 2, and wave C would be a series resembling wave 3.

Wave 2 can retrace as much as 80% (some think even 99%) of wave 1. However, if wave 1 is exceeded with a strong move in above average volume, the trader can be reasonable certain that wave 3 is in progress, and that a strategy can be designed to take advantage of the rally. Yes, one can catch a wave 3 in daily, weekly or even monthly series. As long as you are reasonably certain that the configuration of waves 1 and 2 correspond to a 1-2-3-4-5 and a-b-c series respectively.

Once Wave 3 is underway, the easiest target to estimate as the possible highest point of wave 3 is the upper line of the trend channel. If you are looking at a long-term chart (not a daily chart), the log-transformed chart should be used. Even then, wave 3 will exceed the trend channel, and begin to decline aggressively.



Trust me on this one: people trading in such market, believe that they go nothing but UP, and are willing to bet their pensions on it! There is simply no way to convince them that a big decline is in the works!

Since wave 2 was a rather tame series, and the low of wave 2 being actually above the top of wave 1 (this is called a running triangle), one should expect wave 4 to be exactly de opposite, according to the rules of alternation. Only one thing is for sure: the base line will be violated!

Following the same method, we can look at the components of the proposed wave 3 and see if we can predict with some accuracy when will wave 3 come to an end and wave 4 begin.



Here in cyan color are the labels of the components of wave 3. Yes, wave 3 itself is composed of 5 "impulse" waves. Ironically, subwave 3 did not exceed the established trend channel. This may be a hint that our counts are not properly represented by this series (more on this later).

Using Beckman's nomenclature, Multiyear wave 3, the aggressive bull market, should be called intermediate. Wave 5, the last wave component of intermediate wave 3 is strange in its shape to say the least. However, it still fits the accepted Elliott Wave precepts. Take a look:



The unusual shape of a series of impulse waves is acceptable, as long as the following rules are followed:

1. One of the three waves will be significantly larger than the other two. That is, the extended wave.
2. Wave 3 is usually the extended wave, but when it is not, it cannot be the smallest wave.

In other words, if wave 1 is the extended wave, wave 5 is the smallest. When wave 5 is the extended wave, wave 1 is the smallest. Wave 3 is usually the extended wave.

It may appear then that wave 3 was the extended wave in this series. Wave 5 was the smallest, and was actually smaller than wave 4 and wave 1. Is this acceptable by Elliott rules?

Here is what Neely says about wave 3 extensions:
When wave 3 extends, then it must be more than 161.8% of wave 1. In the current case, wave 3 was 2500 points while wave 1 was 1515 points, a relationship of 165%.

But why was wave 5 so short? is that acceptable by the Elliott theory? It may appear so. Wave 5 was in fact 80% of wave 1. There is no rule that says that wave 5 could not be the smallest or that it must exceed the end of wave 3. In fact, when wave 3 is not exceeded, it is said that there is a wave-5 failure. This brings to mind the dangers of trading on wave 4 bottoms in the hope to have an extended 5th wave. If wave 3 exceeds wave 1 by 161%, wave 3 was the extended wave, and an extension of wave 5 in that situation is unlikely to occur, and a failure is almost assured, since wave 5 need only be about 100% of wave 1 (this is an interesting piece of information for Advanced Get, Type I traders).

If the w2-w4 trend line is broken before wave 3's terminus is exceeded, then a 5th wave failure is indicated. For the 5th wave failure to be proven, the entire impulse pattern must be completely retraced faster than it took to form. Afterward, the market can drift a while or continue on its course, away from the 5th wave failure. To exceed the end of the 5th wave failure, the market would need to consume at least twice the time taken by the entire impulse pattern (1-5); usually much more time is required before a new high takes place!!!



The chart above shows the immediate retracement of the wave 5 of Intermendiate wave 3. Needless to say, the Nasdaq is not anywhere near a retracement back to the 5th failure, even today.