Last week I introduced the concept of Elliott Wave Theory and why it's so effective in measuring stock price movements. If you missed the article here's the link: http://www.optioninvestor.com/page/oin/commentary/newsletter/2006/02-25.html>
This week I want to continue the introduction to EW analysis and show some practical ways to use it.
Practice is the key. There are charting programs out there that make an effort to label the waves for you. I have found all of them deficient and it's mostly due to these corrective waves. Labeling waves by hand on your own chart is the single best way to learn the nuances of them and helps you get a feel for how the market is moving. An example of getting a feel for a wave count is what I call the sniff test. If it doesn't look right it probably isn't. Computer programs have a hard time with the sniff test whereas the human brain is very good (with a little practice). Since the corrective wave count is often not clear until after it finishes, or just before it's about to finish, the human brain is better than a computer at projecting (visualizing) possibilities.
As mentioned above, Fibonacci retracements of a move give us targets for a correction. The most reliable Fib retracements to use are 21.4%, 38.2%, 50%, 61.8% and 78.6%. The combination of these retracements and the Fibonacci projections based on Fib relationships between the waves can be very helpful in identifying price targets. Another useful technique, and quite simple to employ, is the use of trend lines and parallel channels. Trend lines are used by many traders and parallel channels do a good job at showing you the measured moves that the market tends to make.
The wave patterns very often stick to trend lines and parallel channels. In an uptrend take a parallel to the uptrend line and attach it to the first high between the two lows. Referring to the figure below, you first draw a line from the start of a rally through the first pullback which is wave-2. The parallel is then attached to the top of wave-1 and the extension of it is often times a good guide for where wave-3 will stop.
Drawing Parallel Channels
After drawing in the trend lines as described above, once it appears wave-3 may have ended, draw a line from wave-1 to wave-3 and attach a parallel to wave-2. This lower line will often mark where wave-4 will find support. If wave-4 stops at a different place than the parallel line, draw a line from wave-2 to wave-4 and then attach a parallel to wave-3. This parallel line will then often mark where wave-5 will end.
One other thing that I like to do, which is not shown above, is draw in a mid-line centered between the two parallel lines. This often marks minor support and resistance during a move and it very often marks where wave-5 will end (in other words it often does not make it up to the top line). Most programs will draw regression channels that are similar to these parallel channels but I like to draw them manually since they're off the EW count.
Drawing these kinds of parallel channels also works for corrections. Once you have two highs in a pullback and draw your downtrend line on them, take a parallel of that line and snap it to the low that is between those two highs and that often marks where the next leg down (wave-C) will stop. This will be especially true if the market gives two equal legs up or down in its correction (wave-A = wave-C).
Let's look at a couple of real life examples. Actual wave patterns are never as cut and dried, or as easy to identify, as the above patterns and this is where the subjective interpretation comes into play. A combination of EW analysis, Fibonacci retracements/projections and trend lines/channels makes for a powerful combination to determine price moves. Add in your other favorite technical tools for confirmation and you've got a trading system that is hard to beat. Let's take a look at a rally in the RUT to dissect the move and see how the count progressed:
Waves (1) to (3)
After the rally got underway, you can take a guess at waves (1) and (2) and then once we got a deeper pullback underway from the high in mid June we could label that as wave-(3). As an alternative it would have been labeled A-B-C in case it was going to be just a 3-wave upward correction against the decline from earlier in the year. So a red trend line was drawn from wave-(1) to wave-(3) and a purple parallel line was attached to wave-(2). This provided a guide as to where wave-(4) might pull back to. Some times it drops down to it and other times it chops sideways over to it.
And sure enough wave-(4) stopped near that lower trend line (middle of the chart below). Now it became a question where wave-(5) would end.
RUT, Finish Wave-(4) and Start Wave-(5)
As the rally progressed and came up near the upper trend line in mid July, it looks like it ended and first impression is that wave-(5) ended at that high. But looking a little closer at the inside of wave-(5) shows why it wasn't to be trusted as the end of the run up--as the rally progressed we got a quick succession of two highs and pullbacks, labeled waves 1 and 2 and then (i) and (ii) in the above chart. It can't be waves 1-2-3-4 where you'd label wave-(i) on the chart above as wave-3 intead and wave-(ii) as wave-4 because 4 would overlap wave-1 and wave-3 would be the shortest wave. Both would be rule violations.
So the small rally and pullback in the first week of July has to be another smaller degree 1st and 2nd wave as labeled above. This is the reason the high in mid-July would be labeled wave-(iii) and it meant we needed to look for waves (iii), (iv) and (v) to finish wave-3. Moving to the next chart, pick up from where we had wave-(iii) near the upper trend line, again about in the middle of the chart.
RUT, Finishing Wave-(5)
Remember, each impulsive wave, as wave-3 of (5) needs to be, consists of its own 5 waves so after wave-(iii) and then the pullback to wave-(iv), the next leg up was wave-(v). That finished one larger degree wave which was wave-3. Notice the negative MACD divergence against wave-(iii) and wave-(v) in mid July. The 5th wave is almost always negatively divergent against the 3rd wave and it's something you want to check for to add to your confidence that you have the correct count. Use either MACD or RSI for this as they're very powerful tools in EW analysis.
After the completion of wave-(v) and therefore wave-3, we got a sideways triangle wave-4 into the end of July and note how this predicted the last move up to wave-5 in early August. Sideways triangles in this position of a rally are highly accurate in predicting the last move is coming. Once wave-5 completed it finished the next larger degree wave-(5), thereby completing the 5-wave count up from the April low. Note how wave-(5) finished essentially at the mid-line of the up-channel, a common occurrence. Note also the continued bearish MACD divergence at the wave-(5) high.
One last technique to help identify where the final 5th wave will end is to look for the relationship between the 1st and 5th waves. The common relationship between these two waves is that wave-5 will equal 62%, 100% or 162% of wave-1.
RUT, Using Fibs to Project wave-(5)
This chart shows a Fib projection for wave-5 of (5) based on the size of wave-1. QCharts has a tool to do this but you can obviously calculate it by hand. Wave-1, of wave-(5), at the end of June, ran from 625.84 to 646.04 so 20.20 points. From where wave-4 ended at the end of July, at 668.82, adding 20.20 points gives us an upside target of 689.02 which is what the Fib projection on the above chart shows. Had the rally not stopped there, I would have been looking for the upside Fib target of 701.50 up near the upper trend line.
Once price broke down from the parallel channel we had confirmation that the 5-wave move up from April had completed. Many times price will then rally back up to test the bottom of the channel which is usually a good spot to get short. And then once that bounce completes and price continues lower you have your first two points (the high of the rally and then the high of the bounce) to draw your first downtrend line. Wait for the next leg down to complete and snap a parallel line to it and start the whole process all over, this time looking for a corrective pattern that corrects the 5-wave impulsive move up from April to August.
As I said at the start, this is not an easy tool to master. If you're serious about adding this technique to your toolbox, here is a recommended reading list. I would read them in the order listed so as to get a better understanding of the subject without getting overwhelmed with detail that's hard to understand. The last two books are less about the details of wave structure and more about the market and the economy and how you can see the application of EW Theory. You may email me any questions you have on the subject. Good luck and have fun with it.
1) The Elliott Wave Principle Robert Prechter and A.J. Frost