Saturday, January 2, 2016

The Eight-Fold Path Method for Counting an Impulse

A much wiser man than I called the "The Eight Fold Path" the way to enlightenment, and it is in that spirit of sharing that this post is made.

When you survey the world of Elliott Wave analysis, you find that many casual wave analysts will post a chart of "something", with some numbers and letters on it and call it a wave chart, or maybe their "best guess" of where prices are going. But we contend unless people follow a very simple, eight step procedure they will be lost in making objective Elliott Wave counts: counts which are predictive - or like in fourth waves - clearly indicate when prediction is going to be sketchy, at best.

So without further delay - let's outline those simple eight steps, with an example below the steps. All though I didn't explicitly say it in the first posting of this article, I need to make it clear that this method applied to counting Elliott Impulse waves only. In fact, if you follow this method, and the wave does not conform to these guidelines then most likely the wave is a diagonal or part of a larger corrective structure, and not an impulse wave! And knowing that can be a great deal of help as well.

Step 1 - chose the time frame that provides 120 - 160 candles for the wave of interest. Does it have to be exactly 120 to 160 candles? No, but it should not be 80, and it should not be 220. We'll talk more about why at the end of this blog post.

Notice here, we did not say, pick an hourly chart, or pick a daily chart, or pick a 5-minute chart. Elliott patterns unfold in clear waves. Your first job is to decide which wave you are analyzing. Are you analyzing the five-waves of a potential impulse third wave? Are you analyzing the A wave of a zigzag A-B-C structure? Or, are you analyzing the entire bull market wave since March, 2009?

Step 2 - use a bar chart for that wave, and include on it the Elliott Wave Oscillator indicator (EWO). This indicator is also sometimes known as the Awesome Oscillator. Regardless, its just the difference of two simple moving averages; one of 34 periods and one of 5 periods.

Step 3 - locate the maximum value of EWO. This location is reserved for wave iii of 3 in an impulse wave. Note the value of the EWO at this location.

Step 4 - locate a 1.618 or 2.618, or (more rarely) a 4.236 extension following the first wave up. This is most likely where wave 3 in an impulse wave will end, and it will end on a divergence from the EWO value in Step 3.

Step 5 - draw a parallel Elliott channel around waves 1, 2, and 3 in an impulse (which alternatively in every wave can be A, B and C).

Step 6 - watch for the EWO to retrace to near the zero line for wave 4. Computer studies have shown that most wave four's will have the EWO retrace to +10% - to -40% of the maximum value of the EWO in Step 3. Further, verify, by measurement that wave four only retraces 23.6 - 50% of wave three, with 50% being an absolute maximum. Beyond, that level any wave four call has extremely low odds.

Step 7 - in an impulse look for wave four to attack the lower channel line in some manner, but then rebound quickly from it - even if there is a false break. In a true impulse, Wave 5, upward, should be under way and should reach a length of 5 = 1, or alternatively, 5 = 0.618 x (net 1 - through 3). Furthermore, wave 5 must end on a further divergence with the Elliott Wave Oscillator.

Step 8 - Verify, verify, verify that there was alternation present. Most second waves are deep, sharp waves. They are typically zigzags or double zigzags that retrace 50 - 62% or more of their first waves. Fourth waves are 'usually' long, complicated sideways affairs such as triangles, flats or compound waves like flat-x-zigzag. Regardless, there should be alternation in a true impulsive structure. In the rare cases where wave 2 is a flat, then wave 4 will 'most likely' be a triangle or a zigzag to alternate with it.

IF the potential WAVE 4 breaks the channel, and does not rebound,  or the EWO takes on more than -40% of the value of wave iii of 3, on the opposite side of zero, then, most likely, the up wave was only the alternate A-B-C, and there is nothing the Elliott analyst can do except objectively accept the result. 1,2,3 and A,B,C are equivalent until they are not.

This is a chart to illustrate the above concepts in counting an impulse wave - objectively. First, I decided to study the "up wave from November, 2012 in the Dow Jones Industrial Average", and selected the time frame that best put 120 to 160 candles on the chart. That time frame was "daily". Second, the Elliott Wave Oscillator was added to the chart. Third, the maximum of the Elliott Wave Oscillator was examined, and determined to be the location for wave iii of 3. Fourth, we note that wave 3 just exceeds a 1.618 extension. Note that wave v of 3 ends on a divergence of the Elliott Wave Oscillator.  Fifth, we draw a channel around waves 1,2,3 (or alternatively A,B,C until an impulse is 'proven'). Sixth, we note that the EWO comes very close to the zero line for wave 4. Seventh, we note that wave 4 attacks the lower channel boundary, and wave 5 quickly resumes, and makes a wave which is 5 = 1, and  actually, just slightly beyond. Note again that wave 5 ends on another divergence with the Elliott Wave Oscillator. Finally, and Fibonacci eighth, we verify alternation. Wave 2 is a classic zigzag sharp with a 50% retracement, and wave 4 is a 'compound' FLAT wave. Its a flat to blue 'a', then a flat to blue 'c' & black 'a', then a larger flat to black 'c', & black 4.

Look at how time consuming these fourth waves are! And you know what? I almost guarantee you that no person alive predicted this exact structure. This is the very meaning of the "fourth wave conundrum" as we have presented it in our YouTube video. So, in each case you know what everyone is asking for the better part of a whole month, "What's the count? What's the count?", and thinking, "what kind of Elliotician are you that you don't know what the count is, right now? how am I going to make money following you?! I'm getting whipped around like crazy here." And on and on it goes in those pesky fourth waves.

And yes, there are even slight variations on this - such as when the first wave is the extended wave, but this is the only objective way I know to count the Elliott impulse wave. And, it was discovered by Master Trader Bill Williams, using mainframe SuperComputers to crunch all the data. So, when people say they are being "objective", see if they are meeting this type of simple, defined criteria.

And, yes, I hear you Gen-X'ers and Millenials: you got it. This is a procedure. And you hate procedures. Really? Do I have to follow a procedure? Well .. let's just say "only if you truly want to see the Elliott Wave at work."

Ok, now, for those of you interested in only the modest amount of mathematics, I told you I would tell you the secret of why you need to pick the time-frame that best provides 120 - 160 candles on the chart. Here you go: it is because it takes 5 x 34 = or 170 candles to produce five oscillations around the 34 period moving average of the Elliott Wave Oscillator. Yes, the bright ones among you got it right: the five "oscillations" are the five waves of an Elliott Impulse Wave. So, if you are going to have a chance of seeing that fourth wave come back to the channel with the EWO nearing zero, it must happen starting sometime soon after 120 candles are on the chart.

That's all there is to it! It's that simple! Don't make it more complicated than it is!



  1. Thanks a lot and have a wonderful 2016, hopefully with lots of the same brilliant Elliott analysis and educational material.

    1. Thanks Wieb. Have a great & profitable New Year as well! And thanks very much for the support.

  2. Once again, Joe, awesome information that we can utilize and put to our own use instead of relying on others. Thanks for all you do!

  3. This is a great post. I'll try this out stat.
    I thank you.

  4. This is absolutely enlightening, Joe!
    Adding this post to my library on EW.
    With your expertise, you should definitely write a book on the subject.
    My wave4 zigzag-X-triangle "count" on SPX was busted yesterday per your correct odds assessment.

  5. thx TJ for the explanation
    like me many miss caldaro's blog
    we hope to see the blog on again

  6. This comment has been removed by the author.

  7. This comment has been removed by the author.

  8. Hi Could you please explain this statement better? "Computer studies have shown that most wave four's will have the EWO retrace to +10% - to -40% of the maximum value of the EWO in Step 3"
    are you sure that +10% to -40% has to be calculated from the maximum value of EWO in step 3 ? or this +10% to -40% is a zone from zero line? i mean it has to be measure from zero line not from maximum of the peak. Could you please clarify. Thanks
    are you trying to say if EWO peak is at 8 then 10% of 8 is 0.8 and 40% of 8 is 3.2 so EWO can oscillate between +0.8 to- 3.2 from zero line,for wave 4

  9. This comment has been removed by the author.

  10. Joe, thank you, for sharing your knowledge with us. I have a question though, if wave 1 is extended does that mean the largest value on the EWO indicator will be wave iii of wave 1, instead of wave iii of 3? Does this rule apply if we have a 5th wave extension? I have enjoyed reading your commentary, and please ignore the trolls.

  11. Hello, Joe, I have another question in regards to the Eight Fold Method. Does every impulsive wave structure have to adhere to the number of candlesticks mentioned by you? Sometimes, I have noticed on a lower time frame such as the hour, that the amount of candles to finish the wave is 300-400. Please carry on with your commentary, I have learned quite a number of things from you. Thank you.

  12. I don't think there is overlap until 2626 as I had the first wave bottoming on oct 26, the lower low it made after as a b wave of an expanded flat of 2. The fifth wave of the first was an ED which was complete on 10/26. Just a thought