Developing the wave count from the October 2007 high - Part III

Since time is most important the first thing I want to do is find a time relationship between the first two waves since the October 2007 high. I start with the first two CITs on the chart presented yesterday. The time (in trading days) from the October 11 high to the November 26 low is 32 days. From that low to the December 11 high is 11 days. 11/32 = 34.4%. Since I require a relationship to be within 2% of a Fibonacci ratio this is not a valid relationship. 36.2 – 40.2 would have worked but not 34.4.

Since wave patterns can end either before or after a CIT (usually by only one fractal) I next look at the move from the November 26 low to the December 26 high. That is 22 days. 22/32 = 68.8%. Again, this relationship fails. Fibonacci is the key. These results indicate that the first wave down from the October 11, 2007 high ended before the price low of November 26.
Measuring time from the October 11 high to the November 12 price fractal low produces 22 days. From the November 12 low to the December 11 price CIT high is 21 days. 21/22 = 95.5%. This fits as I use 94.1 as a Fibonacci percentage. Next I look for confirmation of this pattern in price. The move from the October 11 high to the November 12 low was 137.56 points. From that low to the December 11 high was 85.04 points. 85.04/137.56 = 61.8%. It doesn’t get better than that!

The above calculations lead to the wave count presented today. As you can see, I am counting the action as an a-b-c Flat pattern where the b wave was itself an Expanded Flat. From the December 11 high the “c” wave has been developing into its own five wave impulse pattern. I talked about possibly taking a long position in SPY yesterday since wave v’ of c will be the last wave of the larger Flat. That trade never came off as Wednesday’s low was violated before Wednesday’s high.

For today, I can tell you that I have the proper time and price relationships where the entire flat from the October 2007 high may have completed yesterday. For trading purposes I will go long the SPY if we break yesterday’s high (without breaking yesterday’s low); or, we open in the lower 1/3 of today’s range and close in the upper 1/3 of the range. In both instances I would like to see a daily chart technical “buy” signal which would be divergence between the RSI and Composite indicators.

Developing the wave count from October 2007 - Part II

After using the higher time frame charts to get a feel for the overall Elliott structure, it is time to look deeper. Yesterday I arrived at the conclusion that either a flat, zigzag, or impulse was unfolding from the October 2007 high. Today’s chart is the daily cash S&P500. The dark red lines are the “waves” as derived from yesterday’s weekly chart analysis. If I were to use the price fractals and CIT’s from the daily chart I would see a five wave move down from the October 2007 high into the November 26 low. Note that in my work a price fractal high must be followed by a price fractal low and vice versa. Wave four overlaps wave one in this situation so I would be forced to conclude that a Leading Diagonal pattern had formed and was either wave 1 or wave “a” of a larger pattern. The rebound up to the December 11 high can be seen to be a three wave zigzag using the fractals. This then would be wave 2 or wave b of the larger pattern. Finally, from the December high it appears (using fractals alone) that we are in the third leg down of the larger wave 3 or c.

However, I don’t believe that analysis (as labeled on today’s chart) is correct. Complexity is added when one realizes that not all Elliott patterns end on the high or low of a move. Triangles and Flats are excellent examples of this phenomenon. To resolve the problem I rely on price and time relationships. Time is more important and Fibonacci the key. In tomorrow’s post I will start that analysis.

One other point that I would like to make today:
1) The cash S&P500 index may be making a short-term low here. I don’t have confirmation from either technical or price action yet, but any move above yesterday’s high would be a strong signal that a tradable bounce is underway. Therefore I will be looking to buy the SPY (in this blog related Investopedia trading account) at 139.13 unless yesterday’s low is violated first. If the trade is executed the initial stop would be at yesterday’s low.

Developing the wave count from October 2007 - Part I

As I have stated before I use price fractals (blue diamonds on my charts) and CIT’s (Changes In Trend, shown by green ellipses on my charts) to help me craft my wave counts. Assuming the October 2007 high began a new Elliott wave to the downside (it was the end of a 1-2-3 or a-b-c from the October 2002 low), the weekly chart shows fractal highs on October 12 and December 14. There is a fractal low on November 30. Therefore, we are in the third wave down from October.

What type of pattern is forming? Since the December high was lower than the October high and the current prices lower than the November low I can rule out a triangle pattern (both expanding and contracting). This leaves either a flat, zigzag, or impulse. Tomorrow I will dig deeper into the pattern but first a few quick points on the weekly chart.

1) Note that the long term moving average (the green line) has been violated and acted as resistance this week. This is the first time this average has not held since the upward move began at the 2002 lows.

2) The weekly RSI is now poised to close below the key 40 area for the first time since the upward move began in 2002. We will have to wait for the week to end before we can say this is fact.

Both of these events continue to make me believe that the pattern from October 2002 to October 2007 is now a complete a-b-c Zigzag.

Time

While we wait to see whether 1370.6 will hold or not I will change my focus to “time”. Gann said that this was most important. To start I will present a quick, simple idea.

Time is indeed the most important and Fibonacci is the key. As most know, 2.618 is an important Fibonacci number. In terms of time, this means that 26, 261/262, and 2618 time periods are also important. If you look at the cash S&P500 chart the time interval from the October 2002 low until the October 2007 high was not only a Fibonacci five years but also 261 weeks.

In my next few postings I will focus more on time relationships in building Elliott Wave counts.

Continuing the thought process

Although I presented a “guest” wave count on Thursday doesn’t mean I agree with it. I do believe that we have most likely concluded a full 1-2-3 (or a-b-c) up from the October 2002 low, but in the manner discussed in posts on January 6 and 7. The question remains whether we are now in a fourth wave or have started a “C” wave decline.

The weakness shown this week has me now leaning towards the scenario that a full a-b-c Zigzag pattern was completed at the October high. This means that the entire move up from the October 2002 low was corrective and part of a larger corrective structure from the 2000 high. This is illustrated in today’s chart.

If we are now in a “(C)” wave from the 2000 high there are only three Elliott wave patterns possible to explain the price movement: An expanded flat, contracting triangle or expanding triangle. Perhaps of some import is what I wrote back in early November in my review of the October 2007 monthly chart: “Most likely the rally from the late 2002 low will be confirmed complete if we break the trendline joining the August 2004 and August 2007 lows. That trendline (slope of 8.608 per month) stands at 1396.4 this month. Another important trendline (slope of 2.48 per week) in my work joins the June 16, 2006 low and the August 17, 2007 low. It stands at 1397.91 right now.”

We have now broken both of these trendlines and are challenging the more forgiving trendline (shown in today’s chart) from the August 2004 low to the June 2006 low. A break of that trendline and the August 2007 low (1370.6) would raise the odds considerably that the move from the 2002 low is in fact a zigzag pattern.

Let's see if 1370.6 can hold.