There are 3 types of corrective patterns that run counter to the dominant trend as identified by RN Elliott (of Elliott 'wave' fame) and I can't dispute his take on it. Elliott also talked about the 'Rule of Alternation' as an important principle of analyzing the markets. In my last (2/4) Trader's Corner which can be reviewed HERE, I wrote about the 'typical' bearish correction as having 3 segments, consisting of a decline (labeled 'a/A'), a rebound (labeled 'b/B'), which is then followed by another decline (labeled 'c/C' in wave terms). I got a few more questions on the nature of corrections and when you can tell if they are over as well as to say something more on the so-called Rule of Alternation.

I went back, as I periodically do, to some 'source' material and reviewed R.N. Elliott's take on the types of corrections to refresh my memory at least one key point; namely, does the second downswing in a bearish correction have to carry farther than the first down leg?

The answer to this question is NO. This point and some other chart/technical analysis considerations suggest that the major indexes won't see lower lows than occurred this past Friday.

I'm reprinting a Subscriber Mail of possible general interest.


I wanted to write and say thanks for the info (on the nature of A-B-C corrections). You were spot on and give you all the props. I watched us bounce up last week on monday and tuesday on weak volume compared to the prior week's sell-off (short covering and weak rebound attempt). We stalled right at resistance and I went short at the close of Wednesday (2/3).(a little aggressive entry for me as I feel I got in the move early as opposed to waiting to see some downside pressure). I bought puts on MEE, CLF, VIP and closed out 50% of my position at the prior week's low and held the rest for an extended target as I believe the selling pressure will continue until at least early next week (2/8-2/9).

What you write is appreciated and used in conjunction with my own homework. Also if you can share a little bit on where I can read more on the "rule of alternation" that would be great. --student of trading and aspiring full-time trader--


"Selling pressure continuing into early next week"...right on the mark (as a prediction from 2/4) as the market made at least a short-term bottom on Monday-Tuesday (2/8-2/9).

To learn more about Elliott's 'Rule of Alternation' in depth I'd check out one OR both of the following two books, which are excellent companion pieces and not huge tomes. As far as Elliott Wave basics these are the only 2 books that I have used:

1. Frost and Prechter: "Elliott Wave Principle, Key to Stock Market Profits". I have the 5th Edition, first published in 1978. There is now a 20th Edition, published in 1998. I ought to update mine, as it's starting to fall apart!

2. Another 'classic' is: "The Major Works of R.N. Elliott", which has a foreword by Robert Prechter. Both books are relatively easy reads.

I don't try to use the more esoteric stuff. I knew Bob (Precter) in New York. We were both members of the Market Technicians Association. He also used to work at Merrill (in better days!). He's fallen out of favor as a market analyst due to some outlandish bear market predictions he's made in the past. If I had to chose one or the other, I'd go with Elliott's own writings (book #2).

I don't use Wave theory in the way the some do, trying to classify every market move in terms of its wave count. When I see an obvious 5-part advance, I find it useful. When I see an obvious 3 part A-B-C correction, it's of interest. Figuring out which wave segment we're in can help predict where prices go next. R.N. Elliott's 'Major Works' is very good. Although he goes more off on some theoretical tangents, Precter's book also lays out the basic wave principles, brings major market analysis closer to now and includes basics on Fibonacci sequences.


As noted, I did a little refresher read and here's what R.N. Eliott had to say on the principle of 'alternation'. "Alternation is a rule of Nature" and he sites a number of examples (e.g., leaves and branches first appearing on one side of a main stem and then on the other). However, the main part of the discussion is on the fact or habit of alternation in human activity. Bull and Bear Markets alternate. "A bull market consists of five waves and a bear market of three waves. The five and the three alternate." Waves 2 and 4 are corrective. If an initial correction in a bull market sequence is "simple", the second correction will be "complex" or of a different type.

Elliott identified 3 types of corrective patterns, the first two having down segments labeled "a", followed by an upswing 'b', which is in turn followed by a second downswing 'c' which completes a correction. The 3rd type of correction he identified is a 'triangle' pattern, is less common and I won't go into it here.

To best show the Zigzag versus Flat type a-b-c corrections I've labeled examples on the S&P 500 (SPX) daily chart. The 'zigzag' pattern is one where the down-up-down segments are of a different length and the pattern obviously fits what we think of as a zigzag in both instances. I am assuming that the 'c' down leg of the most recent a-b-c zigzag corrective pattern has completed itself. Stay tuned on that!

In a zigzag, leg 'c' MAY be longer than the initial decline 'a', but there's no rule that says it should be this way. In an a-b-c correction within an overall bull market, leg 'c' is often NOT longer than 'a'. In a bear market, down leg 'C' is often longer than down leg 'A'.

The a-b-c correction labeled a FLAT above is of the type where down-up-down segments are 'flat' or sideways and they don't typically slant or slant much. Moreover, the flat correction is where we often see EQUAL, or nearly equal, segments in terms of how far each price swing carried. Here is where we might see a double bottom set up.

An ALTERNATING (Rule of Alternation) pattern seen above is that an a-b-c zigzag correction was followed by an a-b-c flat correction, which in turn was followed by a next correction that had the zigzag pattern.


While it may or may not be true that the second down leg of an a-b-c correction or wave turns out to be LONGER (in point terms) than the second decline ('c'), I found myself wondering if it would actually turn out that way. However, there were more technical considerations than just the length of the price swing of the second decline in analyzing IF a low was in place for our recent correction. If I was a 'pure' Elliott wave technician, I would be mostly focused on explaining and understanding the wave 'count' and whether the last downswing 'fit' my interpretation for where the end of the correction should be. I don't go there so much.

An important consideration for me as to a low being in place for the recent correction or not, is suggested by a key trendline. One that appeared to mark support on the weekly S&P 500 (SPX) chart follows another rule of thumb related to trendlines; i.e., a resistance trendline, once pierced, tends to 'become' support on subsequent pullbacks.

While it's also true that progressively lower lows made in a decline might just continue to 'hug' the (downward) slope of the SPX trendline seen above, this is not usually the case. More commonly, within 1-3 bars (time periods) where the lows pull back to the trendline, a recovery rally begins.

A SECOND key consideration in my mind regarding whether a low is now in place, or yet to come, for the correction, was whether a key index that sells off stops after completion of at least a fibonacci 62 percent correction of the prior advance. Besides the 38 and 50% retracement levels, I focus on instances of a fibonacci 61.8 percent correction or a little bit more as marking the possible end of a correction. That 'little bit' more that I see again and again is when a 66% correction is completed; something WD Gann would have understood as he also pointed to the importance of the 1/3 and 2/3rd retracements.

We have the pivotal Nasdaq 100 (NDX) having completed an exact 66% retracement, followed by buying coming in. Although there was some subsequent slippage (running stops) below this level in the next decline (as seen on the chart below), the rebound that has followed looks like the real deal. It will also be important for NDX to clear resistance at 1788.