From Zigzag To Triangle?

02/25/05 08:36:59 AM

by David Penn

Using Elliott wave theory to game the correction of January/February 2005.

Security: $SPX
Position: N/A

My most recent Elliott wave-oriented article ("More Elliott Wave Work," Working-Money.com, February 16, 2005) talked about the correction in the market that began precisely as 2005 began. In that piece, I suggested that the January/February correction might turn out to be a zigzag type of correction, which would take the Standard & Poor's 500 down to a test of the 1140 level (roughly the level of the September 2004 highs).

The litmus test of the zigzag interpretation was whether or not the "b" wave of the correction--a wave that began in late January and appeared to roll over in mid-February--retraced more than 61.8% of the drop from the late December 2004 highs. If the "b" wave failed to do this, then the zigzag interpretation would remain valid. If the "b" wave exceeded a 61.8% retracement, then the zigzag interpretation would be untenable.

As it turned out, the "b" wave rally well exceeded the 61.8% retracement level in February.

So if the zigzag interpretation is out, what kind of correction are we seeing in the S&P 500 in 2005?

Part of the reason I suspected that the January/February correction would be a zigzag is because it is the other, common corrective pattern in Elliott wave analysis. In addition, because it appears to me that wave (2) from mid-September to late October in 2004 was a flat-type of correction, the rule of alternation warns against seeing the next correction as a flat-type also. Given the sharpness of the correction in January--in addition to the other factors mentioned--I thought a zigzag would be a good guess as to how the action in the S&P 500 would unfold as the correction played itself out.

Figure 1: The rule of alternation suggests that if the January/February correction is not a zigzag, it quite possibly might turn out to be a triangle.

Graphic provided by: Prophet Financial, Inc.

Recent price action, however, means that some other sort of nonflat, non-zigzag correction must be considered. Chief among the alternatives, then, is the triangle. The Elliott wave triangle resembles the sort of triangles that technical analysts have been studying and trading for years. Elliott wave theorists tend to look at two different types of triangles: contracting triangles, which consist of converging trendline boundaries, and expanding triangles, which consist of diverging trendline boundaries. Not only that, EW triangles have five subwaves--all of which are corrective and consist of three components or segments.

What additional rules govern EW triangles? Most important, the first of the five subwaves should be the longest and the fifth subwave the shortest. So far, this is the case in the January/February correction, with the decline from late December to late January being longer than the rally from late January to mid-February. There are three more subwaves to go, if in fact this action ends up being an EW contracting triangle. But so far, so good.

Interestingly, triangles are said to only develop in "b" or fourth waves--never in "a" or second waves. Insofar as I believe that the late December 2004 peak represented a wave (3) top, the appearance of a triangle at this point during the subsequent correction would not be surprising at all.

A few caveats, however. If the January/February correction turns out to be both a fourth-wave correction and a triangle correction, then traders should be on guard for a powerful rally once the correction has run its course. "Powerful"? In a previous Elliott wave article for Working-Money.com ("Elliott Wave 2005," January 12, 2005), I made the argument for an end to the 2002-05 bull market (a Cycle B advance, by my reckoning) in the S&P 500 somewhere short of 1260. The argument for a fourth-wave, triangular correction right here in the S&P 500 is consistent with that argument.

How will we know if the triangle thesis is invalid? The first clue will be whether the current decline -- which began in mid-February and can be considered a wave "c" within the triangle -- takes out the January 2005 lows. If it fails to do that, then the triangle argument will remain standing. If the current decline does take out the January lows, then a number of other possibilities -- including the possibility that the 2002-05 bull market ended in late December 2004 and that what I am calling a wave (3) top in December and a wave (4) correction in January/February is really a Cycle B top in December and, in January/February, the first few waves of what will likely be a nasty ride down in a multiyear, Cycle C bear market.

[Personally, I have a funny feeling about the 1180 level. If the triangle turns out to be the correction, then it wouldn't surprise me for a nanosecond if the final wave (the "e" wave) of that triangle found support at or near 1180. If that's the case, then it will only be a matter of seeing how close to 1260 the fifth wave in this last leg of the 2002-05 bull market the S&P 500 will get.]

David Penn

Technical Writer for Technical Analysis of STOCKS & COMMODITIES magazine, Working-Money.com, and Traders.com Advantage.

Title:	Technical Writer
Company:	Technical Analysis, Inc.
Address:	4757 California Avenue SW
	Seattle, WA 98116
Phone # for sales:	206 938 0570
Fax:	206 938 1307
Website:	www.Traders.com
E-mail address:	DPenn@traders.com
Traders' Resource Links
Charting the Stock Market: The Wyckoff Method -- Books
Working-Money.com -- Online Trading Services
Traders.com Advantage -- Online Trading Services
Technical Analysis of Stocks & Commodities -- Publications and Newsletters
Working Money, at Working-Money.com -- Publications and Newsletters
Traders.com Advantage -- Publications and Newsletters
Professional Traders Starter Kit -- Software