Eye of the beholder

I would like to talk a bit about chart patterns and how usefull they can be (or not).

The last months, we're seeing head-and-shoulder patterns in the indices on a regular basis, so it seems.

Let me start with the first one (depicted below). I'll use the NASDAQ-index, but it just as well could've been the S&P-chart. Here we notice an inverse head-and-shoulders pattern formed from February till April.

Technically, the measured move out of the neckline is expected to be at least the same amount of points as the difference between the top/bottom of the head and the neckline. For instance, here we have about a 200 point move from head to neckline. You would expect the target to be 3000 for this chart. However, the move stalled just shy of 2900.

Next chart is the period from April to June. Here we see a HS-pattern that measured about 150 points from head to neckline. The target was 2600 which was met (just).

Third one, from May till July (not completed yet). Here we see a 275 point move from head to neckline. That would give a target (if and when neckline is broken) of approximately 3150.

I'm not really a strong believer in these kind of patterns. The bigger/complexer they are, the greater the chance they will fail. It's all in the eye of the beholder. The last HS-pattern might just as well be seen as a cup-and-handle, still to be completed.

All in all, I put more trust in trendlines and bull- or bearflags. It's more precise and combined with price action and volume, results will be better and you'll need less patience compared to a HS-pattern before it's completed (which can take months).

A nice example here, BIDU. The breakouts were accompanied with good volume and gave results of 10%-20%. 

Some more thoughts on chart patterns; it reminded me of this older post from Peter Brandt, an interesting read.