Trader’s Narrative

A few months ago I started to pay special attention to a remarkable but little followed technical indicator called the Coppock Curve or Guide. Peering ahead and assuming that the market would hold firm, I mentioned that by the end of May we would have a definitive buy signal. The market’s strength was impressive, as we all know and the signal did arrive for a new bull market.

Although of course, it remains to be proven beyond a shadow of a doubt that we have seen the low for this cycle, it looks to have been a valid signal. Returning to our hypothetical forward extrapolation of index levels, the Coppock Curve continues to act very bullish as it continues to rise out of the depths it had fallen to:

Source: InvesTech Research

If we assume that the S&P 500 index will close approximately where it is now by the end of August, then the Coppock Curve would continue to recover extremely fast, reaching -344. That is well above its February 2009 level and closing in on its January level.

That’s assuming that the market is unchanged until month’s end. But what if the market trips up from now until then?

In that scenario, the Coppock Curve still has a good chance of continuing to rise. In fact, the S&P 500 index would have to fall more than 200 points, below 780, for the Coppock Curve to stop climbing. That’s a hypothetical +20% decline.

And that is probably one of the positive consequences of this spring rally. Even if we were to give back 50% or even 65% of the advance, the market would still be able to carve out a higher low and a higher high - the very definition of an uptrend. Such a correction wouldn’t be surprising, especially when you consider the over confidence signaled by the various sentiment measures. But it would only wash out the weak hands and allow the market to continue higher.

While the track record of the Coppock Curve is impressive it is not perfect. If you squint hard enough, you’ll be able to make out the rare two times that its upturn did not mark a significant trend change.

But even so, since many are comparing this bear market to the one that came 80 years ago, it should be noted that the Coppock Guide was one of few technical indicators which allowed for the correct navigation of the bone crushing volatility of the 1929 bear market. During the aftermath there were several intense bear market rallies that fooled many. But they were all ignored by the Coppock Guide as it fell unrelentingly into the deepest level it has ever seen historically. Only in 1932 did it correctly give the all clear.

The Coppock Guide. Just a while ago no one knew it even existed. I’m willing to bet even if you mentioned it in a room full of TA buffs, you would get quizzical looks all around.

But now suddenly it is the belle of the ball! Everyone is talking about it. There is a buzz in the blogs, Bloomberg, MarketWatch (WSJ) have stories on it and even CNBC’s Fast Money, which normally has the attention span of a fruit fly gives it a mention (see below for video). To be honest this is exactly what I was afraid of when I first hesitatingly brought it to your attention in Conditions of a Bull Market. I’d rather it remain the esoteric indicator it used to be (hopefully everyone will forget all about it).

The question remains: is it useful?

I tend to think so but some disagree.

Mark Hulbert looks back all the way to 1896 and calculates the success of Coppock Guide signals on the Dow (see article). His conclusion is that the returns are not indicative of an edge.

However, what Hulbert doesn’t point out is that although the Coppock Guide for the Dow Jones Industrial may be a poor indicator, that doesn’t mean that it is for other, more respectable indexes like the S&P 500. What Hulbert does point out in a subsequent article is that even for the Dow Coppock numbers, the two false signals in the 1930’s skew the results and if we look at more recent data, the indicator does have an edge. So the question is how relevant is the 1930’s to the current market?

Guy’s analysis at the Technial Take also falls into the same trap as Hulbert by looking at the Dow Jones Industrial. I don’t want to rehash why the Dow is inferior to the S&P 500 (price weighted vs. capitalization weighing and small sample size of 30 vs. 500). The only time I give the Dow more respect is in the Dow Theory analysis of the market.

On the other side, there are those that see the Coppock Curve as a valuable guide:

James Stack of InvesTech is one of them. Formerly an engineer by training (who worked for IBM) Stack’s whole approach is a blend of quantifiable edges (whether technical or fundamental) and sentiment analysis. But he doesn’t rely on the Coppock exclusively.

Another is Steve Leuthold of Leuthold Weeden Capital Management. He calculates a similar indicator that he calls the VLT Momentum (”very long term”). I consider Leuthold as one of the ‘grey beards’ who get my respect for their successful navigation of the market over decades. Recently Leuthold has turned even more bullish (after nailing the March low).

Next is Jason Goepfert of SentimenTrader. Recently Jason did a complete analysis of the Coppock signals (using the S&P 500) and showed that it has a definitive edge, especially on a risk adjusted basis. To see the data get a 14 day free trial and take a look.

In the end, the headline is rhetorical because I’m not really going to defend this or any other indicator.

First of all, it is just that, an indicator, with all the inherent flaws and limitations. I don’t expect anyone to exclusively trade off it. In fact I think that would be nuts.

Second, one has to just look at the data. How you want to take advantage of the Coppock Curve, if at all, is really up to you. Above you’ll find enough information to make up your own mind. But treat it as a starting place and do your own research. Then drop me a note - whether positive or negative.

Finally, the Coppock Curve is just one of the conditions for a new bull market - I’ll cover the others soon. I pay much more attention to the weight of indicators, rather than just any one in particular, no matter how impressive its historical performance may be.

Here’s an incredibly shallow analysis of the Coppock Guide (what else did you expect from CNBC?):

After reaching an all time low in January 2009, the Conference Board Consumer Confidence Index reached an all time low (again) in February 2009.

Although we haven’t seen such a dramatic fall in this metric for 42 years, historically, such troughs are signs of the coming of better times in the stock market. In fact, along with several other variables, gloomy consumer sentiment is one of the handful of conditions of a new bull market.

Now we are seeing an almost equally forceful recovery with the latest Consumer Index data from the Conference Board:

The last time we saw this kind of jump in the Consumer Confidence was in early 2003 - coinciding with the bear market’s end. The stock market certainly cheered when the surprisingly higher number was released by the Conference Board on Tuesday. And it is confirmed by other sources like the Gallup poll of US consumers showing a marked improvement since February 2009:

Source: Gallup Consumer Poll

So the question is, is such a sharp recovery in consumer optimism bullish for the market? or bearish?

In a recent column, Mark Hulbert argues that it is bearish. He compared the index with subsequent stock market returns (month, quarter, year, and 2 year periods). The relationship Hulbert found was an inverse one, where falls in Consumer Confidence lead to rallies in the market (and recoveries in Confidence lead to lower returns in the stock market). Given the contrarian nature of the crowd, this isn’t surprising.

But it isn’t the whole story.

James Stack, editor of the InvesTech newsletter and Jason Goepfert of SentimenTrader both focus on something entirely different. The Conference Board releases several different sub-sets of data. The one most people concentrate on is the Present Situations number but there is also the Expectations Index, which asks the respondents to look ahead to 6 months in the future:

They both keep track of another variable: the Expectations minus the Present Situation Index - in other words, the difference between how consumers think they will fare in the future and how they are doing right now. And interestingly enough, this net variable is actually positively correlated to the stock market - and a leading indicator.

I highly recommend both sources. To see the net Confidence chart take up their free trial offers:

You can get a free 14 day free trial of SentimenTrader.

And you can get a free sample of InvesTech’s newest report here.