Trader’s Narrative

Who could have imagined that with crude oil at +$110, the transports would be one of the highest relative strength sectors out there?

Well, if the stock market was strictly logical we would have figured it all out centuries ago. This is exactly what makes trading so fascinating.

Bullish Percent
The recent swing low was in early January, when the bullish percent for the sector reached a measly 5%. Five percent!

Do you know the last time they were that low? Try July 2002.

But since you’ve read my post about timing the market with bullish percent charts, you know all about that and of course, took obscene advantage of it. Of course.

But there’s more going on here. Take a look at this chart:

The top chart shows the relative strength of the transports (to the S&P 500). Notice the higher lows and the higher highs. Then check out the completed head and shoulder formation with a nice quick retest of the neckline.

Believe it or not, it has already made it to the October 2007 highs. In contrast the Dow Jones Industrial average is still well below that area on its chart.

Dow Theory
According to Dow Theory, major signals are given when the two major sectors (sometimes along with the third: utilities) confirm each other.

While the recent action is not bullish per se, at least according to strict Dow Theory, it sets up what is called a “non-confirmation” - in this case, for a decline. That is, because the Dow Transports didn’t confirm the lows that the Dow Jones Industrial Average reached but instead headed up.

What would be truly bearish, according to Dow Theory, is if both the Dow Jones and the Transports print prices lower than their January levels.

So right now what we have is simply the potential for a buy signal, if the Dow Jones continues to rally and rises above its February highs.

This article from today’s Wall Street Journal features some prominent technical analysts (Paul Desmond and Steve Leuthold) opining on whether this is it for the bull market that started in 2002:

Last week’s stock-market carnage — the Dow Jones Industrial Average fell 4.23% for the week to 13265.47 — seemed an overreaction to most analysts, who focus on fundamentals like corporate profits and interest rates. The global economy continues surging, they point out, while most market interest rates remain low by historical standards. What matters, they say, is whether the credit crunch, caused by ill-conceived loans to home buyers and businesses, starts to interfere with growth and interest rates.

But another, less fashionable, breed of analysts sees storm warnings. Known as technical analysts or chartists, because they plot and compare a wide range of sometimes-arcane market data on graph paper and spreadsheets, they liken their work to hurricane tracking. They can see a pattern building, they say; the trick is distinguishing a brutal Category 5 storm from a less-severe Category 2.

Like hurricanes, market tops tend to have a lot of common features, these analysts say. Whether a bear market features a crash, as in 1987, or is marked by a long, painful decline, as in 2000, the end of a bull market usually sends the same signals.
A top commonly looks like a roof, with indexes bumping up against it until they sag. Indexes hit a series of new highs, but over time fewer and fewer stocks join in. Big multinationals gradually take over leadership from the smaller, more-volatile stocks that typically lead a bull market’s early stages.

Sentiment turns from fear to greed as investors push stocks too far. Stocks top out one by one, often with a small band of highly admired issues leading the gains, until there isn’t enough strength left to hold the indexes up. This usually happens when interest rates are rising, pressuring businesses and consumers. Money managers, reluctant to sell for fear of missing out on more gains, get caught in the declines.

The market looked a lot like this before its peaks in 1987 and 2000, and at the end of other bull markets. Today, it has some, but not all, of these characteristics. Hence the fierce debate: Is this a hurricane or just a summer downpour?

Finding an answer is tricky. Tops aren’t just the mirror image of a bottom. When stocks rebound from a bottom, as they did in 2002, they tend to surge. But when they fall from a top, they waver uncertainly, giving the appearance of a pause. The last time stocks topped out, in 2000, speculative technology stocks fell hard. But the Dow industrials declined very gradually, as did the Standard & Poor’s 500-stock index and even the large tech stocks. It took months for analysts to agree a broad bear market was really at hand.

There have been some signs of a roof forming lately. Markets have seen a series of records, with big stocks beginning to lead the way and fewer stocks showing gains. When the Dow industrials hit their record just above 14000 on July 19, many second-tier stocks didn’t join them; indeed, after a strong start, small stocks are down for 2007. Meanwhile, money managers who were skeptical for much of the past year showed signs of greed, setting aside doubts and jumping into the market.

At the same time, until last week, middle-size stocks had been holding up better than small ones, and the gradual topping out hadn’t gotten very far. Financial, consumer, telecommunications and health-care stocks, as well as real-estate stocks and utilities, all had turned down, but technology, energy, basic materials and industrial stocks all were holding up well. Market interest rates had risen, but not heavily.

Moreover, even after last week, the Dow’s worst in more than four years, the index remains up 6.4% in 2007 and 18.2% in the past 52 weeks.

Whether technical analysis is really useful at making sense of such data is a matter of some dispute on Wall Street. Some investors believe it is impossible to forecast the market’s ups and downs. Academic studies have shown that when most people, professionals and amateurs alike, try to move money in and out of stocks to beat market fluctuations, they tend to wind up with losses.

That helps explain why, in this era of passionate investor interest in corporate profit news and Federal Reserve interest-rate decisions, the hard-to-explain work of technical analysis has fallen into disfavor. Some brokerage firms have eliminated their technical research departments altogether. Still, when markets begin to sag, investors rediscover technical analysts.

Thanks in part to technical analysis, Steve Leuthold, who manages $3 billion as chairman of Leuthold Weeden Research in Minneapolis, stunned clients early last week by warning of impending trouble. Two days later came the Dow’s combined plunge of 3.8%, or 519.60 points, on Thursday and Friday.

Mr. Leuthold sees a bear market at hand — that is, a decline of 20% or more from the top. “I don’t think this is going to be the end of the world, but it could be a normal bear market that could go on for a while,” says Mr. Leuthold, who has been working in cut-off jeans and a T-shirt from his summer home near Portland, Maine.

Paul Desmond, president of Lowry’s Reports in North Palm Beach, Fla., interprets the data differently. “If a person’s hair has turned gray, it doesn’t mean they are ready to pass on,” says Mr. Desmond. The market has caught cold, he says, but “isn’t showing the signs you would normally see if it were near death.” He expects a rally in the next few days, and has advised clients that if it is broad and robust, they should begin buying again in anticipation that the market will recover.

The mixed signals have been frustrating for Phil Roth, chief technical market analyst at New York brokerage firm Miller Tabak + Co. Mr. Roth spends an hour or two each day updating 100 different charts by hand. He has been plotting his charts, filled with market indicators, on gray and green sheets of graph paper since he entered the business in 1966. He extends each chart’s length by carefully gluing on more graph paper; these days, they unfold like accordions.

Typically, he says, a bull market shows signs of age after three years or so, tops out and gives way to a bear market. This one seemed to be doing that when it sagged in the spring and summer of 2006. But it rebounded, then shook off similar setbacks in February and June to reach 14000.41 on July 19. Mr. Roth came to attribute its resilience in part to the newly ascendant hedge funds and other private-investment funds that were driving the market.

Of late he had taken note of accumulating warning signs. Even as the market rose, fewer stocks were continuing to reach new highs. The number of stocks trading above their average prices of the past 200 days was slipping. Utility and financial stocks were losing steam. The use of borrowed money for stock investments was hitting all-time highs.

Then, when the Dow crossed 14000 to hit its most recent record, Mr. Roth noted that many stocks in the broader market failed to rise. More trading was being done on falling prices, suggesting weakness.

On Thursday and Friday, his chart-plotting frequently interrupted by phone calls from anxious clients, Mr. Roth saw what he had been looking for: After three consecutive days of record trading volume, the Dow industrials, the S&P 500 and the Nasdaq Composite Index all broke below levels at which they had rebounded strongly in the past. Measures of selling pressure, such as trades on declining prices and the number of stocks that were falling, indicated panicky selling.

He now fears that because so many investors put aside doubts to jump into stocks, and may now be looking for a way out, stocks could experience “a climax in the market” — an even heavier dose of panic selling than investors have seen so far. He expects stocks to fall 10% — which they haven’t done since 2003 — then put in a weak rebound before falling further. And he doesn’t rule out the kind of collapse that occurred on October 19, 1987, when stocks fell 22.6% in one day, the biggest one-day decline ever. “I’m not saying it will happen again, but it could,” Mr. Roth says.

At Leuthold Weeden Research’s headquarters, senior research analyst Andy Engel, working with Mr. Leuthold, tracks more than 180 eclectic data points in heavy ring binders that look like Mr. Roth’s. They watch not only the kinds of market indicators Mr. Roth follows, but, unlike most technical analysts, also broad fundamental and economic measures such as inflation, money supply and corporate profits. They have long been concerned about weakening growth and the risk of higher inflation and interest rates.

Like Mr. Roth, the two worry that many measures of investor sentiment, such as polls and a limited use of options to hedge against declines, suggest investors are overly confident and riding for a fall. Companies also have been issuing more stock, suggesting insiders see this as a good time to sell. Demand for gold-mining stocks, sometimes a leading indicator of declining confidence in the economy, has been on the rise.

Mr. Engel’s and Mr. Leuthold’s indicators gave a false sell signal in the fall of 2005, which made them miss out on some gains. But their indicators were on the money near the 1987 top, giving accurate sell signals, and gave buy signals near the 2002 bear-market bottom. After being surprised by an initial sell signal two weeks ago, the two men waited a week, rechecked the data, then sent out a warning to research clients and cut their own stock exposure last Tuesday.

For his part, Mr. Desmond, of Lowry’s, says his skepticism that this is a bear market stems from extensive research on past market tops. A week ago he warned clients of the risk of a short-term decline, and he still thinks that is what this is. “We just don’t have the signs you normally have at a market top,” he says.

His research shows that, as the market forms its rooflike top, individual stocks slowly fall back despite the indexes’ continuing gains. At almost all market tops starting in 1929, when the indexes finally were ready to fall into bear markets, the vast majority of stocks already had turned down, and many had fallen sharply.

But at the most recent market highs, a number of stock groups such as basic materials, energy and technology still were strong. While small stocks have faded, middle-size stocks are holding up better.

The year 1987 featured a September drop similar to last week’s, followed by the October crash, Mr. Desmond says. That has made some of his clients worry. But in 1987, less-prominent stocks had been slowly deteriorating for months before that, and that process hasn’t advanced nearly as far this time. If this year is going to be an exception, with a bear market starting before this broad deterioration has taken place, Mr. Desmond says, it will become apparent in the coming days, as the bounce-back he expects would prove weak and disappointing.

Timing the stock market can’t be simpler than this.

You probably already know about the concept of ‘percent above moving average’ and how it can pin point inflection points. Here’s a great way to use that concept applied to a very, very narrow subset of the market: the Dow Jones Industrial Average.

Eventhough we’re only talking about 30 stocks out of the thousands traded on stock markets in the US, this group of stocks is so important that their behaviour can accurately reflect the mood of the whole market. And calculating this is so easy, you can do it yourself on the back of a scrap piece of paper. Just scan through the 30 stocks and see how many are above/below their 50 day moving average. Take those that are above and divide by 30. If the number is low (30% and lower) you should not be pressing a short bias or position in the market. If you do, chances are that you’ll get whacked in the face by a snap back rally.

In fact, the lower the percent, the more you should be covering shorts (if any) and thinking about going long. Take a look at this technical indicator and its predictive power:

As you can see, it has a very good track record. It caught every single important inflection point (including the April 2005 bounce which I forgot to annotate). The only exception I can find is the October 2005 example. But even then, it almost kissed the 25% line.

Think of price as stretching against the moving average like a dog pulling away from its handler. In the short term, a dog can pull away from the intended direction of its master but it is only a matter of time that it comes back and rejoins her. In the market, the ‘master’ is the major direction of the market. As long as we know the major trend, we can easier identify short term fluctuations around it.

An important caveat is to not try and flip this around to find tops. Usually tops are notoriously difficult to sniff out for any indicator. So that isn’t a particular knock on this one. It totally missed the May 2006 top and it mis-identified September - November 2006 as a top when in fact the index was on a beautiful trend upwards.