The Limits of Stock Chart Reading

Any time people are compelled to decipher the future, strange methods and theories are sure to abound. The stock market has long been a haven of such folly.  Among the diviners in that arena, there are believers in the notion that planetary movements affect share values, that stock prices move in predictable wave sequences, and that certain geometric patterns on stock charts presage a change in trend. Nor are these habits of thought restricted to tiny corners of the stock exchange. Such is the eagerness to gain a clue into the future that there’ll always be numerous takers for far-fetched prognostications.

Scary Parallel

Source: http://www.marketwatch.com/story/scary-1929-market-chart-gains-traction-2014-02-11

The latest example of this is a chart (see above) that is getting wide dissemination on Wall Street. The chart depicts two prices series, one of the Dow Jones Industrial Average between 1928-1929 and the other of that same index from mid-2012 to the present day. The two lines are strikingly similar in their undulations. Indeed, since the resemblance first caught people’s attention this past November, the correlation has persisted. What this is supposed to portend, of course, is a crash along the lines of October 1929.

Yet this presumes that patterns from the past can be reliably expected to recur in the future. It is, as Ludwig von Mises might have put it, to assume that there are constant relations in economic life — that the fact that events of type B have previously followed events of type A means that B will recur whenever A happens to arise. But there are no constant relations in human affairs. For, unlike natural objects, human beings are continually exposed to novel  experiences, from which they learn and orient their actions accordingly in unforeseeable ways.

Even the original purveyor of the above chart, Tom McClellan (publisher of the McClellan Market Report), concedes that: “Every pattern analog I have ever studied breaks correlation eventually, and often at the point when I am most counting on it to continue working”. Undaunted by this realization that the past is no certain guide to the future, he nevertheless persists in warning us to be wary about the market.

The only historical pattern with any semblance of predictive significance is the proclivity of the stock market to trend in the same direction for a significant period of time. These trends are commonly known as bull and bear markets. Why these exist is actually something of a puzzle. Stock prices, being time-discounted estimates of future company dividends, ought to gyrate randomly in response to new information. To the extent one ought to expect a trend, it should be a very gently rising one mirroring the long-term rate of economic growth.  The reason why this does not occur, however,  is that the central bank generates booms and busts with its monetary policies, booms and busts that the stock market ends up reflecting in bull and bear markets.

All that can be usefully gleaned, then, from a stock chart is the prevailing trend. To gauge that one need not draw precise historical parallels with past price movements. A simple moving average — like a 10 month — might do.  In other words, if a major index like the S&P 500 is above its 10 month moving average, the trend is up. Conversely, if the index is below the average, the trend is down. Even then, there is no guarantee that the indicated trend will continue for any specific amount of time.

2 Responses to “The Limits of Stock Chart Reading”

  1. Joe Walker says:

    Thank you for this article Tomas. I once traded gold on the floor of COMEX and always looked at charts to "see" the future, it can not be done and I learned that the hard way. You can guess and be right, but to think it is because of a "head and shoulders top" pattern or my favorite a "rounding bottom" pattern is silly.

    That said, I do not know when, but gold will skyrocket one day, and not because of a "falling triangle" but because of blatant negligence of our Fed overlords.

  2. BuySmart says:

    I always hate when people try to prove that the market is going to fall using charts from the past. In reality, the time to worry is when nobody else is worried. None the less, I play it safe using a dividend growth strategy, which tends to outperform through market noise.

Leave a Reply

You must be logged in to post a comment.