From the May 01, 2009 issue of Futures Magazine • Subscribe!

The nature and origins of trend following

Although trend following has been a popular trading philosophy for many years, surprisingly little has been written about its origins and history. This is partly due, no doubt, to the scarcity of available information prior to the early 20th century, and because until about 50 years ago trend following as a philosophy had not been completely articulated. By the early 1950s, many trend following methodologies were in common use — and may have been for centuries — but the underlying concept had not been fully defined or even given a name. However, there are things to be revealed by looking back into the history of trend following.

One of the earliest individuals to contemplate trend following was the economist and trader David Ricardo, who flourished in the London markets from the 1790s until about 1818. A large trader in Consols (bonds) and stocks, he accumulated a large fortune from his speculations. Exactly what his methodologies were is not known, but he is credited with one of the most famous sayings in trading: “Cut short your losses; let your profits run on.”

This is good advice, no doubt as it has survived to the present day. Still, although it clearly expresses two central tenets of trend following, it lacks detail. It does not state exactly when to cut a loss or how long to let a profit run.

Move forward a century and west a continent, and you have the following wisdom from the famous Chicago grain trader Arthur W. Cutten: “Most of my success has been due to my hanging on while my profits mounted.” Again, Cutten is saying, “stay with the trend.”

For a third example, Jesse Livermore, a central figure in the history of trend following, stated: “…the big money [is] not in the individual fluctuations but in the main movements — that is, not in reading the tape but in sizing up the entire market and its trend.”

This last quote is from Edwin Lefèvre’s “Reminiscences of a Stock Operator,” a series of articles from the Saturday Evening Post in 1922-23, reprinted many times. Although the speaker is stated to be Larry Livingston, it is generally agreed that Lefèvre’s interviewee was Jesse Livermore. The quotation is Livermore’s interpretation of an oft-repeated statement made by a brokerage-house acquaintance, that “It’s a bull market, you know.” This advice was given whenever some trader was tempted to liquidate a winning position too soon.

The advice from these traders is not about initiations or liquidations; rather, it is about when not to liquidate. The counsel is good, but not complete. Indeed, if we can systematically define our initiations and liquidations, the advice “stay with the trend” becomes irrelevant. It happens automatically. So the question becomes, when did systematic trend following begin?

Unfortunately, we cannot know for sure, so let us ask instead: when did systematic trend following begin to be publicly formulated?

The answer goes back to the end of the 19th century and the development of Dow theory, based on concepts originated by Charles H. Dow in a series of articles in the Wall Street Journal between 1899 and 1902, expanded upon by William Hamilton between 1903 and 1929, and refined by Robert Rhea in 1932. Inasmuch as Dow theory defines a bull market as a series of higher highs, and a bear market as a series of lower lows, the rudiments of a trend following strategy become apparent: one buys on the breakout of an old high, and sells on the breakout of an old low. Of course, there are more rules dealing with confirmation and volume, but these are the basics.

Dow theory is, I believe, the earliest modern expression of an objective trend following system. It defines precisely — as long as a trader can determine a meaningful high or low — the entry and exit levels for trend following trades. Further, the methodology can be generalized and parameterized: different levels of breakouts can be used, moving averages of prices can be used, etc. Dow theory is certainly the grandfather of trend following methodologies.

Dow theory is not very mathematical; rather, it makes logical observations about current and past prices to determine the direction of the market. This innumeracy is not surprising, as the theory was developed long before the advent of the computer. Nor is it surprising that the earliest offshoots of Dow theory continued in this observational, structural mode of analysis. Robert Prechter, for example, states that R. N. Elliott developed his wave methodology through contemplation of Dow theory. Richard W. Schabacker, Robert D. Edwards, and John Magee also recognized Dow theory as seminal to their thinking, with Edwards and Magee’s book “Technical Analysis of Stock Trends,” for example, devoting three chapters to the subject.

Indeed, “Technical Analysis of Stock Trends,” first printed in 1948, as well as its predecessors, Schabacker’s “Technical Analysis and Stock Market Profits,” from 1932, and “Profits in the Stock Market,” by Harold M. Gartley in 1935, are milestones in the development of trend following methodology. Given the focus in these books on technical patterns such as flags, pennants, triangles, head-and-shoulders patterns, it may seem peculiar to associate these books with trend following, but the point behind being able to distinguish such patterns is precisely to recognize signals for trends (see “Building blocks,” above). Edwards noted, “Profits are made by capitalizing on up or down trends, by following them until they are reversed.”

Aha! There we see the terms “trends” and “following” separated by only the space of a word.

A basic trend following tool of Schabacker, Gartley, Edwards and Magee’s was the trendline, a line connecting the “basing points,” or minor tops or bottoms (in terms of Dow theory), with each other. Buying and selling signals were based (with further conditions, exceptions) on the “breaking” of these trendlines, a modification of the more basic Dow idea of a breakout from the price level defined by the top or bottom itself (see “Ride this trend” ). From Schabacker: “So trendlines serve a double purpose.

While they last they define the line of continuation in a movement, and when they are broken they serve notice of a probable reversal and advise us to forget the old lines and start searching for new ones.”

Now we must ask, to what extent were Schabacker, Edwards and Magee innovators of these strategies, and to what extent were they merely describing methodologies in use among traders of their time? A clue is revealed by Edwards, “Your experienced technician, in fact, is constantly drawing trendlines of all sorts,” implying that trendlines had been used by other traders.

Three other books should be mentioned at this point that predate the above mentioned ones. The first book, published in 1928, was “Stock Movements and Speculation” by Frederic Drew Bond. Like other writers of his time, Bond wrote about basic price patterns such as double tops and bottoms, and repeatedly discussed the nature of trends, and how “…the public do not make the trend of the market. They follow it.”

Who, then, makes trends? Well, that would be “plainly the great industrial, commercial and financial capitalists of America, who correctly anticipate the future at least in a general broad way,” Bond says. In other words, trends are created by those who are most directly aware of the fundamental factors influencing the economy.

The second book, by William D. Gann, entitled “Truth of the Stock Tape,” published in 1923, also emphasizes the trend: “The way to make money is to determine the trend and then follow it.” Indeed, the focus in this book, and other works of Gann’s that followed, was, in one way or another, to take trades in the direction of the market’s trend.

The third book was Richard D. Wyckoff’s “Studies in Tape Reading,” published in 1910. Significantly, Wyckoff uses the term “follow the trend” when discussing one Jacob Field, “Prince of the Floor Traders,” but he does not follow through. A few years later, however, Wyckoff was more decisively on the side of trend following, actually publishing a newsletter entitled the Trend Letter. In a similar vein, in a later work, Wyckoff uses an interesting metaphor for trend following: “A small trader should be a hitch-hiker.”

But with respect to using charts as a means for profitably trading the markets, Wyckoff was skeptical:

“Let anyone who thinks he can make money following a Figure Chart or any other kind of chart have a friend prepare it, keeping secret the name of the stock and the period covered. Then put down on paper a positive set of rules which are to be strictly adhered to, ...Record each order and execution just as if actually trading. Put Rollo Tape down as coppering every trade and when done send him a check for what you have lost.”

Wyckoff’s quote suggests that those who use charts must, at least subconsciously, be making a discretionary decision based on what they know of the underlying market and would undoubtedly fail if making decisions simply based on the price chart. Contrast that to today’s systematic trend followers who use charts partly to avoid letting any underlying psychological bias creep into their approach.

The methodology suggested by Wyckoff can now routinely be done by computer. Would his judgment have changed? Possibly, since many years later, Wyckoff was using charts to draw trendlines, or as he called them, supply lines and demand lines.


The early Livermore can be considered to have been at least partially a trend follower inasmuch as he began his trading program in a small way at first, only “adding to his line” if the market went in his direction, and abandoning it otherwise. He also was apparently a “breakout” trader, at one point describing a stock that was bouncing back and forth between two price levels, but observing that eventually either buying or selling would become stronger, and “the price will break through the old barrier.” This breakout, then, would define the “line of least resistance.” Later, he says, “Well, when the price line of least resistance is established I follow it.”

By 1940, Livermore was more of a trend follower. In “How to Trade in Stocks,” he advocated the use of specific buy and sell signals based on his analysis of the perceived trend: “When I see by my records that an upward trend is in progress, I become a buyer as soon as a stock makes a new high on its movement, ...Why? Because I am following the trend at the time. My records signal me to go ahead!”

Central to Livermore’s philosophy was the recording of “pivotal points,” or intermediate highs and lows. These pivotal points were in part the same thing as Dow theory intermediate highs and lows, or Edwards and Magee’s basing points. Initiation and liquidation signals were based on significant movement away from these pivotal points. Thus, Livermore’s formula was not a breakout system, nor a trendline system, but rather a type of “filter rule,” though a bit more complicated than the typically tested sort. The parameters used were arbitrary, but based on much experience.

Livermore’s method certainly has some appeal, but one cannot help thinking that it might have been better understood and traded if it were chart based. But Livermore was not a chartist: “Personally, charts have never appealed to me. I think they are altogether too confusing,” he said.

Many people would disagree. But if you can understand trends without the charts you probably have a better understanding of movements in the markets than most. The rest of us need the graphic illustration to understand the direction.

In the second part of the story of the origins of trend following, we will discuss how all these loose parts got woven together into a clear trading approach and how it is used today.

Stig Ostgaard is an original Turtle trader who has been following trends since the 1980s. He is now a researcher and trader with Last Atlantis Capital Management LLC.

comments powered by Disqus