Bill Rempel, a.k.a. NO DooDahs!

The definitions of trend range from the simple to the philosophical, with simple being something like “put the chart on the wall, back up and see what it looks like from the other side of the room,” and philosophical being something like “markets don’t trend; trend is an idea that we bring to the market in order to evaluate its movement.” I prefer the philosophical definition, because the idea of a trend can be quantified, moving it from the philosophical to the scientific. Here are seven quantifiable ways to define “trend.”

The Efficiency Concept From point A to point B in time, the market has moved from price A to price B, and those prices may be the same (no net change in price) or very different (the market has doubled or even tripled). In the interim, the time between point A and point B, it’s highly likely that the market didn’t take a straight line movement from price A to price B. If we make some estimate of the total price movement in the interim, we can divide the endpoint measurement of price movement by the total to develop an “efficiency of movement” measure. This is the method used by the Thermostat Trading Strategy, which can be found in George Pruitt and John R. Hill’s “Building Winning Trading Systems with TradeStation.” They called their function “Choppy Market Index” and divided the absolute value of the 29-day change in price by the difference between the 30-day high and 30-day low price, then multiplied by 100 so that the range was from 0 to 100 instead of from 0 to 1. Their cutoff to define a trend was “20″ - below 20 and they traded counter-trend, above 20 and they traded in the direction of the trend.

The Average Directional Index (ADX) J. Welles Wilder developed this back in 1978, and it’s an indicator available at StockCharts.com. This indicator starts with two indicators, the +DI and -DI, or the Positive Directional Indicator and the Negative Directional Indicator. Real imaginative, eh? Over the time period used, the +DI and the -DI measure the relative amount of movement up and down in price, and when the two are combined and smoothed with a moving average, the ADX is created.

Many of the most-beloved indicators of trend are based on moving averages of price, and either the current price’s relationship to the average, or the relationship of different-length averages to each other.

The “Grandaddy” - Single Moving Average If you visit Mebane at World Beta, or Kirzner Ferver at (oddly enough) Kirzner Ferver, you’ll see that they use a simple moving average, evaluated monthly, to determine if an asset is trending enough to be safely included in their asset allocation schemes. I strongly suggest you read their archives and browse their research, which is comprehensive, regarding this tool. As I mentioned, Mebane and K.F. both evaluate the price’s position versus the single moving average only once a month; there’s a good reason for this. If you use the price versus a single moving average, and evaluate often, you may wind up getting “whipsawed” - receiving a series of quick, unprofitable, contradictory signals as the price dances around that single moving average. That sucks. One way to control for whipsaws is to check the price’s position versus the moving average only infrequently; but there are other ways …

The Moving Average Crossover The idea here is a prevention of whipsaws by looking for trend signals that fire less often. If, instead of monitoring price versus a single moving average, we monitor the position of a shorter moving average versus a longer moving average, we get fewer signals. If the shorter moving average is higher than the longer average, the trend is UP - and vice versa. This technique is often used by mechanical trend followers on baskets of 20, 30, or even more futures contracts. It’s a very nice diversified methodology if you’ve got the capital to trade it …

Three Moving Averages This is a combination of the above techniques, where a longer-term trend is established by the two shorter moving averages being on one side of a longer moving average, and the trend is confirmed by the alignment of the two shorter moving averages being in rank order. The strongest trend is when all three line up, for example, 10-day above 20-day above 30-day, and a crossover, like a 4/8/12 day crossover, can be used as a trend-following entry system. You could also think of it as measuring trends within a longer trend.

The problem with all of the moving average techniques covered thus far is that while they define the existence and/or direction of a trend, they don’t really measure the quality or strength of the trend, unlike the Efficiency Concept and the ADX, and that quality or strength measurement may come in handy.

The Moving Average Convergence/Divergence and The Percentage Price Oscillator The PPO or Percentage Price Oscillator is really just a percentage-based version of the MACD or Moving Average Convergence/Divergence. The MACD takes two moving averages of price and measures the distance between them, and whether that distance is getting larger or smaller. In this way, it not only establishes the direction of the trend, but also the strength of the trend. Here’s the reason I don’t like MACD - you can’t compare the strength of trend across a large range of prices! In the context of one issue (stock, futures contract, etc.) over a short period of time (weeks or months), that doesn’t matter much, but when prices have changed a lot, like the Dow going from 1,000 to 14,000 over 25 years, the MACD is worse than useless, it’s downright misleading.

The PPO avoids the pitfalls of the MACD when comparing trend over long periods of time, because it divides the difference between the two moving averages by the value of the longer moving average, making it equivalent over longer-term charts (like 25 years of the Dow, or 3 years of GOOG). The other big advantage of the PPO over the MACD is that the PPO can accurately compare the strength of trend across multiple issues that are trading today, so one could use it to determine if the trend in Gold was stronger than the trend in Oil, or vice versa.

This not an exhaustive list by any means, but these are seven of the most common methods for defining trend; Efficiency, Average Directional Index, Single Moving Average, Moving Average Crossover, Three Moving Averages, MACD, and PPO. Happy trend-hunting!

If you liked this post, you might be interested in subscribing to my RSS feed. If you prefer, you can get a nightly RSS email update sent on the days that I post! There are convenient “Subscription” icons near the top of the right sidebar.

To view my actual trades and model portfolios for the different systems I track, visit The Rempel Report. If you’d like to become of member of The Rempel Report, you can register here. At The Rempel Report, I track model portfolios for four different mechanical trading systems, disclosing all results (good and bad) at regular intervals. I also track my personal portfolio, and disclose all trades before I make them. Members receive email notification of new posts and can contribute to the site through comments. Registration is still free!