Martin Pring has performed extensive pattern and indicator analysis over the years, but his market outlook is driven by a long-term economic model that provides the foundation for applying technical tools. Roughly 10 years into his career in the financial markets, analyst and author Martin Pring found himself on the correct side of a couple of historic market moves — the precious metals bull market that peaked in early 1980 and the bond bear market that bottomed in 1981.
Martin Pring has performed extensive pattern and indicator analysis over the years, but his market outlook is driven by a long-term economic model that provides the foundation for applying technical tools.
Roughly 10 years into his career in the financial markets, analyst and author Martin Pring found himself on the correct side of a couple of historic market moves — the precious metals bull market that peaked in early 1980 and the bond bear market that bottomed in 1981.
“We did very, very well, being long gold and silver and short bonds,” he says.
He took away a very valuable lesson from the experience, although not necessarily the one you’d expect.
“In retrospect, it wasn’t skill, it was just pure luck,” he says. “After a few more years of trading you begin to realize you weren’t the genius you thought you were. It was the lucky side of the coin.”
Pring has continued to refine his market models and research in the nearly 25 years since that windfall, augmenting a fundamental understanding of the market environment with technical strategies to help exploit it. For the past 13 years he has been a partner of Pring Turner Capital Group, a registered investment advisor.
He launched his financial career in 1970 at A. E. Ames & Company in Toronto, which at the time was the second largest investment dealer in Canada. While in training, Pring read Robert D. Edwards and John Magee’s classic work on chart analysis, Technical Analysis of Stock Trends.
The book’s primary premise — that price charts represent the ongoing battle between supply and demand — made a great deal of sense to him. But he didn’t know at the time that technical analysis would later play a significant role in his career.
After completing his training with A. E. Ames in 1970, Pring ran a small brokerage office in Owen Sound, Canada. Because of his dissatisfaction with the firm’s published fundamental research, Pring began to do his own technical analysis; if his work provided a sound outlook on a particular stock, he used the stock’s fundamentals to explain to his clientele why he liked the company.
While still in Owen Sound, he met an institutional client of the firm who had a copy of the Bank Credit Analyst (BCA), a monthly research journal that forecasts international financial trends. The publication’s combination of fundamental analysis and chart examples of the longer-term market trends appealed to Pring, so he pitched the publishers on the idea of letting him provide them with technical analysis to complement the journal’s fundamental analysis.
After a period of fulfilling this role on a consulting basis, Pring moved to Montreal to work full-time in BCA’s editorial office. The position eventually led to Pring’s development of technical and monetary models of the stock, bond, and commodity markets (see Pring’s six-stage model). These models put him on the right side of the precious metal and bond markets.
After these windfalls, he struck out on his own as an analyst, although he continued to consult for Bank Credit Analyst. Since then, Pring has worked extensively as a trading advisor, market analyst, commentator, and author. He has written 11 books, including Technical Analysis Explained (originally published in 1979, McGraw-Hill), now in its fourth edition. Translated into more than 10 languages, Technical Analysis Explained is one of three main books used for the Market Technicians Association’s Level 1 CMT (Certified Market Technician) certification.
In 1981 he founded the International Institute for Economic Research and began publishing his newsletter The Intermarket Review. His other books include The All Season Investor (1992), Investment Psychology Explained (1993), Martin Pring on Momentum, and most recently, Martin Pring on Price Patterns (2004, McGraw Hill).
The Intermarket Review, which has monthly and weekly editions, combines a business-cycle approach to the markets with Pring’s momentum and technical analysis. Pring has developed barometers for measuring both the fundamental and technical environments for stocks, bonds, and industrial commodities. This business-model approach was inspired by the work of Leonard P. Ayres, as detailed in his 1940 book, Turning Points in Business Cycles.
“Ayres invented the advance-decline line,” Pring says. “He went back to the beginning of the 19th century and showed there was a chronological sequence between bonds, stocks and commodities. He also [showed how] new stock issues lead the way for the stock market.” Pring took the idea of a “chronological sequence” and developed his six-stage cyclical/economic model.
AT: Let’s start with a quick overview of your market model.
MP: It consists of three markets — bonds, stocks, and commodities — each with two turning points, which results in the six points or stages in the overall cycle. It’s never perfect — each cycle is different, and there are different lead times between markets and moves of different magnitudes — but it works.
The size of the moves of each market is largely determined by the secular, or long-term, trend. For example, in the post-WWII period, the interest-rate market consisted of a series of rising peaks and troughs, which reflects higher bond yields. Since September 1981, however, interest rates have been tracing out a series of lower peaks and troughs.
At this juncture, there is an interesting question to consider: Has there been a reversal in the long-term trend because we’ve witnessed a huge rally in commodity prices? Some of the commodity indexes, such as the Journal of Commerce Industrial Price Index, have made new momentum highs in this cycle. This is one factor that suggests the secular trend has reversed for commodities, which means a rising interest rate environment and rising commodity prices. It’s still too early to be absolutely sure. Most likely, bond prices will experience a trading-range environment as the secular trend reverses.
The Fed may be using a too-tight interest rate policy at this time. My “Growth Indicator,” which is a combination of five economic indicators, has gone below zero. This condition has had an excellent track record of calling tops in short-term interest rates since the 50s. Areading below zero indicates the economy consistently. I’ve looked at differential purchasing power, differential interest rates, current account changes, and I find nothing works. My conclusion is currency trends are driven predominantly by market psychology.
However, I was curious about how the dollar’s trend fits into stage analysis.
When I researched this, I found that since the 60s, nearly each major low in the dollar has occurred at the end of stage five, which is where we were at the end of 2004.
Stage five is characterized by rising yields, rising commodity prices and falling equities, and it occurs when there is a tight monetary policy. It makes sense the dollar would bottom in stage five because the price of the dollar should go up if the Fed is controlling the supply by keeping conditions tight.
AT: Can you use your barometers for market timing?
MP: My barometers measure the environment, not what is happening in the market itself. Usually, what the markets are doing is what the environment indicates should occur, but that’s not always the case.
Because price trends are driven more by crowd psychology than anything else, sometimes stocks do not necessarily react to the fundamentals. It’s the market players’ attitude about the fundamentals — not the fundamentals themselves — that determine stock prices. Otherwise, how can you explain a situation such as the “Nifty Fifty” (a popular group of supposed blue-chip stocks that collapsed in the early 70s)?
Every year their earnings went up — they were one-decision stocks. But they topped in 1973 and didn’t make new highs until 10 years later — despite rising earnings. It was the market players’ attitude toward the earnings of this group, not the earnings themselves that counted.
The same thing applies to my barometers. They measure the environment and reflect traditional bullish or bearish conditions for each of the three markets, which I have further classified as the settings for each of the six stages. For example, Stage 4 is bonds bearish, stocks bullish, and commodities bullish.
One indicator I use is the commercial paper rate, which is the interest rate corporations pay for funding their operations. If this rate is above its 12-month moving average, it’s one component signaling a stage-4 environment.
Because stocks are the most psychologically driven of the three markets, the stock market trend can defy the barometers. Commodities and bonds are more driven by economic forces.
AT: What do you do when you see situations such as the stock market moving counter to what your fundamentals are indicating?
MP: That inspires me to dig a little and see if there’s something I can find that might explain the situation.
I was curious about how the stock market performed in years ending in five. Since 1885, stocks have gone up at a fast clip every year ending in five; the average for the 20th century was 29 percent.
There were only two down years before 1885 — 1875 and 1845 — and 1875 was in the middle of a depression. I thought perhaps this phenomenon was the result of years ending in four being weak, and years ending in five represented rebounds. But when I went back and checked, it turned out that on many occasions the market was actually overbought in years ending in four.
Then I checked to see what monetary policy was like — perhaps the monetary policy was easy in years ending in five. Going back to the 50s, it turns out my barometers were bearish three of the five years ending in five.
So, there is no rationale for why years ending in five have been positive years for the stock market, but the data shows this is the case.
AT: Do you then look to technical analysis to confirm the barometers?
MP: The problem is the barometers are objective, and assessing the technical picture is subjective. I don’t like to override the barometers unless the technical picture is very clear for the case against the barometers.
AT: But some technical work can be defined in an objective manner.
MP: True. One objective technical approach I use for confirmation is the 12- month moving average applied to the S&P 500 index. If the S&P 500 is below its 12-month moving average and the barometers are negative, the environment is bearish and stocks are responding to the negative environment.
AT: Just for the S&P 500?
MP: No, the same technique can be used for commodity indices. For example, I use a “diffusion indicator” that measures the percentage of individual commodities above their 24-month moving averages. The indicator has a consistent track record going back to the 60s. It went bearish for commodities a couple of months ago, but the commodity indices have not crossed below their 12- month moving averages yet.
AT: So far, it seems as if your work really takes a long-term view of the markets — looking back to the 1800s, applying 12-month moving averages as confirmation, and so on. Many people today are much more short-term oriented for both analysis and holding periods.
MP: Computers have shortened everyone’s time span, which I don’t think is necessarily a good thing. More people, including money managers, tend to be much more focused on the short-term. It’s like a drug.
The most successful traders don’t let their emotions affect their trading, but the more short-term oriented you become, the more you watch each and every tick, and the more emotional you become. Instead of trying to make decisions every other hour, many people would be better off if they looked at the markets once a week, made adjustments, and lived through the price swings.
AT: The volatility can shake up non-professional, short-term traders.
MP: It’s very difficult to get away from measuring our portfolios every 30 seconds. And I can’t cast the first stone on that one.
AT: What are your thoughts on developing trading systems for people who really want to focus on short-term time horizons?
MP: Developing a mechanical approach using short-term data is fraught with a lot of problems. I have yet to find something that consistently works. I’m not saying it can’t be done, it’s just that the only approaches I have found that consistently work are the longer-term approaches, especially ones with an intermarket component.
One good example is tracking the three-month commercial paper rate: If the rate drops below its 12-month moving average, indicating rates are falling, and the S&P 500 crosses above its 12- month moving average, it’s very bullish. Historically, it’s a high return with very little risk.
Currently, the model is neutral because the commercial paper rate is above its 12-month moving average, indicating a rising rate environment, and the S&P 500 is above its 12-month moving average. This model went long in April 2003 (when the market embarked on a roughly 10-month rally) and went neutral in the fall of 2004.
AT: Another popular technique is to use valuation, where bullish signals are flashed when the market is undervalued and bearish signals are triggered when the market is overvalued. What are your thoughts about these models?
MP: I don’t use valuation in my barometers, but I do watch things like the dividend yields of the S&P 500. Valuation models are sentiment indicators. You can watch for when they get overbought or oversold, but you have to wait for the actual trend to reverse. As we found out during the stock bubble, something that’s overbought can become even more overbought. It’s when the trend breaks that the psychology really changes, and that’s when valuation becomes useful.
Some analysts will look at a valuation level for the S&P and say the index is slightly above the norm, and therefore the market should come down to the norm. Well, that isn’t enough information. You have to know where the valuation extreme peaked or bottomed. If it was extremely overbought, the market will likely go to an oversold reading instead of just returning to the norm. There is a pendulum aspect to crowd psychology: The mood swings from extreme optimism to widespread pessimism and back again, and it doesn’t happen overnight.
AT: Earlier you mentioned technical analysis can be subjective, but your recently published work (Martin Pring on Price Patterns) details a quantitative approach to analyzing chart patterns. What did your research reveal? MP: I’ve done a lot of work on price patterns, including one- and two-bar price patterns (see “Related reading”). I wanted to put together a complete picture of price patterns and show patterns that work and patterns that appear on a regular basis. You don’t see some price patterns very often — diamond patterns, for example, are rare.
Also, many people don’t understand how price patterns are constructed, so I wanted to explain their basic building blocks — things like peak and trough analysis, trendlines, and volume relationships. If you know the building blocks, then you’re in a better position to understand the patterns and figure out whether breakouts from them are likely to be valid.
AT: What patterns did you analyze, and over what time period?
MP: We tested head-and-shoulders tops and bottoms and double tops and bottoms from 1980 to 2003. We determined the dates and lengths of the primary bull and bear market trends through visual inspection. Then we used the computer to isolate the patterns.
I was fortunate enough to have the support of Rick Escher of Recognia, a software company out of Ottawa that has developed excellent pattern-recognition algorithms. They did the testing for me. The computer identified approximately 5,000 patterns — about 1700 bullish patterns within bullish trends, 500 bullish patterns in bearish trends, and the balance being bearish and bullish patterns in the bear trends.
We found a number of interesting things. First, we found head-and-shoulders and double top/bottom price patterns did work. However, we also found these patterns were more prone to failure if the signals were contra-cyclical breakouts.
For example, if a pattern gave a buy signal in a primary bear market, it tended to fail or produce less profit. If the same signal occurred in a bull market, it tended to work. I already knew this based on experience, but it was good to see the research showed the same thing. My site allows you to scan an online database of more than 15,000 securities for numerous price patterns and technical breakouts.
AT: Pattern recognition by computer is a real challenge compared to what the human eye can see. What are some of the issues you had to address?
MP: One question is the time element. You might see a bullish head-and-shoulders top pattern, but following the breakout of the neckline (the trendline connecting the two lows on either side of the head), it might be two years before the target is hit. The software measured the number of days a pattern took to complete and called this "L," which stands for the length of the pattern.
If a pattern took 50 days to form, which is one “L,” then the question was: How many “Ls” until the target is hit? So, staying with the same example, if the market took 100 days to hit the target, then it was a “2L” pattern. We gave each pattern up to 5L to be completed.
AT: And what did you find out by doing this?
MP: In bull markets 70 percent of the patterns reached their initial pattern objectives within 5L, and just under 25 percent reached four times their initial pattern objectives within 5L.
In other words, if a pattern took 50 days to form, during a bull market it reached four times its objective within 250 days nearly 25 percent of the time.
AT: Were you able to define risk parameters for these patterns?
MP: That was one disadvantage, in that we don’t know the risk of the patterns. In the book, we did devote a chapter to whipsaws and how to anticipate them. For example, an upside breakout at the completion of a pattern on declining volume was a red flag. We also discussed how if short-term momentum was overbought, then an upside breakout from a pattern was less likely to work.
AT: How critical was the volume level on a breakout?
MP: The research showed that a lot of patterns worked without confirmation of heavy volume, but if there was heavy volume on the breakout, it was a very good sign for the pattern.
AT: What about the one- and two-bar pattern you mentioned?
MP: The one- and two-bar patterns are very akin to candles. My goal here was similar to explaining the building blocks of the larger patterns. Understanding how these short-term patterns can represent changes in market psychology puts the trader in a better position to interpret them.
For example, if you see an outside bar (a bar with a higher high and lower low than the previous bar) preceded by a sharp rally, it indicates a change in psychology. You want to see the outside bar opening in the upper half of the previous bar’s range, take out the previous bar’s high, and then close in the lower half of the range after taking out the previous day’s low. This represents a change from a bullish situation to a bearish situation.
For the most part, we looked at these short-term patterns for follow-through over the next five to 10 days. But they do appear at longer-term trend reversal points as well. There can be a domino effect: If the longer-term momentum was overbought, and both the short and intermediate momentum are overbought as well, then a short-term price pattern could be the start of a longerterm change trend.
AT: That makes sense.
MP: Most of technical analysis is just common sense. If a market is making a series of higher peaks, and the last peak was slightly higher, we would say the pattern is not, strictly speaking, a double top. But if the market breaks the low between the two peaks, then the market is really telling you the same thing double- top pattern implies, i.e., that the trading range reflected by the pattern has been resolved in favor of the sellers. These ideas really come from Dow Theory — that uptrends are rising peaks and troughs and downtrends are declining peaks and troughs. If there’s an indication things are changing from rising peaks and troughs to declining peaks and troughs, the trend has changed. Again, using common sense goes a long way toward understanding how to interpret technical analysis.