T Theory is a method of analyzing general investment trends using a time symmetry property I discovered in the early 1970’s. At that time the time symmetry was christened “The Law of Matched Trend Time” because it basically states the duration over which investors can obtain “superior equity returns” will always be equal to the previous time period in which returns were subnormal. A simpler way to put it is to say the market can only “make a strong run ” as long as it has previously ”rested”. As you might expect, the practical purpose of the theory is to anticipate the runs of “superior returns”. This time match property can be shown to be reasonably accurate and reliable historically so I have proposed as a natural property inherent in market trends but of course we have no real knowledge, so results can not be guaranteed. The time symmetry is most easily represented by the graphical “T” thus the name T Theory. In all instances the left side of the graphical T spans the market’s “rest period” while the right side spans the “ run period’ where returns should be the greatest. Since its discovery there have been a number of successes which I have referenced for interested readers. During the early years, I became acquainted with Marty Schwartz, who took a very early interest in the T concept, and applied his exceptional talents to become one of the greatest stock market traders. His exploits, philosophy and comments on my theory were presented in Market Wizards by Jack Schwager1
in 1989. To this day I still receive inquires on how to obtain the “secrets to wealth” that Marty discussed in this book. On this point I am sorry to to inform many of you that there are no easy secrets, only powerful tools with which you might learn to uncover major opportunities before the masses arrive. Marty and I have often talked over the years about the special advantages conferred on us by our US Marine Corps training. It takes a special state of mind to “sign up” for a short boat trip, in a flimsy landing craft, to a beach completely controlled by hordes who have anticipated your arrival and have set up every imaginable way to do you in. Buying into major market opportunities presents a similarly discouraging picture. You may have good reason to anticipate profits, but if a great opportunity does indeed exist, nearly everyone will be against you, including your friends, and the predominant opinion expressed by your peers, including people you respect, will be that you are embarking on a foolhardy enterprise. I believe that T Theory’s major contribution will be to show you why it will always be difficult to buy at major lows, but using its reasoning you may be able to overcome these obstacles. At each and every great buying point you must struggle at the “moment of truth” where you face seemingly overwhelming negative odds. In T Theory this moment of truth is called “The center post of the T”. It represents the point in time where all the bearish negatives of the past have been discounted by the market and is about to be transformed into an emerging, new bull market. Finding this key juncture is a potentially dangerous occupation, but as with all good Marines, one eventually remembers during the heat of battle, to keep ones head down, shoot straight when it’s your turn and always be faithful to your core principles in order to keep casualties to a minimum. In this regard it is wise to keep in mind a few basic T concepts:
Remember: The center post of a T can simply be viewed as a “Mirror in Time” in which the future is seen as an inverted reflection of the recent past. It is essential that you understand at the center post low, the pessimism of the recent past will give way to optimism in the future; the conservatism of the past will give way to speculation in the future, and the poor rates of return of the recent past will turn into superior rates of return in the future. If you don’t get this straight you will have great difficulty making money in the equity market.
The Early History
During the early 1970’s I stumbled onto the curious observation that strong market uptrends often lasted for the same number of weeks or months as the prior correction or decline. This was very apparent in certain cyclical stock market indexes such as the Dow Jones Transportation Average, the Value Line unweighed average of some 1600 stocks, and other cyclical indexes of stock prices back in those days. As I built up a history of these time symmetry relationships it was clear there were certain periods in which the duration of every cyclical uptrend nearly exactly matched the duration of the prior downtrend. Over the years, my job of keeping T Theory current in an ever-evolving environment is finding straightforward examples that illustrate this so-called “Theory of Matched Trend Time”. I am continually researching methods that can be applied to a practical matter of finding superior equity investment opportunities.
Historically my T Theory has evolved through three stages, starting with the Transport chart on the next page taken from an early study which was appropriate for its time. Later the same simple principles were used to establish equity mutual fund projections using the Value Line Unweighed Composite Index as I have shown in the next chart. For the modern era, I have shifted to the Daily Cumulative A-D line for the NY Exchange as the primary indicator of this presumed “natural” law. It can be seen in the color charts that follow. I have also added a brief illustration of the theory as it was applied to very long and very short trends to round out this introduction.
I should point out that one objection to the T concept over the years is that the Law of Matched Trend Time is unlikely to be true because it implies the market can only be strong for 50 percent of the time. Many contend that the market goes up “most of the time”. The reconciliation of this conflict centers around the broad toppy phases of market trends where a positive rate of return may well develop during a sideways trading pattern but the rate of return in such shallow advances is no better than the yield on risk-less Treasury Bills. For this reason, the Law of Matched Trend Time is more accurately stated as: “Superior rates of return (i.e. better than T-Bill rates) can last only as long as the prior period of inferior returns”.
A practical example of the Law’s consequence can be seen in the independent study by Nelson Freeburg in a Technical Analysis of Stocks and Commodities magazine article covering the period from 1957 to 1994. Using interest rate information and relative shorting divergences between the Specialist and the Public (a contrary opinion indicator), Freeburg showed an optimized mutual fund trading strategy justified owning mutual funds 51% of the time over the last 37 years. In fact, it can be shown that any long-term study of investment trends that attempts to make superior rates of returns will move towards a 50/50 split as the Law suggests in order to capture the best returns while avoiding the potential for losses.
The Ts are a great help in picturing how this occurs through the analogy of the mirror reversal I summarized. To provide a clear picture of how this plays out historically let us review the original Ts I discovered in the early days.
The Original Dow Jones Transport Ts (1956 to 1976)
The simple symmetric Ts in the monthly Transportation bar chart below illustrates the way in which the Law of Matched Trend Time could be used to interpret trends. This chart is taken from my early publication and has become somewhat faded with time so I have reproduced the chart’s original 1977 text.
Chart Text: The most powerful yet easy to understand example of the Law of Matched Trend Time is the so-called DOW JONES TRANSPORTATION Ts shown above. These Ts illustrate the basic principle of Magic T Theory that when the market has made an important bottom it will rally only as many months as it had previously declined. For example, the first T centered in 1957 states that after a 20-month decline the next advance will last a matching 20 months, terminating at the T’s right end. In time, the next T is formed and the sequence repeats through history. For convenience, the small Ts are identified by letters A to H. At less frequent intervals very much larger Ts appear when successive highs form a declining tops pattern as seen in 1962. Such Ts are of the utmost importance since they forecast major stock market sell points (i.e. 1968, 1972). Such Ts are identified by the year of the major low at the center of the T.
These simplest examples of the T concept illustrate properties of all Ts, including the modern examples, which should be kept in mind. The T concept only requires the time duration of the advance to match the prior decline time; it places no restriction on the magnitude of the rise or the duration of the original decline. So for example in the Transport Ts chart, the number of months the Transports have declined to a low is pretty much a random number over a very long period of time. It is for this reason that the time symmetry of the T has been lost in a history of random length declines.
However, it is my opinion that the real reason the T’s presence has been lost is that investors simply can not make the connection between the bullish trend and its opposite counterpart, the prior bear market. This closely parallels the “contrary opinion” approach to investing in which one sees opportunities in sold-out markets only when investor sentiment has turned extremely negative. I have found it helpful to think of the T in contrary opinion terms and it may prove helpful to you as well.
From a practical standpoint, the big T with its center post-Low in 1962 illustrates why investors and traders, particularly those who follow trends from a conventional technical approach so often miss out on the opportunities. For the purposes of illustration consider, you might be interested in investing in transportation stocks during 1962. The recent history up to then would have been disappointing with the averages tracing out a 75 month descending tops pattern from the 1956 high. A fundamentally oriented investor could have seen some merit in this situation because the price cap coupled with earnings growth could lead to attractive values by 1962. But most people would consider the trend a “loser” and pass up the opportunity.
From the Magic T viewpoint, we would tend to ally ourselves with the value investor in the left side of the T we specifically define any descending tops pattern as a “Cash Buildup Phase”. During this period we assume investors are shunning the equity investment and putting their assets into cash alternatives like Treasury Bills, etc. As a consequence I interpret long cash buildups as potentially more profitable because more cash reserves could be developing and, through the Theory of Matched Trend Time, we would anticipate the next bull market could last a long time, perhaps becoming greater in magnitude as the cash reserves build steadily higher over time.
All this might seem straightforward when you review the Transport T history but in real time investors often fail to grasp the fact that restraint of the market’s natural rising trend for a long period of time creates tremendous opportunities. In the real world the natural build up is for pessimism, not awareness of cash, and over time it dominates the thinking in a negative way.
A second factor that limits profitability for many is the suddenness of the stampede that develops as the new T gets under way. Often the sharpness of the initial advance, as all that cash buildup tries to get into the market, is viewed as a reason to get out of an “over bought” situation. The Ts I have presented above make it quite clear that once a new bull market has started it is much wiser to be patiently bullish for the equal time projected by the T in order to capture the major profit opportunity. Keeping the full right time span of these Ts in mind is also important because it helps protect against the sudden appearance of the new bear market which the chart shows is likely to begin as the Law of Matched Trend Time expires at the final top.
I leave it you to study all the examples in this study to see for your self how the symmetry of the T can alter your thinking about investments but from my experience it is wise to keep the following in mind generally:
Remember: The greatest profit opportunities come at the end of the longer cash buildups and you must manage your affairs to be mentally ready, and financially healthy so as to be 100% invested at this key juncture. Nearly all the rapid gains are generated in the early stages so concentrate on this aspect in the beginning. But major profits require the holding of good positions for the long duration of the T so don’t be pressured into selling early in the Ts projected rise. Finally be particularly wary of the T’s final rallies into its projected top date (at the right end) Late rallies can represent the speculative “blow off” that leaves the market vulnerable to the next bear market.
The Value Line Ts (1965 to 1977)
The next step in the evolution of the simple time symmetrical Ts made use of the so called “unweighed” Value Line Index plotted below as a proxy for mutual funds. The 1965 to 1977 chart plots the Value Line’s index of 1600 stocks assuming equal dollar investments in each stock and demonstrates how poorly the 1970-72 bull market had been for the average stock. Here it can be seen that in reality most stocks had suffered an ongoing bear market from the 1968 peak to the major 1974 low while the more popular Dow Jones Industrial, not shown, made an all time high in 1973. From this perspective, the Law of Matched Trend Time suggested the recovery from the 1974 low was destined to be more than most investors suspected since a decline of about 6 years total should lead to a general advance of 6 years.
When this chart was published back in 1977, the much larger T centered at the 1974 low was proposed as a more positive projection to a September 1980 peak in line with the simple conclusion that a 6 year bear market for the “ average stock” ought to produce a six year bull market for the majority of stocks. This insight gave me my first opportunity to grow a modest sum of money into something substantial because it was clear that this projection could not be made for the Blue Chip stocks which then were the darlings of the investment community. I was able to correctly deduce that the period in the late seventies was likely to benefit the smaller secondary stocks, which dominate this Value Line Index.
These two examples of the time-symmetrical T illustrate the basic idea and I don’t think I need dwell on the subject further. But before moving on to the more modern examples it is appropriate to note that Magic T Theory’s major task is to isolate major lows since the Theory of Matched Trend Time leaves this aspect of the T as what is known as “the independent variable in the equation”. Once a T has been confirmed the time symmetry pretty much makes the bull market duration fixed and one need only confirm that it is running in satisfactory fashion based on these historical examples.
Much of my future work will deal with the task of identifying what we call the “Center Post Low” of these Big Ts. In my experience, the greatest problem for most investors is dealing with the adversity of the market environment at these major lows from a psychological standpoint rather than a technical or analytical one.
Technical Note: The Value Line Index above has been superseded by the Advance-Decline in the next chart. To make the transition I have compressed the vertical Advance-Decline Line plot 1966 to 1974 in order to illustrate the same Ts you see above. The projected top dates are not changed because the time symmetry patterns are identical for our purposes.
The Modern Advance-Decline Magic Ts (1970 to Sept. 1995)
The modern successor to the Transport and Value Line Ts for the decade ahead are the sequence of Advance-Decline Line Ts covering the period from 1970 to the present as shown below. In the big color chart we use the New York Exchange Daily Advance-Decline Line to define the cash build-up phase which defines the left side of the T. This Advance-Decline Line is plotted along the lower portion of the chart. Each T’s cash build-up phase is defined as the period of declining tops in the Advance-Decline Line and is represented in red ink. An enlarged chart of this history is provided at the end of this study and key data is summarized in the table that follows. (Note T#7’s cash build-up actually “overlaps” the 1987 to 1990 period).
To represent the market’s daily trend over this period I have selected the Daily New York Composite Index of all stocks as the representative measure of equity values since it covers the value of all the stocks represented by the AD Line beneath. The Theory of Matched Trend Time, which forms the whole basis for my T Theory, is demonstrated by linking the T’s projected market rally as the green line having a time duration which is identical to the prior red cash build-up phase below.
Basically, I programmed the upper plot to start from each T’s center post low plotting the NY Index in green ink until the number of days elapsed matched the number of days of the prior red cash buildup phase. At this point the Law of Matched Trend Time says the period of superior returns has ended and I turn the plot red to indicate the daily NY Index is at some risk. The plot remains red until the center of the next T. The cycle then repeats. If one grants the T’s center post location is known, then Magic T Theory states the green line is the predicted period of superior returns while the red upper line is the period of high risk.
Because this history captures all uptrends, while correctly noting all important declines, I believe there is adequate evidence that a “natural law “ is at work, which basically requires the market advances to match prior Advance-Decline Lines. It is important that we show this to be true for all the history above if it is to become the basis for our future work. I think the charts speak for themselves. I am satisfied that its insight will be a valuable tool for the future. In the months ahead we are awaiting completion of the current T which projects the run up to a projected July 1997 projected top.
The larger version of the chart makes it clearer that the projected top dates are not precisely accurate. But as you would expect from a natural symmetry, the errors are equally distributed so the projected tops should be taken as an average estimate, not a hard date. Actually is doesn’t matter much anyway. The maximum profits are easily seen as accruing from the center post of each new T and the maximum risk is easily seen as coming during the decline into what becomes the center post low of the next T. So our real task is to develop a method of anticipating the next T. The starting point is to gather the historical specifics.
The Advance Decline T’s Performance Summary
The table below summarizes the key characteristics of each graphical T in the prior chart. The cash buildup duration in market days are, of course, defined with hindsight but the projected top date and gain figures are the logical consequence of the Theory of Matched Trend Time. There are some consistent characteristics of these Ts which we may look to repeat in the future. However it should be kept in mind that the successive center post lows are theoretically allowed at random intervals so low pattern projections are subject to large errors. Nevertheless two important observations come out of the tabulated data. First, the most common place for a T’s center post low is near year-end or in the last quarter of the year.
The second point is that whenever a completely new Cash Buildup Phase is necessary, this history suggests that at a minimum about 200 market days are typical. Note the data for Ts #4, #5 and #8. There are about 260 market days in a calendar year so a wait of 200 market days is a bit much in some circumstances particularly if the market has just made a speculative peak. Two hundred days is enough time for the business cycle to turn down or interest rates to rise dramatically by the time the next T is available to rescue the market from any such adversity. Also there is a another problem in trying to make too many judgments based on cash buildup times. T’s can have tremendous overlaps in time as the 7th T illustrates. This necessitates that we look at alternative scenarios and let the market’s technical trends point to the actual configuration of the new T. I spend considerable time studying the potential for new Ts as this aspect of the theory has the greatest potential for the future.
The Longest Range Mega Ts
During the late 1970’s and going into the 1980’s I became aware of the potential for a great new bull market that my theory visualized as from the great topping pattern in the Dow Jones Industrial Average at the 1000 level extending from 1966 to the early 1980’s. At the time it was popular to consider the market’s inability to get above the 1000 levels as a negative sign. However I recognized that this sideways pattern could be the rising phase or sash buildup for a huge Ts that could dwarf all my more modest interpretations. This massive time span concept was christened the Mega-T and right now represents one of the key concepts that I am extending as we approach the millennium. As always the concept is a pretty simple time symmetry but one that has to be pulled out of the very long term uptrend for stocks price. In this introduction I have extracted by 1994 Study of the Mega-T. During 1997 I will be extending this study which covers the 1920 to 1994 period all the way back to the 1800’s. In its final form I expect to show that the makes underlying nature to form time symmetrical patterns is a long established property and hopefully one that will be continued in the distant future.
During the 1970’s most of my initial research efforts were devoted to an understanding of how the stock market’s time symmetry principle I had uncovered could be applied to medium duration trends. This so called “Magic T” concept rested on my discovery that the market can only produce superior rates of return for as long as the market had previously demonstrated a subnormal rate of return. Having documented a number of successes in applying this idea for trading trends it occurred to me one day that the principle was also at work in the long Dow chart below once I added the simple definition of a “normal up-trend”.
I saw that laying down a simple straight line approximation to the long-term average rise in this blue-chip average from the 1920’s allowed me to use the T’s time symmetry property to explain how major investment trends evolve. By interpreting the market’s long-term price appreciation potential as a steady modest uptrend as shown below, I could apply the T Theory criteria which simply states the post war bull market “run” could only last as long as the prior depression induced subnormal trend or “resting” phase.
In order to illustrate the match between the resulting price appreciation and the prior “buildup phase”, I adopted the graphical “T” . Only the time span of the T’s “arms” is significant since the principle does not imply any particular degree of price appreciation. So for example, the first T in the next chart must have its left end at the 1929 peak while the second T the left end must start in 1966 where the rising trend ended and a “subnormal” sideways trend began.
The center post of the Ts lies in the last low before the character of the tops trend shows improvement relative to the existing norm. It is relatively easy to see the significance of the 1982 low but not so easy to see the 1949 low unless you reference these lows to the historic norm.
As a practical matter, one simply observes that long periods of under-performance provide a building opportunity for superior investments when the so called “buildup” phase is ended. Closer study is required as the center post of the eventual T is reached and the beginning of a more rapid advance is confirmed. Once under way it is best to invest with the new uptrend for a long period of time and treat corrections as normal given the history of Ts.
To track the up trends which are expected to be above average in the right side of the T, I have added a 150 Week moving average as a guide to correction lows. Note for the most part Bull market corrections hold at the smoother moving average during the projected strong periods of 1949-68 and 1982-1994. But it is likely corrections outside this favorable period will plunge much lower. Much of the technical work in my T Theory is occupied with understanding what to expect in terms of the risk/reward ratio within the right side of a new T vs the much riskier environment one can expect when the T expired such in the decade following 1968.
A second topic, and one most important for the years ahead, is the identification of potentially important “build-ups’ in alternative investments which can provide the superior potential after the current Mega-T has expired in 1998. By then the long term investment potential in US stocks is likely to turn “subnormal” as was the case after 1968. We will want to concentrate our attention on whatever areas show the biggest or best potential for a new bull market trend in a new T. This is a key topic for my 1997 studies particularly in the develoFpment of Ts for individual mutual funds.
To illustrate the practical significance of what I call “ Magic T Theory” let us review some of the key investment turning points identified in the next chart. What surprised me most was how knowledgeable investors failed to realize the popular trends of the day had exhausted its potential and exposed investors to tremendous downside risks. At the same time as risks were maximum, a minimum understanding of the T concept would have redirected ones investments away from these popular over-priced “fads” towards the new or emerging growth investments having greater potential.
We want to understand how investors were misled, how the better alternative investments could be identified, all with an eye to the current T’s projection that the great bull market started in 1982, will expire in late 1997 or early 1998. At that time the tremendous investment in common stocks will have been completed according to the T and money will tend to flow out of stocks and mutual funds into other investments which are just completing their buildup phases.
As a Magic T approaches its right end point, it is not uncommon to see investor’s interest turn away from “blue chip” stocks that have been picked over during the years of “superior performance” and begin speculating on relatively unknown “fad” stocks. During the late 1960’s such a speculation manifested itself in the creation of so-called “Go-Go funds”. These funds jumped from fad to fad until the exhaustion of the T in 1968 left the fad stocks vulnerable to the severe downside reaction for the next 6 years. At the time of the 1974 low, aggressive growth funds and stocks had suffered enormous losses (up to 80%) leading some to characterize this key low as “The End” of Western Civilization. These kinds of events could repeat after the current T expires in early 1998.
As of 1994 (when this was written) I have not see a repeat speculation underway but that’s partly due to us having advanced only 12 years from the key 1982 low whereas the T effectively calls for a 16 year potential, leaving another 4 year period for speculation to rear its head. At the next major market low, history warns us to look for signs of speculative investing and if it does occur, it likely is warning of a coming top in line with the Ts’s general.
Nevertheless, history alerts us to the possibility that at the next major low investors may try to jump on the speculative bandwagon rather than purchase familiar blue chip favorites. If so small capitalization, high growth potential stocks and funds may become the leader in the final phase of the current Mega-T. We will check this in future studies by comparing the relative strength of all sorts of alternate investments as an aid to discovering any important new speculative trends.
Another historical observation worth remembering is the institutional speculation that developed in the late sixties in the so-called “Nifty Fifty” stocks which were viewed at the time as “perpetual” growth stocks to be bought and never sold. It is easy to see why institutions could assume a perpetual growth concept. The long upward trend of stocks from the major 1932 low made it clear by the sixties that simply holding stocks with highly visible earnings potential was a superior strategy.
However the nature of the T makes it clear that some arbitrary time limitation should exist as to the “superior growth” phase. Over time investors had been conditioned to believe continuous growth was possible and they chose to dismiss the 1929-1932 destruction as an aberration. However in 1973, the perpetual growth assumption underling the enormous popularity of these “nifty” IBM, Polaroid, Xerox, etc. stocks began to unravel and those who held on saw enormous destruction as these companies lost their leadership and effectively became the great “dinosaurs” of the investment world.
The moral of these examples that investors tend to overlook the fact that very long growth phases characterized by a generally rising bottoms tend to lull investors into a false sense of security. The concept of the T puts a time limit on such trends and so alerts us to the approximate time when the growth assumptions run out. As for the “magic” of the estimate, only history will record whether the current Mega-T will lead to a repeat of these episodes after 1998.
The Short Range Ts
As a final example I am summarizing my 1996 work on the Short Range Ts which use daily volume data to pick out the time symmetries that span 6 to 12 weeks.
During 1996 I reached back to some of my earlier work on the Theory of Matched Trend Time to see if the shorter trends, typically 6 to 12 weeks or so, might show a useful time symmetry properties. During the research I discovered that the daily Advance-Decline Line was best for developing the Long Range Ts , but so-called “on balance volume” information gave the better results for developing the Short Range Ts you see in the next two charts.
As you may remember from Long Range T Theory, the daily Advance-Decline Line declining tops trend is interpreted as a Cash Build Up Phase, which over time, builds up potential for the market averages to sustain an up trend for an equal number of weeks, months, or days as the case may be. For these new Short Range Ts, the On Balance Volume Oscillator simply takes the place of the Advance-Decline Line and we measure the Cash Build Up Phase by the time span of descending tops.
For these shorter Ts the usual rules apply but they develop much quicker. First, the Center Post Low of any new T must lie to the right of the last T’s projected peak so that the market has the opportunity to decline to a full oversold condition at the Center Post of the next T. Second the Center Post of the new T must be free to move forward in time to reflect the market’s natural ability to adapt to new fundamentals while maintaining the basic time symmetry between a specific market rally and its prior Cash Build Up Phase.
A third observation warns of potential pitfalls in applying these Ts. History shows the gains within the right side of the successive T are unusually variable. Generally the market does well after the downward sloping Cash Build Up line is cut to the upside as long as the market is projected to still be headed high by the more significant Long Range Advance-Decline Ts. During the bear markets that separate these bullish periods, the projected rally tends to be failure prone or erratic at best. This just goes to prove the old saying “ you can’t make a silk purse out of a sow’s ear”. In other words, a projected advance in the right side of a Short Range T will not materialize as expected if the key technical and fundamental ingredients are not in place.
The most important technical ingredient to insure success in any new Short Range T is an oversold market condition. The most important fundamental ingredient to insure a good rally in the right side of any new Short Range T is prospects for declining long term interest rates. This usually defined as the yield on the 30 year bond. Tracking the bond yield will also be introduced during 1997. As you might expect, the period of Cash Build Up in T Theory would normally be expected to take place in an environment of rising long term interest rates since stock investors look to long term bonds as a normal investment alternative. High or increasing bond yields therefore might be expected as part of the cash asset buildup process. I have seen Ts in bond yields and am interested in pursuing them because they may shed light on prospects for market lows, gold verses bond preferences and upside potential for these Short Range Ts.