A Little History Lesson On Technical Analysis

Is technical analysis voodoo science, or is the methodology being accepted in today’s world of investing or portfolio management?

Thanks to the “new economy”—Internet, business television and social media—technical analysis has been embraced by many bright investors and savvy financial institutions. I recently did a Google search on “online stock charts” and got pages of names in seconds.

Aside from most of the online discount brokers—including the big banks—there were BigCharts.com, FreeStockCharts.com, StockCharts.com, Yahoo Finance and Google Finance. Most offered historical price data and plenty of technical studies such as moving averages, price oscillators, along with the popular RSI and the MACD indicators.

It was only 30 years ago that such a search was impossible—because there was no Internet, no search engines and, basically, no computers.

Yes, I know that Jobs and Wozniak had built the Apple II by then but aside from games, it had no business software, except VisiCalc. The more business-friendly IBM PC was introduced in August 1981, but for the most part the tools of technical traders still consisted of chart paper, a calculator and a newspaper.

One of those early technicians was Donald R. Stark.  I had the good fortune to work with Don when we were rookie stock broker/advisors at Richardson Securities just a few years before the great 1973-1974 bear market.

Don was one of those goofy guys who drew charts all day and basically ignored the fundamental-based research cranked out by Richardson’s Winnipeg head office. When the great 1973-1974 bear hit, we all scattered with Don settling in with Draper Dobie Ltd where he met the soon-to-be cycle legend Ian S. Notley. When Dominion Securities acquired Draper Dobie in 1977, they also had the good fortune to acquire the genius of Stark and Notley who then set out to create the original RBC Capital Markets’ respected Trend & Cycle Department (RBC).

They initially used a Digital Equipment Corporation (DEC) microcomputer, which used telephone lines to communicate with remote terminals. In-house, these were known as “Computer-produced charts from AUTOGRAPHIX terminal, DOMINION SECURITIES GRAPHICS LIMITED.”

Notley’s work on cycles was legendary as he expanded on sources such as the 1978 Dick A. Stoken publication entitled Cycles, What They Are, What They Mean, How To Profit By Them. That was the same year Robert Prechter collaborated with A.J. Frost in publishing the Elliott Wave Principle. Their source material went back to R. N. Elliott’s 1946 publication, Nature’s Law – The Secret of the Universe.

Don Stark was Ian Notley’s right hand. I still have a number of their brilliant 1980 through 1984 Trend & Cycle publications which I still review to-day. Their great top-down calls, ranging from calling the bond market “the buy of a generation” to predicting the re-structuring of the huge American multinationals and the early transition from the “old” economy to the “new” economy are the stuff from which legends are created.

In 1987 the Stark and Notley team left Dominion Securities Pitfield to part as friends, with Ian moving to New England and Don joining me and Karl Wagner at Market Fax Info Services. Don spent another five years there crafting his quantitative and inter-market skills before moving on as a sub-advisor to several respected fund managers.

Unfortunately, we have lost Don, Ian and Karl but some of their contributions to technical analysis can be found on the Canadian Society of Technical Analysts web site (CSTA.org / organizational / in memorium).

Today in spite of the “new economy” the Trend & Cycle platform at RBC remains basically unchanged from the original Notley and Stark methodology.

An example is a chart from an RBC publication dated May 26, 2014, displaying an opinion of TSX-listed CNR that I have “cloned” (Chart #1) using SuperCharts software.

The top histogram (OS 5/15/3) is a 5-period simple moving average, less a 15-period simple moving average with the difference smoothed by three. This study can be used on daily, weekly or monthly data. The squiggly cycle line overlay is the Coppock Curve which is most effective on monthly charts.

The lower line—below the CNR plot—is a simple spread (relative analysis) of CNR against the broader S&P/TSX Composite index.

Not a bad approach–using price momentum, cycles and relative analysis all on one chart—which is suggesting CNR has positive price momentum, an up (quadrant 2) cycle and flat to rising relative strength (established outperform) condition.

Today for my own work I have modified the original Stark & Notely work which is displayed in a second example (Chart #2) where I have placed two money flow studies over the OS 5/15/3 histogram and re-located the Coppock Curve. I then replaced the single spread line with three relative average lines—using the shorter thin line as a trigger for a relative rank change—say from early out performance to established out performance.

Of course all this technical work is just noise unless it is proven to work in the real world of investing or portfolio management.

A good “real money” example was the Union Securities Hybrid Investment Program, a discretionary program managed by a portfolio manager (PM) using both "fundamental" and "technical" analytical tools to select securities. The PM and my firm—Getting Technical Info Services (GT)—would split the Hybrid Portfolio with about one half of the portfolio being fundamental (PM) selections and the other half being technical (GT) selections.

The Union Securities Hybrid Investment Program was closed in mid-2012 when in October 2012 all of Union’s client accounts and assets were transferred to another IIROC Dealer Member.

I did however track an original “Mandate #4” Hybrid portfolio beginning with all the equity components at November 9, 2011 and held hypothetically through to November 12, 2013. The Union Hybrid Mandate #4 program was in the aggressive growth category that allowed up to an 80% exposure to equities and 20% to fixed income.

The chosen benchmark was the S&P/TSX Composite Index. At inception November 9, 2011, there were 12 fundamental components (mostly exchange traded funds) and there were 15 technical components which were a mix of Canadian and U.S. listed equities.

During the study period of November 9, 2011 through to November 12, 2013, our benchmark, the TSX Composite, had a capital return of +9%.

Over the same period, the 12 fundamental selections had a 2-year capital return of -8 %,

Over the same period, the 15 technical selections had a 2-year capital return of + 32%.

Observations:

The fundamental (PM) would have done okay except for two problems that seem to be common in many fundamental methodologies. Our PM was drawn into a value trap (RIM or BlackBerry) and our PM was caught in the need-to-own-gold as a hedge trap.

All of the technical (GT) selections were based on technical studies found in the original Trend & Cycle platform, such as price momentum, cycles, relative analysis along with my newer money flow numbers.

Over the same 24-month period the Hybrid Program also enjoyed five technical selections that were the subject of takeover bids, namely Gerdau Ameristeel, El Paso Corp, Biovail Corp. Viterra Inc. and ShawCor Ltd.

Now I know the technical studies I used are over 30 years old—ancient in today’s modern economy—but much like the Coca Cola formula, if it works, why mess with it?

Bill Carrigan, CIM is an independent stock-market analyst.

info@gettingtechnical.com