NEW YORK (TheStreet) -- The impressive 3.95% rally in the Dow Jones Industrial Average (DJIA) this week (see chart) caused my brain cells to remember a market study done by a research boutique and money management firm in ivy-covered Princeton, N.J. -- Delafield, Harvey and Tabell (DHT).
They found, using the Dow Jones Industrial Average and going back to 1966, that if you went long the market every time the DJIA rose 4% or more in one day and held for the next 12 months, the return was an average of 23%! There were 22 accurate signals between 1966 and 1999. Every one led to a higher market one year later -- not shabby for such a simple indicator.
I learned about this simple, but very effective technical indicator from Ken Tower, CMT, who was the chief market strategist when this tool was "discovered" in the 1980s. DHT was not immune to the changing financial landscape. It was bought by U.S. Trust and over the years the principals retired. U.S. Trust was eventually taken over by Charles Schwab. Ken Tower did the original research. We tracked down Ken, who is now CEO and Chief Investment Strategist at QAS in Jersey City, N.J., for an update.
Technical indicators come and go. For example, odd lot statistics used to be a great tool of technicians to spot what the little investor was doing, but the indicator was shelved when it became easy to buy other instruments. This 4% rule is far from perfect and infallibility left this indicator back in 2000. Tower told Real Money that 4% is not a magic number, so 3.95% is fine. In looking for signals, Tower would scan the data for rally days greater than 3.5%. With the experience gained from poring over many market monitors, Tower found most indicators with precise trigger points are unreliable.
Overall, there are 67 signals from January 1966 to date (not counting this week's move) and the average one year ahead was +15.38%, but there were a string of failures in the 2000-2002 bear market (average one-year loss 14%). This was followed by 10 accurate signals (average +19.61%) between July 2002 and March 2003. Then another set of failed signals came between March and September 2008 (average loss 27%).
Finally, a string of 22 positive signals came between October 2008 and November 2011 (average gain 26.8%).
"So the 50-year record is far from perfect, but I think it's still useful," Tower said. "The only unfavorable signals occur in massive bear markets. If the signal doesn't work it tells you what type of market you are in." With single-day rallies of 4% occurring more frequently in modern times, it may be necessary to adjust the threshold or add a secondary factor, he opined.
Was there a simple indicator to identify a top successfully?
"No. Unfortunately, in equity markets, market tops and bottoms are very different. What works for one doesn't work for the other, " Tower added.
So what now for the Dow? This recent rally in the Dow was thanks in part to the Federal Reserve and China. As the Fed seemed more willing to be pragmatic on interest rates and the Chinese seemed keener on stimulating their economy, stocks were buoyed. We'll have to wait and see if either of these macro supports will hold and, for now, we'll have to conjecture the moves from the only sure signs printed in the past in the charts.
--By Bruce Kamich and Sebastian Silva