Zweig Breadth Thrust
March 20, 2009


One way to judge the strength or weakness of a stock market move is to study the "internals".  This is simply breaking down the current move into groups such as up stocks, down stocks, up volume, down volume, etc.  In this way, we can see how the move transpired at a more granular level.

Why do we look at internals?  The short answer is that not all moves are created equal.  A 100 point Dow move up with wide breadth (a lot of stocks moving up, or a lot of volume to the upside), is generally more bullish than a 100 point move with few stocks participating.

Today I present an internals indicator called the Zweig Breadth Thrust.  Funny name, but it triggers so infrequently, that we should take it seriously.  It just triggered as of Wednesday's close.  The idea behind the indicator is that stocks turn up sharply from oversold, but are not yet overbought.  The calculation is quite simple: 
  1. Find the number of advancing stocks
  2. Find the number of declining stocks
  3. Calculate the 10-day, exponential moving average (MA) of each
  4. Zweig Breadth Thrust = (MA of Advancing) / (MA of Advancing + MA of Declining)
I use this indicator in 2 ways:  traditionally and in a more subjective way.  First, the traditional method.  Dr. Zweig found that anytime the value goes from below 40 to above 60.5 within 10 days, the likelihood of a new bull market is quite high.  It doesn't happen very often, (only 15 times since 1945, including this week) but when it does, he claims that the average return over the next 11 months is 24.6%. 

Using Wednesday's Dow close of 7486.58, a 24.6% return would get us back to about 9330 in 11 months.  That seems conservative, as it only gets the market back to where it was around election day, November 2008.  It had already dropped quite a bit going into that month. 

Another way to view breadth indicators is to look for divergences.  Often, breadth indicators will put in a higher low when the market creates a lower low at important turning points.  This is called a positive divergence.  Looking back to 1986, I found 8 occurrences of positive divergence, including the current one confirmed this week.  Of the seven previous instances, the average 1-year gain in the stock market following the divergence was 27.3%.  Clearly, these rare events are worth paying attention to. 

The table below contains the date of the divergence, the following returns, and how far the market was down from it's recent high at the time of the divergence.  I found no correlation between the returns and how far down the market was.

  S&P500 pct down
Divergence 1 year return from high
12/4/1987 22.8% -33.5%
10/11/1990 29.1% -19.9%
4/11/1997 50.6% -9.6%
10/8/1998 37.3% -19.2%
4/4/2001 2.1% -27.8%
10/9/2002 33.7% -49.2%
8/12/2004 15.7% -8.2%
3/9/2009           ? -56.8%

By putting together the historical record of the "thrust" indicator plus the divergences, I consider it significant that they both happened concurrently in the past week.  While you can't invest/trade based on only one indicator for very long, we can add this breadth data to all the other information from the economy, sectors, commodities, individual company earnings and guidance, and Fed actions.  All things considered, while the economy may still take at least a couple quarters to become healthy, the market appears to be getting more bullish in anticipation.

Dan Grill


Below, you'll find a long-term chart of the S&P 500, and screenshots of all 8 divergences.  The S&P chart shows green arrows where I found these inflections, so you can get a perspective of the indicator's timing capabilities.

S&P 500, Monthly, 1984-2009

1987 Divergence, Zweig Breadth vs. S&P 500

1990 Divergence, Zweig Breadth vs. S&P 500

1997 Divergence, Zweig Breadth vs. S&P 500

1998 Divergence, Zweig Breadth vs. S&P 500

2001 Divergence, Zweig Breadth vs. S&P 500

2002 Divergence, Zweig Breadth vs. S&P 500

2004 Divergence, Zweig Breadth vs. S&P 500

2009 Divergence, Zweig Breadth vs. S&P 500