Rare Market Buy Signal with Perfect Track Record Triggers for Just the 18th Time Since 1945

Last week's wild roller-coaster ride ended with three days of rocket-launch rallying across the major stock and bond indices. However, the index-level price jumps weren't the most important feature of the action. Breadth was.

Specifically, we saw numerous breadth thrusts during the rally that combine to suggest, according to historical data, that the lows prior to that rally are consistent with prior instances that form a data set overwhelmingly aligned with a key bottom in the market.

Breadth thrusts have become something like the gold standard for technical and 'market fundamental' signals. One such indicator, known as the Zweig Breadth Thrust, has a perfect track record for predicting key market bottoms and strong gains ahead going back over 75 years.

Other measures, such as the SPY Breadth Thrust, the TCTM Composite Thrust Model, and the 91% McClellan Oscillator Crossover Breadth Thrust, also have very strong historical records.

Last week triggered all of them.

 

The Fly in the Ointment

However, the one fly in the ointment--and it may or may not matter for assessing the value of this breed of signal--is the notion that everyone now closely monitors these signals because of their strong historical track records.

In the past, we have seen other very strong signals gradually degrade in their predictive utility once everyone starts to watch them closely because the market has a way of eliminating the value of something that everyone is using. For example, there was a time when moving averages actually worked consistently as systemic trading signals. Now they represent a zone that is heavily gamed and aggregate results are no better than random.

The mechanics of that process of degradation have to do with competition for access, which leads naturally to a competition for predicting the signal and accessing it ahead of the crowd, which, in turn, leads naturally to a process where traders scrambling for exposure ahead of the actual signal end up causing the signal—a self-fulfilling prophecy.

In other words, would it have been a Breadth Thrust if no one was trying to buy everything under the sun in anticipation of it becoming a breadth thrust after some initial symptoms of a breadth thrust appeared but before the genuine signal, itself, became fully manifest? Or, instead, was it that anticipatory buying that caused the signal to appear? And, if the latter, does that degrade the potency and meaning of the market signal?

An even simpler way of conceptualizing this problem is to simply ask yourself: Is it possible for a market signal to exist with predictive force where all market participants can benefit?

Since markets don't work that way (it's impossible for a market 'buy' signal to exist, for example, where everyone is on the right side because then there would be no one to buy from), some have started to question the reliability of these kinds of signals now that they are so popular.

 

But Perfect is Perfect

All of that said, this discussion is merely theoretical at present. In fact, the strict definition of the Zweig Breadth Thrust still has an intact perfect track record of calling key bottoms and predicting strong gains over following months.

Coming into last week, this signal had been triggered 17 times since 1945. The average gain following each of these thrusts was 24.6% in an average timeframe of 11 months. The S&P 500 has never been down a year later following one of these signals.

The measure is calculated by dividing a 10-day exponential moving average (EMA) of the number of advancing issues by the number of advancing plus declining issues.

In other words, over the prior 10 days, the number of stocks showing gains on each day is computed as an average with a statistical skew prioritizing more recent days in that set over older days in that set, and that data point, for each day, is then taken as a percentage of the total set of all advancers and decliners. The resulting number triggers the signal if the sequence evolves from statistically very rare very low values to statistically very rare very high values during that single ten-day period.

Let's say, for example, that the number of NYSE advancing stocks over the past 11 days is the following set (there are 2400 stocks on the NYSE): 495, 635, 1570, 609, 1314, 540, 1702, 1677, 1432, 2121, and 1977.

Alongside this series, then, is the series of 10-day EMA values for that series (which obviously incorporates the prior 10 days of advancers values at the start): 852, 983, 915, 987, 906, 1051, 1165, 1213, 1378, and 1487.

Since we know the advancers plus decliners must add up to 2400 for NYSE, we can now compute the series of Zweig Breadth Thrust scores for that 10-day period, which is simply the above list of 10-day EMA data points for each day divided by 2400: 0.36, 0.41, 0.38, 0.41, 0.38, 0.44, 0.49, 0.51, 0.57, and 0.62.

A Zweig Breadth Thrust occurs when, during a 10-day period, the Breadth Thrust indicator rises from below 0.40 to above 0.615.

The data series we have focused on above is, in fact, the data series for the 10 days from 10/23/2023 to 11/03/2023 (the ten days culminating in, and including, Friday of last week). In other words, because we see scores in this ten-day series that go from 0.36 to 0.62, we have an official Zweig Breadth Thrust--only the 18th time it has happened since 1945.

 

Think About It

At this point, I would encourage you to look away from the math and the data points and simply think about what such a moment implies: a reading of below 0.40 is very rare. You will only see that when you have a series of days where the vast majority of stocks are all declining together, day after day, for a week or more. It happens only a few times each year.

By the same token, a reading above 0.615 is very rare. It only happens once or twice a year, when the vast majority of stocks are all going up together over a series of days strung together.

They are both very rare independently. But when both of these dynamics happen over the same series of just 10 days, it has tended to be a monumentally important signal for the market.

Again, this signal was triggered on Friday, born of desperate bearishness and across-the-board selling that dominated the market from 10/18/2023 thru 10/27/2023. Then we stabilized. Then we hit positive news from the US Treasury Department in its quarterly refunding update, and then we didn't hear anything from Fed Chair Powell to spoil the incipient party.

That sent a rush of buying into the bond market. And, because interest rates have been the most important intermarket correlation factor, that opened the floodgates for buyers in the stock market.

The important part of this story is that this flood of buying interest appears, at least from this data set, to be sufficiently broad-based to trigger a rare signal from one of the most historically reliable measures around over the past 70-80 years.

 

Yeah, But, Still...

A cautionary point: 18 is not a large enough sample, just on a statistical basis, to create enormous single-factor confidence. In addition, as we covered above, this indicator is now much more widely watched.

Historically, mainstream indicator popularity is negatively correlated with continued predictive strength.

Have we passed that threshold with the Zweig Breadth Thrust indicator yet?

Time will tell. But the triggering of this signal to close last week certainly helps to foster confidence for those who have other reasons to like this market in the months ahead.