SentimenTrader ---- A quick pullback recovery, overwhelming up volume, overbought financials

 

TradingEdge Weekly for Aug 30 - A quick pullback recovery, overwhelming up volume, overbought financials

Key points:

• The average S&P 500 stock has staged an impressive pullback recovery
• During the recovery, advancing volume swamped volume flowing into declining stocks
• The Correlation Pattern Match tool shows what happened after similar 18-day stretches in the S&P 500
• Looking at a long-term cycle in stocks
• Avoiding the "ugly 17" months during cycles has proved to be a boon
• Financials are overbought, and that's a good thing
• Real estate has struggled during September
• Looking for clues in gold based on short-term returns

An amazingly quick pullback recovery

The persistence and breadth of the August rally allowed the equal-weight version of the S&P 500 to close at fresh record highs, erasing its pullback.

Dating back to 1957, this was one of the fastest-ever roundtrips from a pullback. The index took only 23 trading days to go from a record high to at least a -5% pullback and back to a new record. There were only four other times it made such a quick recovery, and each of them was a good sign for the index. 

To generate a larger sample size, the table below shows recoveries from pullbacks that took a bit longer but still less than two months. While returns were still generally positive, the risk/reward was uninspiring. On the plus side, only three signals showed meaningful and persistent losses over the medium to long term.

Regardless of what the much more popular capitalization-weighted index was doing then, these quick recoveries in the equally-weighted index tended to precede outperformance for the equal-weight version. Over the next two months, the equal-weight version of the S&P sported a higher return than the cap-weight version 75% of the time.

Advancing volume surge

Volume in advancing stocks on the NYSE outpaced volume in declining stocks by a ratio of 10/1. Dean showed that similar volume skews near a high produced a 93% win rate for the S&P 500 over the next six months.

Our data shows a ratio exceeding this level on 207 occasions since 1950, when volume data became more reliable. However, context is essential. When the S&P 500 was 1% or less from a 3-year high, the number of precedents drops to 15.

Following precedents when the NYSE up-down volume ratio surpassed 10/1, with the S&P 500 less than 1% below a 3-year high, the world's most benchmarked index tended to maintain its upward bias, rising 93% of the time over the subsequent six months. 

Over the next six months, the signal encountered a maximum loss exceeding -5% in 4 out of 15 cases, with no instance reaching a -10% loss, suggesting that a significant correction is less likely.

Growth-oriented sectors like Consumer Discretionary and Technology outpaced all other groups and the S&P 500 over the favorable six-month period.

Matching patterns

The Correlation Pattern Match tool now includes a trend filter, adding valuable context to the analysis. Dean pointed out that a CPM study found 85 other S&P 500 periods that mirrored the last 18 sessions with a 98% correlation.

This tool excels in price pattern analysis, allowing users to assess the likely outcome of historical patterns from a risk/reward perspective. Given this new enhancement, let's examine all other periods that match the previous 18 sessions for the S&P 500 with a 98% correlation when the S&P 500 traded above its 200-day average. The CPM displays a blue background on the chart wherever pattern matches are detected.

Following a 98% correlation match to the last 18 sessions when the S&P 500 exceeded its 200-day average, the world's most benchmarked index displayed a consistent upward bias over the following year. While the 1-year time frame is the only interval to show significance relative to random returns, several other horizons exceeded 1.5, suggesting a favorable environment but not an overwhelming edge.

Suppose we examine precedents since 1989. In that case, the current price pattern has achieved more favorable returns and consistency, with several win streaks surpassing ten consecutive gains. 

In the following year, the S&P 500 increased 94% of the time, suggesting that these sharp price accelerations above the 200-day moving average were associated with sustainable long-term uptrends. As always, one-year time frames are susceptible to some volatility in between, and this one is no different, with nine maximum losses exceeding -10% since 1989

Stocks' 20-year cycle...

Jay is on vacation this week but re-ran Part II of a series on the 20-Year Cycle in the stock market, which originally ran on 2021-02-11. It highlights a 42-month period that occurs every 20 years, starting on 1902-09-30.

Part II of our 20-year pattern starts on September 30th, 1902. From this date, we will look at the next 42 months as a seasonally "favorable" period. Then we will progress in 20-year intervals. In other words:

• Sep 30, 1902 to Mar 31, 1906
• Sep 30, 1922 to Mar 31, 1926
• Sep 30, 1942 to Mar 31, 1946
• Sep 30, 1962 to Mar 31, 1966
• Sep 30, 1982 to Mar 31, 1986
• Sep 30, 2002 to Mar 31, 2006
• Sep 30, 2022 to Mar 31, 2026
• Etc.

The chart below displays the cumulative results of buying and holding the Dow ONLY from Sep 30th of "even" year 2 through Mar 31^st of "even" year 6, from 1899 to the present.

Cumulative Dow Price performance during the 42-month favorable period through 12/31/2019 is a gain of +1,756%.

...And an ugly 17 months

Jay further revealed some months that are consistently poor performers. That's the bad news. The good news is that by avoiding these particular months, an investor might be able to gain an edge.

The months we will look at are what I refer to as the "Ugly 17". These are 17 months out of every 240 months (i.e., 20-year cycles starting in 1900) that have tended to witness poor stock market performance cycle after cycle.

The first month in the "Ugly 17" is "September of Year 0." The others are:

Now let's consider the performance of the Dow ONLY during these "Ugly 17" months as a whole. The chart below displays the cumulative % +(-) for the Dow Jones Industrial Average ONLY during these same 17 months during each 20-year cycle starting in 1900.

The chart below displays the cumulative % return for the Dow Jones Industrial Average since the end of 2019 for both the "17 Ugly Months" (blue line) and "All Other Months" (black line). From 2019-12-31 through July 2024:

• All Other Months have gained +125%
• The "17 Ugly Months" have lost -24% 

The next "Ugly Month" in the sequence is May 2025.

Overbought, and that's a good thing

Dean noted that more than 87% of S&P 500 Financial sector stocks reached an overbought condition, with the index at a high. Similar overbought scenarios produced an 87% win rate for Financials over the subsequent six months.

On Monday, more than 87% of S&P 500 Financial sector stocks notched an overbought stochastic indicator, with the sector closing at a new high, a scenario that has occurred 15 other times since 1958. As shown in the chart below, the favorable price momentum tends to continue, resulting in an impressive annualized return of 63.5%. 

Following precedents when more than 87% of S&P 500 Financial sector stocks achieved an overbought status, with the index closing at a 3-year high, the sector tended to maintain its upward trajectory, rising 87% of the time over the subsequent six months. 

Over the next six months, the signal encountered a maximum loss exceeding -5% in 6 out of 15 cases and -10% in 4 out of 15 occurrences. During the favorable six-month window, S&P 500 Financials outperformed the S&P 500 in five out of six intervals.

Real estate in September

Real estate has historically been a weak sector during September. 

The chart from Jay below displays the cumulative total return for the real estate sector if held only during September starting in 1945 through 2021.

The trouble in the real estate sector often does not begin until later in September and sometimes bleeds well into late fall. The chart below displays the annual seasonal trend for ticker IYR (iShares U.S. Real Estate ETF), which tracks the Dow Jones Real Estate Index. As you can see, there has been a tendency for weakness between trading Day of Year (TDY) #179 and TDY #226.

2007 and especially 2008 accounted for most of the net loss. Nevertheless, the more significant point is that not much of anything great happened during this period in other years.

Watching gold traders' reactions

Gold's rise has captured more investors' attention, and it's impacting sentiment levels.

The Optimism Index powers gold's Risk Level. Over the past 20 days, that Risk Level has averaged 7, the 2nd-highest in five years. The last time it reached this level coincided with the 2020 peak.

When the 20-day average Risk Level reached 7 for the first time in months, it wasn't a clear-cut sell signal. Three times, this triggered right before substantial runs in the metal, highlighted with green boxes in the chart below.

The table below shows returns in gold after these signals. There were three dates with double-digit gains either six or twelve months later, in 2007, 2016, and 2019. All the others showed losses, at least over the next few months.

What sticks out is that the three signals that witnessed substantial gains all witnessed minimal selling - if any - after the 20-day Risk Level first crossed 7. Even a month later, all three saw no more than a -0.76% drop in gold as buyers never let up. All the other signals saw more significant losses than gains, or close to it.

This suggests that if buyers continue to bid up gold in the weeks ahead, we have some compelling evidence that this will be one of the exceptions when other factors overwhelm sentiment and potentially lead to sustained gains.

After these gold signals, the S&P 500 did quite well. Over the next year, it rose 87% of the time by an average of +16%. Gains were widespread among its sectors, with financials among the more consistent beneficiaries-which is a bit surprising, given assumptions that big gains in gold come ahead of economic trouble, where financials are usually among the hardest hit.

About TradingEdge Weekly...

The goal of TradingEdge Weekly is to summarize some of the research published to SentimenTrader over the past week. Sometimes there is a lot to digest, and this summary highlights the highest conviction or most compelling ideas we discussed. This is NOT the published research; rather, it pulls out some of the most relevant parts. It includes links to the published research for convenience, and if you don't subscribe to those products, it will present the options for access.