In the never ending search for a potential market top, some analysts use sentiment analysis. This type of analysis relies on sentiment indicators which are tools that are designed to identify how investors feel about a market. Sentiment is independent of fundamental information and price data and is based solely on feelings, which can provide significant information not factored into the fundamental or technical picture.
Sentiment analysis is based on logic. The general idea is that when everyone is bullish, everyone is fully invested in the stock market and the market is unlikely to go much higher since there aren’t many buyers waiting to push prices up. In essence, sentiment analysis tells us that an excessive number of bulls means a large number of market participants are fully invested and there isn’t enough money available to keep pushing prices up.
Likewise, when too many investors are bearish, there are most likely a large number of participants who have completed selling. In that case, the market is running out of sellers and a bullish reversal is expected.
Because sentiment is a feeling, analysts generally rely on surveys to obtain data they can use in their trading decisions. Examples of sentiment indicators include surveys of individual investors, newsletter writers and investment advisors to determine if they are bulls or bears. These surveys include questions to determine if the survey participant think the market will go up or down. Usually the question includes a time element, such as “I feel that the direction of the stock market over the next 6 months will be:
Up — Bullish
No Change — Neutral
Down — Bearish”
There are other examples of sentiment indicators including commitment of traders (COT) data and data from the options markets including the put-call ratio.
COT data applies to the futures markets and tells investors what different groups of investors are doing the market. All positions are assigned to commercial participants, large speculators and small speculators. With that data, analysts can determine whether or not a price trend is likely to continue. COT data can be difficult to interpret as the chart below shows.
The chart above shows that commercials (the green line at the bottom of the chart) are now holding a short position in the market while both small (the red line) and large speculators (the black line) are bullish. In futures markets, the commercials are considered to be the smart money while speculators are then the dumb money. Based on that, the current data is slightly bearish for stocks.
COT data can offer extremely timely signals as it did in early 2016. As the chart above shows, commercials were unusually bullish at that time while small speculators were increasingly bearish as the market was forming a bottom. In this case, smart money was right and dumb money was wrong. Sentiment data would have helped identify an important market bottom.
COT data is available at the Commodity Futures Trading Commission (CFTC) web site. For the stock market, analysts review COT for futures contracts on the S&P 500, Russell 2000, Nasdaq 100 and Dow Jones Industrial Average indexes. Gold traders or investors accumulating precious metals may find the data from the gold and silver market to be useful.
In the options markets, we can also measure market activity to assess sentiment. Put options are positions that benefit from a market decline and call options benefit from an increase in prices. The put-call ratio compares the interest in puts and calls in an effort to identify whether traders are bullish or bearish.
Some analysts apply a simple approach to sentiment analysis and try to identify extremes in the data. To see how this works, we will consider a survey that is readily accessible, the American Association of Individual Investors (AAII) Investor Sentiment Survey. This is a weekly survey asking individuals whether they are bullish, bearish or neutral over the next six months. The most recent data is available at the AAII web site for free. Results are summarized in a diagram and historical data is available as a spreadsheet that can be downloaded.
From the diagram, we can see than generally about 40% of individuals are bullish, 30% are bearish and 30% are neutral.
Some analysts simply look for extremes and believe that if, for example, most responses are bullish the market is near a top. A significant move could see more than 70% of investors with the same opinion. This approach will rarely provide a signal.
Researchers have found there are other times the data provides signals. One study found, “the best contrarian signal occurred not when investors were unusually optimistic or pessimistic, but rather when they described themselves as being neutral (defined by the survey as expecting stock prices to be unchanged over the next six months). Unusually high levels of neutral sentiment have been followed by a median 26-week rise in the S&P 500 of 8.6% and a median 52-week rise in the S&P 500 of 17.7%. In contrast, the S&P 500 had a median return of 5.2% and 10.7%, respectively, over all 26-week and 52-week periods throughout the survey’s history.” The study also noted that large cap stocks were up 83% of the time six months after neutral sentiment reached an extreme. A year after neutral readings reached an extreme value, large caps were up 88% of the time.
That is an important insight. Indecision is the best indicator to watch for. When a large number of investors are neither bullish nor bearish, the market tends to deliver the best results. This is actually in line with the popular saying on Wall Street that the market’s job is to confound the maximum number of investors at any given time. When a large number of investors have no idea what will happen, stocks tend to soar. This ensures the maximum number of investors as possible will be wrong.
In order to be useful, we need to define what a high reading is.
AAII has conducted the survey every week since July 1987, providing nearly thirty years’ worth of history. This is a data set that lends itself to statistical analysis and one tool that has proven to be useful is the standard deviation. For neutral readings, the average reading is 31% with a standard deviation of 8.7%. This indicates neutral readings above 40% are generally bullish.
Other important levels in the AAII survey to watch are 40% for the bears and 50% for the bulls.
Sentiment analysts also have access to other surveys. Investors Intelligence has published the Advisors’ Sentiment Report since 1963. This report surveys the market views of over 100 independent investment newsletters (those not affiliated with brokerage houses or mutual funds) and reports the findings as the percentage of advisors that are bullish, those bearish and those that expect a correction. The report is interpreted as a contrarian indicator. A large number of bulls is considered to be bearish and a high bearish reading is considered bullish.
While the Advisors’ Sentiment Report is useful, it is a subscription service. Many investors find the information in the free AAII Sentiment Survey provides enough information for their sentiment analysis. This data can be supplemented with the free data available in the COT report.
Sentiment analysis should be combined with fundamental or technical analysis rather than used as a stand alone trading strategy. In this way, it can supplement your market research, providing rare signals independent of the price action.