Sentiment is a factor that significantly influences the market, yet it is difficult to analyze, partly because it is poorly anchored to statistical models. Sentiment Analysis is certainly the most challenging of the three disciplines that make up market analysis (technical, fundamental, and Sentiment). In this article, we will provide an overview of Sentiment and offer practical guidelines for analysis.
What is Sentiment
The term, from this point of view, is revealing. The word Sentiment evokes concepts such as emotions and feelings, and indeed, that's what it's all about. Sentiment refers to the whole range of emotions that often irrationally guide investors and therefore profoundly impact prices. Now, each trader is unique, with their own character and emotional sphere. However, some reactions are shared and therefore objectively analyzable, somewhat like market movers.
There can be no doubt about the power of these reactions, partly because sometimes markets undergo oscillations too significant to be attributed to purely technical factors. It's evident that often, something else is at play, perhaps seemingly underground. Emotions and feelings, precisely. Among the most frequent, the following stand out.
Excitement. This sentiment occurs when market performance exceeds expectations, and investors, caught up in euphoria, buy excessively, supporting prices. These movements, if prolonged, generate bubbles which, when they burst, create chaos and depression in the market.
Depression. It occurs, precisely, when investors are seized by a sense of disappointment, when the market performs far worse than expected. At this point, worried about the downward trend, investors sell. If sales reach a critical mass, depression turns into fear, and fear degenerates into panic.
Disorientation. It almost always occurs during periods of volatility, perhaps after the worst has passed, but the market is still in a settling phase. Investors don't know what to do: sell, buy, or hold their position?
The main characteristic of Sentiment is that it generally develops in the very short term. Emotions explode, involving ever-larger groups of investors in a domino effect, then fade until they give way to rationality. For this reason, Sentiment analysis, if not useless, assumes less importance if the trading horizon is medium or long term. The truth is that price oscillations generated by reactions tend to be offset within a few weeks.
However, if you are an intraday trader, Sentiment Analysis must be your daily bread. Hence the need to understand how it works and how to practice it.
While realizing that Sentiment has an effect on prices is quite peaceful, it is less easy to understand how big this effect is. It would be naive to think that the market, in a given period, is influenced only by technical factors and rational motivations or, perhaps in a different period, only by emotional and irrational factors. At first glance, to understand the extent of Sentiment, one can compare the intrinsic (estimated) value of an asset with the value it assumes in that given period. If the discrepancy is significant, Sentiment has played a larger than normal role.
To understand how Sentiment acts and its effect on the market, it is useful to give an example with stocks, which are among the assets most susceptible to the phenomenon. The expression "stock market panic" is known, after all, even by laypeople. Let's imagine a rising market: prices increase, and of course, the percentage of investors buying stock X also increases. If the price of stock X exceeds a maximum or a psychological threshold, even investors who did not intend to buy X will do so, further supporting the price. The perception will then spread that X is destined to rise for a long time to come, a perception that strengthens over time and overshadows the elements that, on the other hand, would suggest the end of the upward trend or even a reversal. In simple terms, the asset performs beyond its real merits. It is supported not so much by technical elements but by the favorable sentiment of investors.
Is Sentiment predictable?
In general, common sense tends to think of emotions as something spontaneous and unpredictable. Regardless of whether this belief is true or false, does it also apply to the market? The answer is no. Sentiment, within certain limits, of course, can be predicted, calculated, and exploited at will. It is objectively difficult, but it can be done. However, it is necessary to understand what factors influence Sentiment. Understanding what influences Sentiment means predicting, even here within certain limits, how prices will move.
So, what influences Sentiment?
Information. Meaning, of course, specialized information. News can influence Sentiment; in fact, it is the most powerful factor of all. After all, and this applies in all fields, information influences public opinion. And the same goes for the market. Now, the news that, more than any other, moves investors' minds is that which is confused - certainly legitimately - with opinions. If a commercial bank publishes a paper in which it declares itself very pessimistic about the future of a particular asset, investors will react accordingly and start freeing their portfolios of those assets.
Experience. Meaning, this is obvious, not in an individual sense but a collective one. It is shared experience that generates shared reactions. If investors appreciate a phenomenon that has generated chaos in the past, they will try to defend themselves as "one man," perhaps by selling a certain security en masse. From there, it's a short step to generating a large-scale domino effect.
The
context. The external environment also influences, i.e., events, phenomena, and problems that concern not so much the market as what gravitates around it. It is evident that precarious contexts make investors more susceptible. One of the many examples in this regard concerns the debt crisis of 2011. If the economies of Italy and Greece had not been in recession, and compromised by a precarious political situation, would the debt crisis have exploded with such virulence? Probably not.
Studying Sentiment
As we have already specified, Sentiment Analysis is more complicated than technical analysis and even fundamental analysis. There are no precise statistical models to refer to, but this does not detract from the fact that it nevertheless enjoys elements of tools that make it somehow objective.
In simple terms, Sentiment Analysis also has its indicators. Certainly very different in constitution and methods of use from those of technical analysis, but still able to generate signals.
An important indicator is given by
indices. These, in fact, represent a summary of the performance generated by a basket of assets, almost exclusively stocks. Within it, some assets perform well, others less so, but the overall result still signals a direction. Monitoring the most important indices, therefore, is one of the essential tasks of those who do Sentiment Analysis, as they allow us to understand if the atmosphere is changing or taking on excessive turns. Among the most important indicators is the DAX, which offers a representation of the German stock market. It is also useful, above all for Forex traders, as it is positively correlated with the euro-dollar. If the DAX rises, the euro-dollar rises; if the DAX falls, the euro-dollar falls.
An equally important indicator is represented by the open and long-term positions
of some large international operators. In truth, this type of indicator, rather than offering an overview of the current Sentiment, offers a rather reliable signal of what the Sentiment will be in the immediate future. This is true because the mass of retail investors, who together are able to move large sums of money and influence prices, generally move after the big investors, almost like a delayed pursuit. Where can information on open positions be found? In the COT Reports. Large investors, in fact, according to anti-Insider Trading regulations, are required to provide precise information on their movements. This information is public, so it can also be used by common traders who engage in Sentiment Analysis.
Finally, surveys. What better way to understand what investors think or fear than to ask them directly? Obviously, it is not possible to grasp the specific Sentiment of a huge number of investors, but selective sampling is very useful. Here are the surveys most used by Sentiment analysts.
AAII Bull and Bear. Conducted by AAII, it is a survey that involves a certain (variable) number of Wall Street investors who are asked the same question. The survey results are published in many specialized publications, including the famous Bloomberg.
CNN Fear and Greed Index. The survey is more complex as it is not only the result of an investigation but also of data processing that aims to derive an index of "greed." If the result is close to 100, for example, it means that investors are nervous and impatient, therefore inclined to move according to emotions and not reason. If the index is close to 50, it means that the situation is stationary and emotionally neutral.
XTB Market Sentiment. It is a composite and very extensive survey offered by the xStation 5 platform. It is automatically processed based on some market evidence.
The right approach to Sentiment Analysis
We have seen what Sentiment is and how it is studied, focusing on some indicators in particular. It is now time to provide some general advice on how to use this important tool, on the attitude to have towards it, on the most correct approach to use. A tool, in fact, can be useful, but if used superficially or, conversely, if abused, the risk is that it can do more harm than good. Here then are some general indications on the timing and methods of use of Sentiment Analysis.
Don't use it often if trading is long-term. When collective and shared, investors' feelings and emotions have a very strong impact on the market. It's now established. The main point is the following: for how long? Well, for a limited period, unless events occur that are so important and decisive as to produce a truly lasting influence. In the long term, if not even in the medium term, the market tends to return to the ranks, thus reabsorbing excessive movements caused by irrational reactions. For this reason, long-term traders, or even swing traders, would do well not to waste too much energy on Sentiment Analysis. The latter is instead a resource to be used intensively by traders specializing in intraday and especially scalpers.
Never use it alone. Sentiment Analysis, certainly when it is useful to do it (see previous point), is only one pillar of a larger construction, such as market analysis as a whole. Sentiment analysis is bound to produce unreliable signals if it is not accompanied by fundamental analysis and technical analysis. Technical analysis, for example, serves to understand what the psychological thresholds are that can generate emotional reactions (the reference is to supports and resistances). Fundamental analysis, on the other hand, provides an overview of the context and generates information on market movers, which in turn can produce waves of euphoria or waves of panic. In short, everything is connected, even when it comes to trading. Therefore, never use Sentiment Analysis alone, but relate it to technical analysis and fundamental analysis.
Don't trust surveys too much. Statistics is a perfect science, but only in one sense: it gives an exact evidence of probabilities. Precisely, it offers probabilities, which do not necessarily indicate a truth or an exact prediction. Surveys need to be treated for what they are: a source of indications, often broad. Surveys, in fact (and this applies to all surveys, not just those concerning trading), have a more or less wide margin of error that depends on the intrinsic limitations of sampling.
