Why are technical indicators so popular among retail investors?

Technical indicators such as MACD, KDJ, Bollinger Bands, and RSI that are common in stock trading software have almost become standard tools for retail investors. Compared with complicated and difficult fundamental analysis, these "red and green arrows" and "buy and sell signals" are intuitive, easy to understand and easy to operate. For ordinary investors with limited information processing capabilities, this "visual judgment" lowers the threshold for participation and also caters to the psychological appeal of "making decisions quickly".

More importantly, these indicators create an “illusion of certainty” by setting clear thresholds and crossover points. For example, the MACD golden cross or KDJ rebounding from a low level are given the signal meaning of "can intervene", even though the success rate of the signal has no statistical support. After one success, investors tend to "deify" technical indicators and become more dependent or even superstitious in their next decision-making.

This behavior is highly consistent with the "availability bias" in behavioral economics. That is to say, people are more inclined to believe in visible and quantifiable tools, while ignoring the statistical basis and market structure logic behind them. Technical indicators, due to their "visual visibility", have become the psychological dependence of many people in an information overload environment.

Is the indicator construction logic divorced from the real structure of the market?

Taking MACD as an example, it is based on the difference between two exponential moving averages, and buy and sell signals are constructed through the intersection and divergence of "DIF" and "DEA". It sounds rigorous, but it is essentially just a weighted smoothing of prices without introducing any exogenous variables or economic mechanisms. In other words, it doesn't explain "why prices rise or fall", it just tracks "whether price movements are continuing."

Similarly, KDJ constructs the "overbought and oversold" range based on the randomness of the stochastic indicator. It only generates signal ranges through historical high and low points, but ignores a basic fact: there is no strict functional relationship between the future path of the price and the past range state.

In financial economics, this "historical path dependence" model is widely questioned. The efficient market hypothesis (EMH) states that all available information is already reflected in current prices, so historical prices have no additional predictive value. If an indicator is constructed based solely on past prices, its predictive performance should theoretically be zero.

Furthermore, the market is the result of a multi-dimensional game. Participant behavior, policy intervention, macro variables and emergencies jointly determine asset pricing, and technical indicators mostly perform "mathematical transformations" on a single price sequence, which naturally makes it difficult to penetrate the price appearance and touch the true market essence.

Is the "signal" just a "posterior explanation" in disguise?

Looking back at the performance of technical indicators in practical applications, we will find a common phenomenon: successful cases are widely spread, while failed cases quietly sink. This "survivor bias" makes indicator signals appear to be effective, but in fact they are probably just "explanations after the fact."

For example, during a certain stage of market rise, the MACD golden cross may appear in advance many times, but the time point that actually triggers the reversal is often marked after the fact. Investors "correspond" the signal to the trend afterwards, creating the illusion of "successful prediction." However, in actual trading, this lagging signal can easily lead to buying at a high point or selling at a low point.

What’s more noteworthy is that most technical indicators are not automatically generated, but are the product of preset parameters. For example, the common "5-day moving average" and "14-day RSI" etc., the cycle selection itself is highly subjective. Different combinations of parameters can give completely opposite signals in the same chart - this exposes the "indicator signal" not as a natural law, but as a posteriori narrative of artificial rules.

In other words, indicator signals do not discover the "causal relationship" of the market, but are constantly looking for "correlation fragments" to splice into meaningful graphics. This approach is enough to comfort investors in terms of narrative, but it can easily lead to path dependence and misjudgment in decision-making.

How does behavioral economics explain "belief in indicators"?

Behavioral economics reveals that human decision-making is often deeply affected by cognitive biases and emotional drives, and technical indicators cater to several typical biases.

The first is the “illusion of control.” Technical indicators create the illusion that "signals control prices" and make investors believe that they can control the market through tools. Although the nature of market trends is chaotic, with the "blessing" of MACD or Bollinger Bands, investors have the cognitive illusion that "I can predict".

The second is "confirmation bias". Investors who use technical indicators are more likely to choose indicator signals that are consistent with their own existing views. For example, when they are bullish on the market, they are more likely to focus on the "golden cross" and ignore the "divergence"; and when they are bearish, they are more likely to magnify the risks suggested by the "dead cross".

There’s also “hindsight bias.” Once a trade is successful, investors are more likely to attribute the success to the prompt of a certain indicator and ignore other market information or luck at the time. This ex-post reinforcement mechanism causes the trust of technical indicators to be psychologically amplified, even if its predictive performance is not stable.

Eventually, the indicator itself becomes a “psychological anchor” whose function shifts from assisting decision-making to maintaining beliefs.

Do empirical studies deny the predictive value of technical indicators?

Academic evaluations of technical analysis have historically been divided. Some early studies believe that technical indicators have certain predictive power in specific market stages, especially in areas with obvious trends and low volatility. However, a large number of subsequent studies pointed out that technical indicators did not show significant excess returns in long-term backtests.

A classic study on the U.S. stock market shows that after accounting for fees, slippage and capital size, the annualized return of a trading strategy based on MACD and RSI is significantly lower than the "buy and hold" strategy. In the Chinese market, although short-term sentiment fluctuates greatly and technical indicators make it easier to capture trading opportunities, their success rate is also highly dependent on sample selection and market conditions.

Another important finding is that the "signal lag" of technical indicators causes them to act more like "following confirmers" than "trend leaders." Especially in the current era dominated by high-frequency trading and algorithmic investment, the data based on technical indicators often lags behind price behavior, making it difficult to provide a basis for real-time decision-making.

In addition, some empirical studies point out that in markets where technical indicators are widely used, their effectiveness decreases. In other words, when everyone looks at RSI, RSI becomes invalid. This "indicator reflexivity" reflects the market's high adaptability and also reveals the fundamental reason why technical indicators cannot provide a long-term "advantage structure."

Do trading signals still have instrumental value?

Are "indicator signals" completely useless despite their many flaws and biases? not necessarily.

In actual investment, if technical indicators are used as "market rhythm sensing tools" rather than "mechanical trading instructions", they still have certain auxiliary value. For example, MACD can be used to observe whether momentum is increasing and whether the trend is slowing down; Bollinger Bands can help determine the convergence or expansion stage of market fluctuations. The key lies in whether the user can put the indicator into a broader analysis framework and understand it, rather than using it as a buying and selling criterion in isolation.

Furthermore, in quantitative investment, technical indicators are often used as one of the "signal characteristics" to build models together with fundamental factors and industry factors. This approach emphasizes "information fusion" rather than "indicator dominance" and avoids the oversimplification of "single factor signals".

For ordinary investors, the real value of indicator signals may not be to predict the market, but to help them establish trading discipline and review paths. In a market full of noise, even an imperfect signal system can maintain psychological stability better than a completely random decision-making method.

But the premise is: we must know that it is a tool, not a truth; a reference, not an instruction; a part of the map, not the direction itself.