On The Merits of Technical Analysis

Technical analysis is astrology for men. 

Attempting to predict future asset prices based on past prices and data, is equivalent to predicting how successful a relationship will be based on star signs.

1. Humans recognize patterns, even when there are none. 

There’s a name for this, pareidolia, which is defined as: the tendency for perception to impose a meaningful interpretation on a nebulous visual stimulus (so that one sees an object, pattern or meaning where in fact there is none).

Humans are extremely pareidolic. Consider the picture below on the left. Most of us will immediately see a face in the smoke, when obviously, there isn’t actually one. The smoke is basically random, yet we impose meaning in it. 

The same thing applies to trading. When people see graphs like the above right, they are prone to see patterns, even when there aren’t any meaningful ones. Yet technical analysts would call this a “descending triangle”-pattern or whatever, and attempt to use it to predict future price. 

Even if stock prices completely followed a random walk, people would be able to convince themselves that there are patterns having predictive value. In laboratory experiments, subjects are reported to have found patterns in purely random sequences of stock prices [1].

This great video by Veritasium really showcases just how prone people are to superimposing their preexisting beliefs on patterns.

Technical analysis should be called pareidolia analysis, because that’s exactly what it is. 

2. Luck makes superstition.

Technical analysis occasionally works. The problem however, is that successful price predictions from technical analysis are entirely due to luck. When these technical analysts get lucky, it will confirm their preexisting beliefs, such that they become even more convinced of the merits of technical analysis. This is a perfect example of confirmation bias. When these traders are rewarded completely at random (by luck), they will continue their behavior as though it was what caused their success. 

In one famous experiment,  pigeons were placed into cages, wherein a food hopper attached to the cage may be swung into place so that the pigeon can eat from it. A timing relay holds the hopper in place for five seconds at each interval. If a clock is now arranged to present the food hopper at regular intervals with no reference whatsoever to the bird’s behavior, operant conditioning usually takes place. The bird tends to learn whatever response it is making when the hopper appears. For example, if the bird was preening itself when the food appeared, they will continue preening itself in hopes of getting more food. Another bird may have been pecking the ground when the food appeared, and so will continue pecking the ground expecting food [2].

I argue that the same exact phenomenon happens with technical analysts. When these traders get randomly rewarded after spotting a “bullish triple bottom”-pattern, they keep looking for that pattern hoping for another reward. They attribute successful trades to skill, when in fact their successes are entirely the result of chance alone.

3. Technical analysis has zero predictive power. 

Almost every study on the profitability of technical analysis finds the same thing: mixed results. Technical analysis worked on some time periods but not others. Technical analysis worked on some markets but not others. Some technical analysis patterns work but not others. Technical analysis showed profitability in some studies, but not in others. All in all, the results are what you’d expect from randomness [3][4][5][6][7][8][9].

And if you want random results, you should consider gambling instead, because then you’re not fooling yourself about what you’re doing. The profitability of day trading alone is questionable [10], and technical analysis certainly doesn’t seem to help [11][12]. In fact, scientific experiments have found that belief in technical analysis is associated with cognitive biases [13].

There is no technical analysis method that has even been shown to work consistently over time. So whenever someone presents a graph with some lines on it, as evidence for future price, it should clue you in to the fact they have absolutely no idea what they’re talking about, whatsoever. 

Technical analysis is cognitive biases, pareidolia, and sheer ignorance mixed into one.