Is trend your friend?

By Piotr Zielonka and Krzysztof Szymanek
Posted on April 25, 2017

The quality of our life depends largely on our ability to predict.
We are all eager to know what the future hold. The talent to anticipate is highly valued in all cultures and societies. 

Clairvoyants employ tricks designed to convince their clients about their extraordinary abilities. Seers continually predict occurrence of spectacular events, knowing that in case of a hit-coincidence, the audience will not take into account their of previous mistakes. Seers predict mostly catastrophic events, by which they attract attention. Visions of the future are therefore formulated vaguely and often cannot be tested. Sometimes seers 'predict' an event that occurred earlier. This is called a Texas sharpshooter fallacy. The shooter fires in the direction of a barn and then draws a circle around the tightest cluster of hits.  The observer is thus given the impression that the shooter will hit the exact same point each time.

“People differ in the way they make predictions.”

They may believe in a trend continuation or in trend reversal. The first attitude illustrates a well-known stock market brokers’ saying: “Trend is your friend”. An expectation that the observed trend will continue is the most common strategy for naive forecasting. As studies show, to opt for the prediction of a trend reversal, one has to make a significant effort to go beyond the automatic perception of reality. For instance, people who repeatedly have had bad luck, may remain in the belief of the trend continuation. Apparently there once lived a man who succeeded with nothing, who sat on freshly painted benches, bought defective cars, and occasionally was bitten by a dog, the dog was rabid of course. However, one day, he incidentally dropped his sandwich and .... he couldn't believe his eyes. A slice of his sandwich had fallen butter side up. His luck had changed. He went to the relevant expert, asking him to explain his change in luck. The expert after much thought, said: "Well, the trend of your woes may not be reversed. Perhaps you buttered the wrong side of the bread”. 

"An expectation that the observed trend will continue is the most common strategy for naive forecasting."

Taking this into account, after facing a series of similar events be it good or bad, people sometimes opt to believe in trend reversal. This attitude is illustrated by the stock market adage: “What goes up, must come down”. According to HerodotusAmasis (ruler of Egypt in 6 BCE) noticed that his friend, Polycrates (ruler of Samos in 6 BCE) was too successful. In Polycrates’s life good fortune never alternated with misfortune. Good luck followed him everywhere. Amasis had never heard of anyone so fortunate. Thus he advised Polycrates to choose what he valued the most; in other words, what would draw the most regret from him, when lost and asked him to throw this away in order to evoke misfortune in a controlled manner. Polycrates obediently followed the advice and threw his most precious ring into the sea. A week later, a fisherman caught a fish and shared it with the Polycrates. While having a dinner, Polycrates discovered the ring inside the fish. Amasis then learned that a person is incapable of reversing the trend of events in order to alter their fate.

“An important question arises: what circumstances make people predict trend continuation or trend reversal?”

It turns out that in order to make prediction people first try to guess the nature of the source of uncertainty. Be it deterministic or random. If an individual assumes that a sequence of events is deterministic, they expect trend continuation. Basketball fans often believe that a player who has just put the ball into the basket has a greater chance of further success in following attempts than a one who has not. However, if an individual believes that the nature of an observed series is random, they mostly expect a trend reversal. Gamblers have a belief that in a game of roulette after the ball falls in black several times in a row, one should definitely bet against black, reasoning that a long streak of blacks has to be eventually followed by red. Of course, the ball hardly remembers what it did before. In the case of independent events no prediction strategy really works. One should then predict the more frequent event.

What is interesting, even if observed events follow a random series, people are prone to convince themselves that these are patterns having predictive value. In laboratory experiments, participants are reported to have found patterns in purely random sequences of stock prices. This being technical analysis. It is the examination of past price movements to predict future price movements. It is more an art than an objective analytical tool. Technical analysis is therefore open to interpretation. Two technicians will look at the same chart and paint two different scenarios or even see two different patterns. The only commonality would be that both technicians would not be able to come up with a logical support justifying their positions.  It seems that the reason why technical analysis is such a popular tool for stock market predictions is the spurious relationship between its signals and real stock markets price fluctuations. 

Spurious relationship is a common fallacy of human cognition related to forecasting. For example, do you remember Paul, the octopus which was able to predict the majority of game results (11 out of 13) during the Soccer World Cup 2010? Before each match Paul was placed in front of two boxes containing food ornamented with flags of the rival soccer teams. The box Paul chose first would stand as his prediction of which team would win the game. Before each match his prediction was announced. It turned out that Paul achieved a 85% accuracy rate. 

We should say that Paul did not predict but actually successfully guessed the results of the matches. Prediction should be justified before the time comes, to which it refers, to predict. It must have an objective basis, indicating a high probability for its implementation. Guessing turns out to be accurate only at the moment when it is fulfilled. It seems important to settle whether we predict or blindly guess.

Even if we refrain ourselves from predicting or guessing, we are very likely to fall into another trap, called hindsight bias. Whatever is highly unlikely but then happens, is accompanied by a statement: “It was to be expected”.  This type of thinking is based on the erroneous assumption that the future can always be predicted based on the events of the past.
In many cases we cannot predict the future, but it does not stop us from trying. What we really can do is to reject prognosis which are passed on doubtful assumptions and seem more guessing than forecasting.