"It's hard to make predictions – especially about the future," as the old Danish proverb popularized in other countries by physicist Niels Boer says – but it seems that "experts" of various kinds, such as television commentators, tend to fail in particular. Why is this happening, and why aren't they learning a lesson? We used Claude and Forefront to provide a possible answer to the question.
Superforecasting: The Art and Science of Prediction, by Philip Tetlock and Dan Gardner, published in 2015, offers a fascinating look at what separates successful forecasters from everyone else through extensive research on forecasting tournaments run by the US government's Advanced Intelligence Research Projects (IARPA).
The authors identified certain traits common among "superforecasters"—people who consistently outperform others in predicting geopolitical and economic events. One of the book's key insights is that prediction is limited by a variety of biases and cognitive limitations. Even the most knowledgeable experts in a particular field may find it difficult to accurately predict the future.
Making predictions involves dealing with uncertainty, and taking into account a number of complex factors that can affect how events occur. Our brains are not naturally wired for this type of probabilistic, scenario-based thinking. Instead, as the authors explain, we have tendencies to ignore information that contradicts our existing theories, to be too sure of our views, and not to change our beliefs when new data emerge.
The book identifies some of the mental traits that help superforecasters overcome these cognitive traps: they tend to approach problems from a position of curiosity, with an open mind rather than "confirmation bias" (a tendency to refer only to facts that support the preferred conclusion); They are comfortable dealing with numerical probabilities instead of making final statements; And they have a "growth pattern," in which they treat mistakes not as personal failures, but as opportunities to learn and improve their methods.
The two emphasize that the media encourages absolutism over presenting various probabilistic possibilities. That's because announcing decisive predictions with false confidence, no matter how small the probability of them, attracts an audience. An admission of complexity is not "photographed" well.
The authors compare the different types of commentators to foxes and hedgehogs: foxes, who approach a topic with a variety of details of knowledge and try to connect them to several plausible possibilities, versus hedgehogs, who arrange all ideas around existing ideas in order to reach the conclusion they tend to decide on in advance. They recommend that experts adopt self-criticism, searching precisely for ideas that contradict their own, presenting various options with relatively high probability, and updating forecasts frequently as the data changes.
Wrong? We'll fix it! If you find a mistake in the article, please share with us