Superforecasting: The Art and Science of Prediction

Superforecasters tend to possess what the psychologist Carol Dweck calls “a growth mindset” as opposed to a fixed mindset; they believe that they can improve their potential as long as they work hard. Those with a growth mindset attend to new information that can increase their skills and knowledge.

A growth mindset is invaluable to a superforecaster. To obtain skill in forecasting, one must combine trial and error to scrutinize their technique. Without such scrutiny, experience can only take a person so far. One reason why forecasters struggle to obtain sufficient feedback is ambiguous language, whereby vague terms like “probably” and “likely” are deployed instead of numbers (181). If a forecast turns out to be wrong, such unclear language makes it difficult to go back and learn from errors. Another obstacle is hindsight bias, whereby learning the outcome of a situation distorts our perspective of what we knew before; this bias is a person’s belief that they “knew it all along” when they did not, in fact, know it all along. This bias makes it difficult to learn from mistakes. Still, like Dweck’s growth-minded subjects, most superforecasters are eager for feedback. Their willingness to continually improve makes them akin to computer programs that have perpetual beta, or trial mode.