Can I predict the future?

Episode Summary

Executive Summary: The episode investigates whether “super forecasters” are real and what makes them accurate. It explains that forecasting skill comes less from raw intelligence or expertise and more from curiosity, open-mindedness, humility, analytical thinking, probabilistic reasoning, and using outside views and base rates. The hosts test these ideas through riddles, a public prediction challenge, and a live forecast on a Tarantino film, showing prediction can be learned and improved.

Main Topics: What a super forecaster is (Priority: 5/5): A super forecaster is someone who predicts future events far better than most experts, by analyzing evidence and revising judgments as new information arrives. Why experts often fail (Priority: 5/5): The episode highlights Philip Tetlock’s finding that many professional forecasters were no better than chance, showing that expertise alone does not guarantee accurate prediction. Traits of strong forecasters (Priority: 5/5): Curiosity, intellectual humility, open-mindedness, willingness to update beliefs, and recognition of bias are presented as the key psychological traits behind forecasting success. Analytical and probabilistic thinking (Priority: 4/5): The host is tested with trick questions to show that good forecasting requires careful reading, not intuition, and that estimating odds is central to accurate predictions. Crowd forecasting and collective intelligence (Priority: 5/5): NESTA and Good Judgment’s crowd prediction work suggests groups can outperform individuals by pooling information and reducing bias, especially in uncertain political contexts like Brexit. Learning to forecast better (Priority: 4/5): The episode argues forecasting is partly trainable: modules on base rates, outside view, and cognitive bias can improve decision-making in both public and personal life. Live forecast example: Tarantino box office (Priority: 4/5): Greg makes a forecast about Once Upon a Time in Hollywood’s opening-week revenue, then compares it with a top super forecaster and the actual result.

Key Arguments: Super forecasters are not psychic; they are unusually good at evidence-based probability judgments. Raw intelligence is not enough for forecasting; it can even increase confidence in incorrect conclusions if not paired with humility and self-correction. Good forecasters treat beliefs as hypotheses that should change with new evidence. Crowds can outperform individuals because different people contribute different pieces of information and cancel out individual bias. Outside-view thinking and base rates are more reliable than relying only on immediate impressions or personal intuition. Forecasting skill can be improved through training and practice, not just innate talent.

Data Points: Top performers in Good Judgment Project: Top 2% - Participants in the US-funded forecasting tournament who were labeled super forecasters Forecasting trial length: 4 years - The Good Judgment Project ran forecasts over a four-year period Crowd accuracy example: 1.17 euros per pound - Crowd forecast for the euro-pound exchange rate on 1 April 2019 matched the actual value closely Crowd probability on exchange rate: More than 50% - For 62 of 67 days, the crowd assigned over 50% probability to the 1.1–1.2 euro range Number of options in exchange-rate question: 5 options - Used to explain why a 50% probability was much better than random Random baseline for exchange-rate options: 20% each - Each of the five choices would have had an equal chance if guessed randomly Forecast training duration: About 1 hour - The online super forecaster training course Greg completed Host prediction confidence: 45% and 35% - Greg assigned probabilities to two box-office bins for the Tarantino film Michael Story’s comparison forecast: 40% to 60 million USD bin - Michael judged that box-office range as most likely Actual box office result: Just above $40 million - The film’s opening-week gross, placing it near the boundary of the lower and middle bins Crowdscience listener prediction outcome: Top 50% - Greg’s forecast was good enough to be above median on the question Galton ox-weight example: 1 pound away - The crowd average guess at a country fair in 1906 was nearly exact Forecasting quiz result: Jack is married and looking at Anne - Used to answer the riddle about whether a married person is looking at an unmarried person

Pivotal Quotes: "A super forecaster is someone who predicts events far better than the experts." — Cicely: The listener’s initial understanding of the concept prompting the episode "They are somewhat distinctive psychologically... super forecasters tend to see their beliefs as testable hypotheses that should be revised in response to evidence." — Philip Tetlock: Explaining the core mindset of highly accurate forecasters "The thing with a base rate is you might have a set of instances where something's happened over the last 10 years, but if they're gradually trending up or gradually trending down, that's a kind of useful piece of information." — Michael Story: Describing how expert forecasters use historical patterns and trends

Implications: Forecasting is a practical, learnable skill that can improve decisions in politics, business, and daily life. Listeners are encouraged to use evidence, base rates, and crowd wisdom rather than overtrust intuition or expertise.