How to expect the unexpected The science of making predictions -- and the art of knowing when not to

In this smart study, Yates (The Math of Life and Death), a mathematician at the University of Bath, explores the cognitive biases that lead to incorrect predictions. Intuition often conflicts with reality, he explains, suggesting that humans have a linearity bias, or "propensity to believe that things will stay constant or continue at a consistent rate." He contends this bias contributed to the delayed uptake of antiviral measures as Covid-19 cases ticked upward in early 2020, noting a study that found subjects who underestimated the speed of exponential growth were less likely to practice social distancing. Elsewhere, Yates illustrates how people misunderstand probability by discussing the 1967 case of a man who appeared to have predicted a real-life plane crash in a dream. Drawing on research about dream frequency, Yates estimates that "we might expect over 66,000" such dreams across the world in the month before a crash, suggesting that what appeared to be too accurate to be a coincidence was actually to be expected. The survey of the mind's biases intrigues, and the author excels at demonstrating their real-world effects, as when he posits that the tendency to assume things will always be "just the way they are now" contributes to many people's reluctance to observe hurricane evacuation orders and other precautions. It's a safe bet that readers will take to this. (Oct.)

From tarot cards to forecasts of Armageddon, a mathematician examines the psychology of predictions, debunking myths and setting an agenda for clear thinking. It's human nature to want to know what is going to happen in the future. However, writes Yates, author of The Math of Life and Death, doing so with a useful degree of probability is extremely difficult--or even impossible. The author pleasantly explains the tricks used by psychics and charlatans, which usually involve telling paying customers what they want to hear. He tracks numerous apocalyptic predictions and the reasons given by the forecasters for their obvious failure. There is also a tendency of people to see patterns in events and data that don't exist. Random distribution can throw up apparent causations and connections, but they are really no more than background noise. Humans think in linear terms, assuming that the future will be like the present and therefore precise extrapolations are possible. Not so, says Yates. There are too many variables to consider. True, linearity is needed for everyday existence, but when it comes to making predictions, it is more hindrance than help. The author examines the different types of delusional thinking and outlines the mathematics of probability, and he devotes a useful chapter to chaos theory. The only field with a scientific basis seems to be short-term weather forecasting, although even there, things can go disastrously wrong. Math-based models can be important tools, with the proviso that the output is only as reliable as the input. In the end, there is no perfect prediction method. The best we can do is think broadly, be prepared to change our minds in light of new evidence, and understand our own biases. Yates' tour of the predictions business covers much interesting ground, which he tills with an entertaining sense of humor. Copyright (c) Kirkus Reviews, used with permission.