Everything is predictable How Bayesian statistics explain our world

This beguiling mathematical romp from science writer Chivers (The Rationalist's Guide to the Galaxy) surveys the far-reaching applications of the statistics theorem elaborated by the 18th-century English minister Thomas Bayes, who showed how to estimate the probability that a hypothesis is true by considering new data alongside "prior" assessments of the hypothesis's accuracy. (For instance, the theorem might determine the probability that a middle-aged woman has Covid by considering a positive test result alongside the virus's prevalence rate among middle-aged women generally.) Bayes's theorem produces startling insights that can upend conventional wisdom, Chivers writes, noting that the equation explains why "a cancer test can be 99 percent accurate even though 99 percent of the people it says have cancer don't." Examining the theorem in a raft of offbeat contexts, the author suggests its focus on evaluating new information in the context of previous beliefs sheds light on why vaccine skeptics are unmoved by evidence demonstrating vaccines' safety and efficacy, and why contestants guessing which door hides a prize on Let's Make a Deal should always switch their pick after the host reveals one of the losing doors. Chivers's dive into probability theory is heady but lucid, and conveys arcane concepts in commonsensical prose. The result is a stimulating take on making sense of a murky, uncertain reality. Photos. Agent: Melissa Flashman, Janklow & Nesbit Assoc. (May)

An instructive look at "how likely something is, given the evidence we have." In this compelling account, science writer Chivers, author of The Rationalist's Guide to the Galaxy and How To Read Numbers, introduces us to Thomas Bayes, who developed "perhaps the most important single equation in history." The author explains that life is not a chess game. It's like poker, where we make decisions based on limited information. "The usual way to explain Bayes' theorem is with medical testing," writes the author. For example, does a woman with a positive mammogram have breast cancer? No test is perfect, but it must be nearly 100%, right? Wrong. Readers may be surprised to learn that a test that is 90% accurate (typical of a mammogram) isn't the same as there being a 90% chance that it's correct. Bayes predictions require additional information--in this case, the incidence of breast cancer in the population. Chivers may not be exaggerating his subject's importance, but this is one of the longest of many popular books on Bayes' theorem. Delving almost too deeply, he delivers a history of scientific prediction as well as the ongoing controversy within the statistics community between pro- and anti-Bayesian factions. He also offers a marginally relevant but jaw-dropping account of the current state of science, where ignorance or deliberate manipulation of statistics by ambitious researchers has produced an epidemic of studies announcing results that often can't be reproduced. "Science," he writes, "is explicitly about making predictions--hypotheses--and testing them….The problem is that in science, we like to think that there is an objective truth out there, and the Bayesian model of perception is explicitly subjective. A probability estimate isn't some fact about the world, but my best guess of the world, given the information I have." An ingenious introduction to the mathematics of rational thinking. Copyright (c) Kirkus Reviews, used with permission.