Predictive Analytics, Revised and Updated

Preface to the Original Edition

Yesterday is history, tomorrow is a mystery, but today is a gift. That's why we call it the present.

—Attributed to A. A. Milne, Bil Keane, and Oogway, the wise turtle in Kung Fu Panda

People look at me funny when I tell them what I do. It's an occupational hazard.

The Information Age suffers from a glaring omission. This claim may surprise many, considering we are actively recording Everything That Happens in the World. Moving beyond history books that document important events, we've progressed to systems that log every click, payment, call, crash, crime, and illness. With this in place, you would expect lovers of data to be satisfied, if not spoiled rotten.

But this apparent infinity of information excludes the very events that would be most valuable to know of: things that haven't happened yet.

Everyone craves the power to see the future; we are collectively obsessed with prediction. We bow to prognostic deities. We empty our pockets for palm readers. We hearken to horoscopes, adore astrology, and feast upon fortune cookies.

But many people who salivate for psychics also spurn science. Their innate response says “yuck”—it's either too hard to understand or too boring. Or perhaps many believe prediction by its nature is just impossible without supernatural support.

There's a lighthearted TV show I like premised on this very theme, Psych, in which a sharp-eyed detective—a modern-day, data-driven Sherlock Holmesian hipster—has perfected the ...