It turns out that as Amazon figured out how to deliver things to your door faster, Americans got proportionally better at stepping into traffic, and for eighteen years straight, these two things moved together like a particularly grim dance partnership. One might imagine the universe has a peculiar sense of irony, or one might simply accept that humanity will find a pattern in absolutely anything, which is rather more likely. We have correlated online shopping with death.
What's actually happening here is far more mundane and therefore rather fascinating. Between 2005 and 2022, the United States experienced sustained economic growth, population increase, and a dramatic rise in vehicle miles traveled—all of which would naturally lift both Amazon's revenue curve and, regrettably, pedestrian fatality statistics. More people moving around, more cars on roads, more money circulating through the economy, more packages being rushed toward your front porch in increasingly efficient delivery windows. To put the scale of it plainly: pedestrian deaths rose from about 4,900 annually in 2005 to nearly 7,500 by 2022, while Amazon's revenue climbed from roughly $8 billion to over $469 billion—two entirely separate human systems both accelerating together because the underlying economy was accelerating.
The correlation between Amazon's explosive growth and rising pedestrian fatalities tells us almost nothing about Amazon, and almost everything about our appetite for detecting meaningful relationships in data that merely happened to trend upward during the same period. We are pattern-matching creatures living in an increasingly complex world, which means we will forever find signals in noise, connections in coincidence, and meaning in the movements of unrelated things. Sometimes they just both go up.
As an Amazon Associate, getspurious.com earns from qualifying purchases. Learn more.
Want to learn more about why correlations like “Amazon annual revenue” vs “Pedestrian traffic fatalities” don't prove causation? Read our guide to statistical thinking.