New housing construction startsBlackBerry global smartphone market share
It turns out that as Americans stopped building houses, they also stopped carrying small rectangles that could send emails, which is either a profound statement about the relationship between real estate and portable technology or simply evidence that the universe is playing an elaborate prank on anyone foolish enough to collect data. Between 2007 and 2016, housing construction and BlackBerry's grip on the smartphone market moved in perfect opposition, as though one were the evil twin of the other, which is to say they had almost nothing to do with each other whatsoever.
What's genuinely interesting here is that both trends were actually driven by the same lurking variable: the financial crisis of 2008 and its long, exhausting aftermath. Housing construction cratered because banks stopped lending and people stopped buying, while BlackBerry's market share evaporated because the iPhone arrived in 2007 and everyone realized they wanted one instead, and these two separate catastrophes just happened to move in opposite directions like synchronized swimmers who had never met. The 2008 recession eliminated roughly 80 percent of housing starts at their nadir, and by 2016 BlackBerry's market share had dwindled to less than 0.5 percent globally—two entirely different industries responding to entirely different forces, yet moving together as though choreographed by someone who understood neither industry particularly well.
This is what happens when you have a decade of data and enough statistical tools to find patterns in literally anything: two unrelated things will occasionally move in lockstep, and you'll find yourself explaining it to people at dinner parties with an increasingly hollow sense of purpose. The real lesson is not that housing and BlackBerry phones are secretly connected, but that humans are magnificent pattern-recognition machines who will find meaning in the static if you give them ten data points and a regression line. We saw correlation and built a narrative around it, which is exactly what we shouldn't do. Yet we did anyway.
As an Amazon Associate, getspurious.com earns from qualifying purchases. Learn more.
Want to learn more about why correlations like “New housing construction starts” vs “BlackBerry global smartphone market share” don't prove causation? Read our guide to statistical thinking.