You’ve probably heard the saying “celebrities die in threes.” This, of course, is one of the more silly things that a human can utter. But in case you needed someone to fact-check this one for you, the New York Times went to the trouble in an article from 2014:
Since 1990, 449 [”celebrities”] have died. In 75 cases, two of them died within three days of each other. But in only seven cases did three of them die within a five-day period. According to my colleague Boris Chen, a statistician, this is about what you’d expect by random chance.
After the deaths of any two celebrities close together, some people are waiting for a third celebrity to die. Maybe one will, and maybe one won’t. But this idea that celebrities always die in threes is a logical fallacy right up there with the Texas sharpshooter fallacy, where we see patterns that don’t actually exist by focusing on data that support our theory while ignoring anything that would contradict it.
As strange, curious humans we’re always looking for patterns, leaving us prone to believe a lot of superstitious nonsense. But when it comes to the “celebrities die in threes” garbage the first obvious question is, how do you define who is and who is not a celebrity? The New York Times combed through their obituaries and considered anyone whose obit ran longer than 2,000 words to be a celebrity.