Quantcast
Coincidence Threshold - Dilbert Blog

Coincidence Threshold

You probably saw the story about the woman who won the lottery for the fourth time. Some say the odds of that are 1 out of 18 septillion. I think almost everyone who reads this blog assumes it wasn’t luck, and that there was some criminal activity involved. But it made me wonder if humans have some sort of innate and similar view of where coincidence fails as an explanation. Let’s test that.

Let’s say a mysterious man appears at your door and tells you that you have a special power. If you write the name of a person on a piece of paper and put it in a sock overnight, that person will die. You try it once out of curiosity, picking the name of some scoundrel from the news so you won’t feel bad if it works. And sure enough, that person dies of a heart attack that very night.

Your first thought might be that as a general rule, scoundrels live risky lifestyles. You might have gotten lucky on that one. So you try it again the next night with a new name, and that person is also dead by morning.  You try it twenty times, never telling anyone else of your experiment, and each time it works before sunrise the next day.

You can imagine a variety of explanations for your experience. You might be crazy, or dreaming, or experiencing selective memory. Maybe the mysterious man who told you of this method is somehow watching you and putting out the hit order on the name you choose, for reasons that you can’t understand. Maybe you’re just good at recognizing when people’s time has come. Maybe you are part of some sort of science experiment or reality show. Perhaps there are a dozen other explanations.

My question to you is this: How many times could you repeat this experiment alone, with whatever controls or privacy you could muster, before you believed your actions were causing specific people to die?

For me it would be somewhere around the fourth person.