Gary Smith wants to change the way you look at the world.
Just about every day, you encounter a statistical phenomenon known as regression to the mean — in medicine, business, sports, education, investing, even parenting. But you probably don't know about this concept. In his new book What the Luck? The Surprising Role of Chance in Our Everyday Lives, Smith, a professor of economics at Pomona College in southern California, explains that once you grasp what regression to the mean is and stop underestimating the role of luck in your life, you'll look at things in an entirely new light.
For example, let's say a student scores 95 percent on her first test, and follows up with an 80 percent on the next exam. Maybe you assume that this is a very smart student who got cocky, started slacking off, or just stopped studying altogether. How else could an A student become a B- student? Well, maybe because they weren't an A student in the first place.
Subscribe to The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.
SUBSCRIBE & SAVE
Sign up for The Week's Free Newsletters
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
The more likely case, Smith says, is that the A — far from a "mean" grade — was the outlier. And a person scoring far from the mean on her first test and closer to the mean on the second is an example of regression to the mean.
"People tend to overreact to things because they think they're permanent, when in fact they are temporary," Smith said. "If a student received a 100 percent on a test, I overreact by saying she's the best student I've had in five years. When she has a somewhat lower score the next test, I think, 'What happened? Am I doing a bad job teaching? Is she unlearning things?' You can overreact both ways. Think about the role of luck and that a student who gets a 95 the first time should expect something somewhat lower. It doesn't make them a bad student, just that their first score was probably exaggerated."
It happens in business, too: People overreact to the stock market when a good earnings report is announced and a company's stock price jumps up — and when the price falls after one bad quarter.
"You have to find out for sure if it's a fluke," Smith said. "The more data we have, the better idea we have."
The key is not to overreact — to anything, really. Everything that happens is just one more data point. Try applying that worldview to the presidential race, Smith suggests.
"Every four years, we have these great hopes and think [a new president is] going to solve all of our problems," Smith said. "Their popularity is lower at the end than at the beginning of their term because it turns out they weren't Superman or Superwoman." Instead, they were closer to the mean than their supporters hoped and their critics feared.
"I'm not going to say Trump or Clinton will be the next Lincoln or Washington," he said, "but its not going to be the end of the world. They are probably not going to ruin the country and have us all move to Canada."
You see regression to the mean in sports, too. "Somebody will win the Rookie of the Year award, and everyone thinks they're the next Michael Jordan or LeBron James," Smith said. "The next year they don't do quite as well, and they call it the sophomore slump. The analysts ask, 'What happened? They got lazy, they didn't try as hard.' Really, it was probably a little bit of luck the first time around." There's also the Sports Illustrated jinx — some say that after an athlete or team appears on the cover, they don't perform as well during their next few games. "To get on the cover of Sports Illustrated, you have to do something really, really extraordinary," Smith said. "You had to be very fortunate. Luck is fleeting."
People tend to remember their lucky streaks and attach significance to them, Smith said. In Las Vegas, for example, gamblers often suffer from selective recall — you'll be able to remember the time you won $500, but not the 10 other times where you lost $100.
"People think they are smarter than average and are going to win," he said. "If they win, it proves how smart they are." But if they lose, they search for excuses. "You explain away your losses and gloat about your victories. You're just as overconfident, even though you lost your money."
So what's the lesson? Don't overreact. Because "just by luck alone, there is no predictive power of what's going to happen next."
