Join Yash Patel for an in-depth discussion in this video The experts, part of The Psychology of Living In A Data-Driven World.
- [Instructor] We, as a human race, have been historically awful at making predictions. So why do we keep doing it? Well, in 2011, the authors of the popular series Freakonomics, Stephen Dubner and Steven Levitt, argued that there are just huge benefits to people who can make predictions correctly, and there is no punishment for people who make bad predictions, mostly because bad predictions are immediately forgotten. So there is this huge incentive for someone to keep making predictions over and over and over again, no matter how right or wrong he or she is.
Think about it. In terms of probability, you only have to be right 51% of the time. People will remember that 1%. Okay, so if people are so bad at making predictions, why do we have these experts that we rely on? That, my friends, is because we don't really know that much. The world is too complicated to be predicted with accuracy, and we, as humans, are wired to avoid uncertainty. But these experts, they know one thing. This one thing gives them a significant advantage over plebeians like us, who know little to nothing.
And here is the crazy thing. Even the experts are wrong. All the darn time. Look at these political pundits. They know politics really well, I'll give them that, but they can't make predictions worth their weight. Yet, they show up repeatedly on Fox, CNN, MSNBC. Pick your flavor. Again, it's because they know slightly more than us. They know that one little thing that's going to give them an edge, and they engage with our visceral emotions. I'll admit Nate Silver's team is an anomaly.
He's pretty spot on with a bunch of his research. Pretty sure he's some kind of wizard, and I am really jealous. I can't even set my fancy football lineup optimally half the time. So I should back up, though. If you don't know Nate Silver, he runs that site, FiveThirtyEight. He rose to fame in the 2012 presidential election when he predicted the outcome of all 50 states and the District of Columbia correctly. But Nate doesn't play to emotions. His team analyzes data, and a lot of it.
And they take into account polling methodologies, sample sizes, error calculations. You know, that mathy stuff. However, even armed with all of this data, he still whiffed during the 2016 elections with the incorrect prediction for president. Just goes to show you anybody can be wrong. So why are we so bad with predictions? Let's take a look at that next.