"The illusion that we understand the past fosters overconfidence in our ability to predict the future."
Imagine I organize a coin-flipping contest. In this contest, 1,000 people square off in a March Madness-like bracket, with 500 first-round matches. The winner of each match is the one who flips the most heads in a row. After about 10 rounds we find ourselves watching the championship between the best two head-flippers.
We know this is a competition that doesn't rely on skill. Yet, the final two out of 1,000 will feel a sense of control over the outcome. We know, because we are outside of the situation, that no matter what happens there has to be two final people, and then a winner. That's just math.
A funny thing happens when you are the one experiencing the random math. Imagine you did "win" two matches, then three, and four, and so on. Our experience is that we flipped seven heads in a row, several times.
The outcome of our experience had to happen to someone, but the fact that it was us is random. Though we feel like we have some control, we're missing the bigger point.
Let me tell you something you already know; this phenomenon doesn't just happen with flipping coins.
We are very good at spotting patterns. Random things happen all the time and we're been wired to think there's meaning behind it. We might call these coincidences.
The ability to spot patterns may have helped out ancestors; after all, if you noticed that it only rains when there are clouds in the sky it gave you an advantage over those who didn't know. It's easier for us to believe there's a pattern or a reason behind something than to have to admit to ourselves that probability and randomness are all around us.
We spot patterns in stock prices. We see patterns in politics. We think there are patterns in others' behaviors. Some even think patterns exist between which socks they wear and the outcome of sporting events.
Pattern Recognition is Easy
It's easy for us to recognize patterns. But raining only when there are clouds isn't the same thing as it raining any time there are clouds.
True, there are patterns that we detect that make sense. However, there are more that are mere random, have no base in reality and aren't able to predict the future.
Small Sample Sizes
It's quite intuitive to think experience is the best teacher. Every job interview I've ever had featured a discussion about my experience. They wanted to know if I had enough experience. They wanted to know if I had the right kind of experience.
To understand the problem with experience and our view of the world I have to use a little bit of jargon - I apologize ahead of time. If we want to make sure something is a fact and not just an outlier we need to know about the whole universe. Since that's impossible, we have to get what statisticians call a sample. If our sample is not biased, meaning the sample represents the whole universe, then our sample can give us good information - if it's sufficiently large. The smaller our sample is, the more likely it is that our observations are extreme.
Jaron over...Let me try again with an example and plain English.
I'm going to use an example of a common belief, that the rich are bad people. You may hold this belief or one like it - the only way to get rich is to take advantage of people, rich people only care about money, rich people are greedy, and so on. For our purposes, let's assume a person named Kim believes rich people are bad.
It's completely possible that Kim's experience is accurate. That is, every time Kim met someone "rich," that person was a real jerk.