It’s hard to predict the future, but you can be better at predicting the future. All you need is a few delicious avocados. Even the “experts” are bad at predicting the future Wharton professor Phillip Tetlock wanted to make the future easier to predict. So he held “forecasting tournaments,” in which experts from a variety of fields made millions of predictions about global events. Tetlock found that experts are no better at predicting the future than dart-throwing chimps. In fact, the more high-profile experts – the ones who get invited onto news shows – were the worst at making predictions. But, Tetlock found that some people are really great at telling the future. He calls them “Superforecasters”, and regardless of their area of expertise, they consistently beat the field with their predictions. Tetlock also found that with a little training, people can improve their forecasting skills. The superforecasters in Tetlock’s Good Judgement Project – people from all backgrounds working with publicly-available information – make forecasts 30% better than intelligence officers with access to classified information. Creative work is uncertain. Does it have to be? As someone working in the “Extremistan” world of creative work, I’m always trying to improve my forecasting skills. If I publish a tweet, how many likes will it get? If I write a book, how many copies will it sell? The chances of getting any of these predictions exactly right are so slim, it doesn’t feel worth it to try to predict these things. But that doesn’t mean I can’t rate my predictions and make those predictions better. Introducing the Avocado Challenge If you would like to be better at predicting the future, I have a challenge for you. I call it the Avocado Challenge. Elon Musk recently asked on Twitter “What can’t we predict?” I answered “whether or not an avocado is ready to open.” 12 likes. People agree with me. https://twitter.com/kadavy/status/1309643017599569920  Here’s how the Avocado Challenge works. The next time you’re about to open an avocado, make a prediction: How confident are you the avocado is ripe? Choose a percentage of confidence, such as 50% or 20% – or if you’re feeling lucky, 100%. To make it simple, you can rate your confidence on a scale of 0 to 10. State your prediction out loud or write it down. Now, open the avocado. Is it ripe? Yes or no? Scoring your avocado predictions You now have two variables: Your prediction as stated in percentage confidence, and the outcome of avocado ripeness. With these two variables, you can calculate what’s called a Brier score. This tells you just how good your forecast was. The Brier score is what Phillip Tetlock uses to score his forecasting tournaments. Two variables: confidence and outcome It works like this: Translate your percentage confidence into a decimal between 0 and 1. So 50% would be 0.5, 20% would be 0.2, and 100% would just be 1. Now, translate the avocado ripeness outcome into a binary number. If the avocado was not ripe, your outcome value is “0.” If the avocado was ripe, your outcome value is “1.” (You may wonder: How do I determine whether or not an avocado is ripe? I’ll get to that in a minute. Let’s pretend for a second it’s easy.) Calculating your Brier score Once you have those two variables, there are two steps to follow to find out your Brier score: Subtract the outcome value from your confidence value. If I was 50% confident the avocado would be ripe that confidence value is 0.5. If the avocado was in fact ripe I subtract the outcome value of 1 from 0.5 to get -0.5. Square that number, or multiply it by itself. -0.5² = 0.25. Our Brier score is 0.25. Is that good or bad? The lower your Brier score, the better your prediction was. If you were 100% confident the avocado would be ripe and it was not, your Brier score would be 1 – the worst score possible. If you were 100% confident the avocado would be ripe and it was ripe, your Brier score would be 0 – the be