mitchell vitez

dark mode

blog about music art media

resume email github

Start Predicting Everything

Back in college I had a dumb skill. My friends would ask me what time it was, and without looking at a clock or my phone or my watch or whatever I’d often be able to get the time exactly right, down to the minute. I didn’t have some innate skill for timetelling or anything like that. What I did have was more practice guessing the time than most people around me.

For whatever reason, there was one time when I wanted to check what time it was, but before doing so I made myself guess, just to see how close I could get. I’m pretty sure that first guess was very far off the actual time, so I made a lighthearted commitment to try to improve my guesses.

This became a habit. Every time I wanted to check the time, I would first guess and see how far off I was. Sure enough, the guesses started getting closer and closer. I didn’t feel like I was doing any hard work. Maybe my brain was doing something behind the scenes, but knowing the time never felt especially important to me or like something I should explicitly practice knowing. I just kind of fell into a loop with good instant feedback, the kind that naturally makes your estimates better over time.

People sometimes ask for the current time, and it’s fun to be the first person to answer correctly, so of course it was easy for me to transition this little guessing game into a somewhat-rewarding interaction with others. Once people know you have a dumb skill like this, they like to put it to the test, which meant even more practice for me!

Now, I’ve mostly lost this skill because I haven’t kept practicing. But I think it’s cool to know that even something as abstract and weird as guessing the current time can be managed by your subconscious mind somehow, in a way that lets you improve, and maybe mildly impress your friends.

I’d bet that this generalizes fairly well. If you want to get better at predicting something, just start predicting it. You don’t have to understand the mechanisms behind your improvement, but I think you probably do need to have fairly accurate and immediate feedback for this to work well.