N=1 is Significant

A close friend hates that I sometimes judge people based on a single data point. My favorite example is Zoom Calls. A person who is one minute early is very different from one who is one minute late.

When someone is late, maybe their last call ran over, they overbooked, or they didn't leave enough buffer. Each possibility points to poor planning or that your meeting wasn't a priority. Is this world shattering alpha? No, but it helps you better get who you are dealing with and how important you may be to them.

When it comes to understanding people, paying close attention to one individual's behavior is often more useful than relying on broad studies. An anecdote about a single person in context can tell you more than a study of one hundred subjects in a lab.

Psychology research often uses college students as participants, but anyone who has met undergrad psych students knows they are hardly typical. What can we really learn from the behavior of 20-year-olds in artificial settings? General findings like “most people think they're above average” might be true in aggregate, but they won't help you understand the specific person sitting across from you.

Not all anecdotal evidence is equally revealing. Take poker and chess. A single poker win says little about your actual skill. But a few chess games against a rated opponent can give a clear sense of your max elo.

Let's take an extreme example. If you see someone smoke crack once, you don't need much inference to guess they're a crackhead, even though your sample size is just one.

Is this methodology perfect? Of course not, but it has strong predictive power.

Ultimately, people are creatures of habit. Knowing that a single outcome can reveal a pattern helps you relate to others more effectively. When I meet someone who works out at 5 a.m., I don’t invite them for a 10 p.m. drink—I suggest a 7:30 a.m. coffee. It shows them that you get them, and you're more likely to connect.

Being observant is being considerate. This has helped me build better relationships, even if it annoys some friends when I jump to conclusions a bit too quickly, but one data point is sometimes all you need.