My brain has been lying to me with confidence my entire life. Not occasionally. Constantly. And for a long time I didn’t notice, because confidence is the one thing that feels exactly the same whether it’s earned or not.

I spent a while going after this. Reading, catching myself mid-decision, writing down the places I’d been confidently wrong. The first thing I learned is that the lies aren’t random. They come in patterns. Once you see them, you can’t unsee them.

Three traps. Each one a different way my brain lied to me.

Here are three of its favourite lies.

Bias #1: Survivorship bias, your brain only counts the winners

Steve Jobs did it. Zuckerberg did it. Gates did it. Dropouts, all of them, and your brain connects the dots: dropping out is the move.

What it doesn’t connect is the millions who dropped out and now make less than their graduating classmates. Nobody writes a magazine profile about that guy.

But here’s the half of survivorship bias nobody talks about. It’s not just the failures you don’t see. It’s most of the successes too.

I work at a company called Procol. Real business, real revenue. You’ve never heard of it. You wouldn’t recognise the founders if they were standing next to you in a lift.

The founders everyone can name are famous for being famous, not for building the most. The actual economy runs on companies you’ve never heard of, and my brain keeps pointing at the famous three and calling that the data.

I catch myself doing the same thing with startup founders. I’ll read about someone who quit their FAANG job, built a SaaS, and hit $1M ARR in 18 months, and my brain says “that could be me.” What it doesn’t say: for every one of those stories, thousands quit, burned through savings, and quietly went back to job hunting. I just never meet them, because they don’t post about it on LinkedIn.

The media doesn’t report reality. It reports interesting deviations from reality. And my brain treats those deviations as the norm.

Bias #2: Sunk cost fallacy, throwing good after bad

The clean version first. You buy a movie ticket for $20. Fifteen minutes in, you know it’s garbage. Do you leave?

Of course not. You sit there for another two and a half hours because you already paid. Here’s the math your brain refuses to do: that $20 is gone whether you stay or go. By staying, you’re spending more time on top of it. If your time’s worth $20 an hour, you just spent $50 to “recover” $20. A $20 mistake now costs you $70, and $50 of that was voluntary.

This is the one bias I’m actually good at, and it shows up in the smallest place. I start a book. A hundred pages in, it’s not getting better, or it’s just not my taste, for a thousand different reasons. I drop it. Why spend any more time on it. The $20 is gone. Walk out.

Books are easy, because the cost is small and the exit is clean.

Then there’s the version you can’t price. Relationships. You can’t put a number on emotions, which makes sunk cost fallacy almost impossible to see in real time. You’ve invested six months, a year, three years. You know it’s not going anywhere. But leaving means all that time was “wasted,” so you invest another year. And another.

The thing I learned, the hard way, is that familiarity and love are two different things. They feel similar enough that you can spend years confusing one for the other. The math is identical to the movie ticket: the cost is already sunk, the only question is whether you keep paying. Except you can’t see it, because there’s no receipt.

Bias #3: Confirmation bias, seeing what you already believe

I was watching vlogs by Paramvir Beniwal, a travel YouTuber who’s been to over a hundred countries. One of them was from a village in Africa, and the thing that hit me was how much it looked like a village in a poor part of India. Same kind of houses, same kind of dust, same kind of people getting on with their day.

So I looked it up. Per capita income in that part of the world is in the same band as Bihar, around $880 a year. Girls’ education, for every ten years boys spend in school, girls spend about nine. The numbers weren’t the catastrophe I’d been carrying around in my head.

Here’s what I realised about my own head. I’d absorbed one Africa, the one on the news, which only shows up for famines, wars, and coups. My brain had filed the whole continent under “uniformly terrible,” and every headline confirmed it. A travel vlog was the first time I’d looked at the same place from a different angle, and the picture wouldn’t hold.

That’s confirmation bias. Not a bad dataset, but a dataset you only ever look at from one side. Every fact that fits, you nod at. Every fact that doesn’t, you find a reason to dismiss.

There’s a line in Factfulness that lands differently now. The authors describe couples across the United States, Iran, Mexico, Malaysia, Brazil, Italy, China, Indonesia, India, Colombia, Bangladesh, South Africa (different languages, different scriptures on the nightstand) and they’re all doing the same thing in their bedrooms. Whispering about their dreams for their kids’ future.

Every language has a word for “happy.” Every language has a word for “grief.” That’s not a coincidence. It’s evidence that humans run on nearly identical hardware, and most of what we think divides us is just what we haven’t looked at closely enough.

I’m not saying the world is fine. It isn’t fine in the village I saw in that vlog, and it isn’t fine in Bihar either. But the world isn’t as bad as the news needs it to be. It’s also not as good, everywhere, in uniform. It’s somewhere in between, and you only find where by actually looking.

So what do I do with this?

I don’t have a cure. The brain still lies. But I’ve got three questions I run on myself now, one per lie.

When a success story sounds too clean: how many people tried the same thing and failed, and how many succeeded without anyone noticing?

When I’m holding on because I’ve “already invested so much”: if I were starting fresh today, would I make the same choice?

When a belief feels obviously true: what data would change my mind?

If I can’t answer that last one, I’m not thinking. I’m just confirming.