Why the smartest people often make the worst decisions
You’ve probably seen a smart person make a surprisingly bad decision. Maybe a friend who saw straight through other people’s hang-ups, then spent years bogged down in his own. A brilliant colleague who defended a doomed project for months, with arguments no one could counter. In the end he couldn’t counter them either. Or maybe you, one evening, looking back at a decision you knew was wrong from the start, wondering how you’d managed to convince yourself so thoroughly.
We usually explain this away with stress or emotions. As if intelligence were a good tool that had briefly stopped working. As if what kept you stuck was just a malfunction.
It’s not a malfunction. The tool is working exactly as designed, just not for what you think.
You see other people’s biases. Not your own.
Psychologist Keith Stanovich spent over a decade building tests for rationality, separate from intelligence tests. What he found is uncomfortable: the ability to make sound, unbiased decisions is largely independent of IQ.
In 2012, with West and Meserve, he tested what’s known as the bias blind spot: the tendency to spot reasoning errors in others while remaining blind to your own. They measured seven different cognitive biases. The result: the most intelligent participants didn’t have a smaller blind spot. They had a larger one. On six of the seven biases tested, the more cognitively sophisticated the person, the less they saw their own errors.
Stanovich coined a word for this: dysrationalia. The ability to act irrationally despite perfectly adequate intelligence.
Sound like anyone you know?
The more confident the expert, the more wrong they are
Philip Tetlock, a psychologist at Wharton, ran one of the longest studies ever conducted on the quality of expert judgment. Over eighteen years, he tracked 284 specialists: economists, geopolitical analysts, political scientists. He collected 28,000 predictions, then compared them with what actually happened.
The most uncomfortable finding: the most confident experts, those who reasoned from one big unifying idea, predicted worse than chance. Not on par with chance. Worse than chance.
Tetlock called them “hedgehogs,” after the Greek poet Archilochus: “The fox knows many things, but the hedgehog knows one big thing.” The hedgehog has a theory about the world. They’re convinced of it, and they’re consistent. And they’re wrong more often than someone flipping a coin.
The “foxes” were the ones who doubted, who drew from multiple frameworks, who had no grand theory. They did better. Not spectacularly better. But better than chance, at least.
The expert’s intelligence didn’t protect them. It made their mistakes better defended and harder to let go of.
The most effective lawyer you ever hired
Daniel Kahneman, Nobel laureate in economics, spent his career documenting how the human brain takes shortcuts that look like thinking. His conclusion: intelligence doesn’t neutralize cognitive biases. It dresses them up better.
Here’s how it plays out. Say you’re on the fence. You know you should leave that job, that relationship. The information is there. The signals are clear. But when it comes time to decide, something kicks in.
“It’s not the right time.” “I’ve put too much time into this to walk away now.” “People who leave in this situation usually regret it.” “And there have been some positive signs lately, to be fair.”
Every sentence is reasonable. Every argument holds up. The whole thing looks like a balanced assessment. Except it isn’t one. You’re not weighing pros and cons. The verdict is already in. What’s running through your head isn’t thinking. But you don’t know that yet. All you feel is that you have good reasons to stay.
Someone with an IQ of 100 will find two reasons not to change anything. Someone with an IQ of 140 will find seven, and each one will be flawless. The outcome is the same: nothing moves. But the second person is absolutely convinced they’ve thought it through.
Reasoning doesn’t do what you think it does
If you recognized something in that inner monologue, what comes next will explain it. And you’re not going to like the explanation.
In 2011, researchers Hugo Mercier and Dan Sperber published a theory that shook cognitive science. Their thesis fits in one sentence: human reasoning didn’t evolve to find truth. It evolved to win arguments.
Read that again slowly.
For centuries, we’ve treated reason as a tool of discovery. A way to reach sound conclusions through reflection. Mercier and Sperber say: no. It’s a tool of persuasion. It evolved to convince others and evaluate their arguments. Not to see clearly on your own.
That’s why those four sentences feel like thinking but aren’t. They’re arguments. Your brain builds a case, exactly the way it would to convince someone else. Except there is no one else. You’re alone. And the only person left to convince is you.
The proof is in a paradox psychologists have observed for decades. When we reason alone, we’re bad at it. We confirm what we already believe, we ignore what’s inconvenient. But put the same person in front of someone who disagrees, and something shifts. Suddenly they can reassess, take in the opposing view, drop a position they were defending five minutes ago.
Because reasoning was built for arguing with someone else. Not for thinking on your own.
You know that moment: alone, late at night, “thinking things through” to make the right decision. That’s precisely when you’re most vulnerable. A persuasion tool with no one to push back. And the smarter you are, the sharper that tool gets.
The question that remains
If intelligence doesn’t protect you from bad choices, if expertise amplifies confidence without improving accuracy, if reasoning itself is a persuasion tool disguised as a truth-finding tool, then what are you actually deciding with?
Maybe what you need isn’t better analysis. Maybe it’s an angle you haven’t tried yet.
And maybe the hard part was never to think harder. It was to stop thinking alone.
The thing is, that moment when you’re alone with your “good reasons” rarely lands at an hour when there’s anyone around to answer. That’s why Focal Shift created an AI Advisor. Not to think for you. To ask the question your intelligence is too busy avoiding. At any hour.
Focal Shift (focalshift.eu, available in English and French). A deliberately simple tool, built on anything but simplicity.