You Can Get Better at Judgement—Just Not the Way Self-Help Sells It

We keep asking whether judgement is fixed or growable. Can we get better at it, or are we stuck with what we've got? The question feels urgent—especially if you've watched someone repeat the same bad calls or wondered why your own decisions sometimes misfire. I've seen smart people make the same bad call twice and wondered what would actually help. But the way we usually frame it is the problem. The real answer isn't "yes, you can" or "no, you can't." It's that judgement can improve, but only when we stop treating it as one thing and start focusing on the parts that actually change—and the parts that don't.

The Question That Won't Go Away

Self-help has a strong default: you can change. Mindset, habits, deliberate practice. The counter-view shows up in everyday talk too—that some people "just have good judgement" and others don't. So we end up with a binary: fixed or growable. Either we're born with a set amount of whatever this is, or we can add to it. That binary is seductive. It's also the wrong way to look at it.

Why "Judgement" Is the Wrong Unit

Judgement isn't a single skill or a fixed quota. In the research it shows up as a bundle of processes: what you pay attention to, what you believe in advance, how you use feedback, and how much practice you have in a given domain. Some of those pieces are relatively stable across contexts. Others shift a lot with training and structure.

So "can judgement improve?" only makes sense once you say which bit you mean. General "good judgement" in the abstract is less trainable. What is trainable is calibration—how well your confidence in a decision matches how often you're right—and domain-specific judgement in areas where you get clear feedback. If you treat judgement as one lump, you'll either overclaim (everything can change) or underclaim (nothing can). Both miss the picture.

What Actually Improves

What does move the needle? Calibration improves when people get clear, timely feedback on outcomes. The Good Judgment Project and related forecasting research have shown that when people get systematic feedback on their predictions—how often their "70% confident" calls are right about 70% of the time—they get better at matching confidence to accuracy. Domain expertise improves with deliberate practice in that domain, not with generic "think better" advice.

And structured processes help: pre-mortems (imagining the decision has already failed and asking what would have caused it), checklists, red teams, and simple decision rules. So does the environment: who's in the room, what's default, what's framed as optional. A lot of "better judgement" is better feedback, better structure, and better context—not a bigger dose of an invisible trait.

So why does so much advice miss the mark?

What Self-Help Gets Wrong

A lot of self-help sells improvement without saying what's actually changing. "Improve your judgement" or "think more clearly" stays vague. It also tends to ignore selection: the people who stick with a program are often the ones who were already motivated and able to benefit. That doesn't mean improvement is impossible—it means we should be careful not to overclaim. And it often underplays the role of the environment. You can work on yourself and still be in a system that rewards overconfidence or hides feedback. This isn't anti-self-improvement; it's against vagueness and silver bullets.

What Doesn't Change Much

Some aspects of how we judge stay fairly stable. Biases and random variation in decisions don't vanish with a weekend workshop. Not everyone improves to the same level—there are real limits. Acknowledging that isn't defeatism—it's what makes the "what does improve" claim credible. We're not saying everything is trainable. We're saying enough is trainable to be worth the effort, and the rest is worth working around (with structure and environment) rather than pretending we can fix it inside the person.

What to Do Instead

So what should you do? Seek feedback on important decisions when you can. Use structure: for example, run a pre-mortem on one upcoming decision—spend two minutes asking "If this goes wrong, what will have caused it?" before you lock it in. Improve the environment: who advises you, what's default, how outcomes are reviewed. And focus on specific domains where you care about getting better, rather than on "improving your judgement" in the abstract. Avoid programs that promise to upgrade your brain without saying what's actually changing or how.

The Real Question

The question isn't "Is my judgement fixed?" It's "Which parts of my judgement can I improve, and how—and which parts should I work around?" You have agency without needing to believe everything is infinitely malleable. That's the trade on offer: drop the binary, keep the hope, and aim at the levers that actually exist.

This week: Pick one decision you'll make soon and spend two minutes on a pre-mortem—what would cause it to fail?—before you lock it in.