Assuming You Might Be Wrong

Category: Judgement Is a System Component Humility acts as a functional advantage because uncertainty improves decisions.

A small weather-forecasting office issues a flood warning for a river valley, and the forecaster who signs it off does something that, to an outsider, looks oddly unconfident for a person whose job is to be certain.

She doesn’t just publish her number and move on. She writes, next to it, the things that would make her wrong: the upstream gauge she’s relying on hasn’t reported in three hours, so the figure feeding her model might be stale; the rainfall band could track ten miles north, in which case the whole valley is fine and she’s just evacuated a village for nothing; the soil saturation reading is a day old and a dry day would change everything. She names, in plain sight, the conditions under which her own warning is a mistake — and then she keeps watching those specific things rather than her forecast, because they’re the things that will tell her she’s wrong before the river does.

To a certain kind of observer this looks like weakness. Surely the confident forecaster, the one who states the number flatly and stands behind it, is the better one to have. Surely all this hedging is just a professional covering herself.

But watch what it actually buys her. When the upstream gauge comes back online two hours later reading lower than her model assumed, she already knows what it means, because she’d already marked that gauge as the thing most likely to be lying to her. She revises down, fast, while the confident forecaster next door — who never wrote down what would make him wrong — is still emotionally attached to the warning he committed to, and will defend it for another hour out of sheer momentum. Her humility isn’t modesty. It’s a faster correction mechanism. She catches her own error before the world catches it for her, and that is the entire difference between a good call and a disaster.

That habit — holding the conclusion firmly enough to act on, loosely enough to drop — is not a personality trait she happens to have. It is a designed condition, and it is one of the quietest engines under environments that make good decisions for a long time.

The Principle

Humility is not a virtue that good decision-makers happen to possess. It is a functional component of good decisions — the deliberate practice of holding your conclusions provisionally, actively assuming you might be wrong, and keeping a live watch for the evidence that would prove it. The people and teams who do this make better decisions not because they are nicer or more honest, but because they correct faster, and correcting fast is what good judgement actually is.

The widespread assumption is that confidence and competence travel together — that the person who is most sure is most likely to be right, and that doubt is something you feel when you don’t quite know what you’re doing. So we reward certainty, promote the people who project it, and quietly distrust the ones who say “I think” and “probably” and “unless.” That framing has it almost exactly backwards. Certainty is not evidence of being right; it is, very often, evidence of having stopped looking. The decision-maker who assumes they might be wrong keeps looking — and because they keep looking, they find the disconfirming thing early, while it’s still cheap to act on. Humility, here, is not about how you feel. It’s about what you keep doing after you’ve reached a conclusion.

Why It Is Inevitable

This isn’t an optional refinement the best decision-makers happen to add on top of being good. It’s something they’re more or less forced into, because the alternative fails in a way that is structural and unavoidable.

The world supplies information in the wrong order. You almost never get all the relevant facts before you have to decide; you get some of them, you decide, and then more of them arrive after, while the decision is already in motion. This means that being wrong is not a sign of having decided badly — it’s the ordinary condition of deciding at all. Every conclusion you reach is reached on partial information, which means every conclusion carries a standing probability of being wrong that no amount of care at the front end can remove. The only question is whether you have a mechanism for noticing when that probability cashes out.

The confident decision-maker has no such mechanism, because confidence is the absence of one. Having decided, they stop treating the question as open; the disconfirming evidence that arrives later lands on a closed mind and bounces off, because a closed mind isn’t watching for it. They don’t fail because they were stupid or careless at the moment of decision — they may have decided well on what they had. They fail because they treated a provisional conclusion as a finished one, and so had no way to catch it when the world moved underneath it.

The humble decision-maker is simply the one who accepts the order the information actually arrives in. They decide, because you have to, but they keep the question ajar — they hold the conclusion as the best current guess rather than the settled truth, and they keep a watch on the things that would overturn it. That isn’t a personality difference. It’s the only posture that survives contact with how information actually behaves, which is why any environment that wants to make good decisions over a long horizon is pushed toward it whether it intended to be or not.

How It Shows Up

• Conclusions are stated with their conditions attached — “this, unless that” — so the thing that would change the answer is named at the moment the answer is given, not discovered later.
• People say “I might be wrong about this” and mean it operationally: they’ve actually identified what would prove it and are watching for that thing.
• Challenge is invited rather than tolerated. Someone who has decided actively asks “what am I missing?” and wants a real answer, not reassurance.
• The strongest dissent in the room is sought out, not managed down — the person most likely to think the decision is wrong is asked to say why, in full.
• New evidence that contradicts a held position is treated as useful news rather than an attack, and the position updates visibly and without drama.
• Forecasts and predictions are revisited against what actually happened, so being wrong is noticed and learned from rather than quietly forgotten.
• Seniority does not confer immunity from the question. The most experienced person in the room is the one most likely to say “but I could be reading this from an old pattern that no longer holds.”

Why It Causes Benefit

When this is in place, an environment gets something that looks paradoxical: it is more decisive and more accurate at the same time, and the humility is the reason for both.

It’s more accurate because the humble decision-maker is running a correction loop the confident one isn’t. Having assumed they might be wrong, they keep looking for the evidence that would prove it — and so they find it early, while it’s small and cheap, before it has compounded into something expensive. The confident decision-maker finds out they were wrong at the same time everyone else does: when the thing has already failed in public. The humble one finds out privately, in advance, because they were the only person still watching. The gap between those two moments — between catching your own error and having it caught for you — is enormous, and it’s almost the whole of the difference in outcomes. Good judgement, in practice, is mostly just the early detection of your own mistakes.

It’s more decisive — and this is the part that surprises people — because holding a conclusion loosely actually makes it easier to commit to. The decision-maker who knows they’ll catch and correct an error doesn’t need to be certain before acting; the watch is their safety net, so they can move on a provisional conclusion without waiting for an impossible completeness. It’s the person with no correction mechanism who has to be sure before they act, because for them a wrong decision is unrecoverable — they won’t notice it in time. Humility, counter-intuitively, lets you act faster, because it makes acting on incomplete information survivable.

And there’s a compounding effect that runs through the whole environment. When the senior people visibly assume they might be wrong — when they ask what they’re missing and update without ego — they make it safe for everyone else to surface the disconfirming thing. The information that would correct a bad decision tends to sit with the most junior person in the room, who can see the flaw but daren’t name it to someone who’s clearly already sure. An environment that prizes humility unlocks that information; an environment that prizes certainty buries it. So the humble environment doesn’t just make individually better calls — it sees its own mistakes from every level, early, because nobody is afraid to be the one who points at them.

How To Cultivate It

• Ask for the conditions, not just the conclusion. When someone brings a decision, ask “what would have to be true for this to be wrong?” — and treat a good answer as a sign of strength, not weakness. The conclusion with its failure conditions attached is more useful than the confident one without them.
• Reward updating, never punish it. The person who changes their mind in the face of new evidence has done the hard and correct thing, and must be seen to gain standing by it, not lose it. The moment “you’ve changed your position” becomes a charge, people stop updating and start defending.
• Make seeking dissent a job, not a courtesy. Assign someone to argue the other side; ask the most sceptical person in the room to go first; build the search for disconfirming evidence into the process so it doesn’t depend on anyone feeling brave enough to volunteer it.
• Separate confidence from competence in how you promote and praise. If certainty is what gets rewarded, you will select for people who have stopped looking. Notice and value the ones who hold their conclusions provisionally — they are not less able, they are more correctable.
• Keep score honestly against what actually happened. Revisit predictions; note where the decision was wrong and why; make being wrong a normal, recorded, unembarrassing event rather than something quietly buried. You cannot learn to assume you might be wrong if you never check whether you were.
• Model it from the top, especially. The most senior, most experienced person saying “I might be reading an old pattern here” does more to make humility safe than any process, because it proves that admitting uncertainty costs nothing — and conceals nothing about competence.

What Good Looks Like

An environment where conclusions are held firmly enough to act on and loosely enough to drop, and where nobody confuses the two. Where the things that would make a decision wrong are named at the moment the decision is made, and then actively watched — so errors are caught early, privately, and cheaply, rather than late, publicly, and at full cost. Where challenge is sought rather than survived, where the most junior person can name the flaw the most senior one missed without fear, and where changing your mind in the light of new evidence raises your standing rather than lowering it. Where confidence is treated with mild suspicion and provisional holding with quiet respect, because everyone has understood that certainty is usually just the place where someone stopped looking. The decisions made here are both bolder and more accurate than a more confident environment would manage — bolder because the correction mechanism makes acting on incomplete information safe, more accurate because the looking never stops. And the humility that produces all this is not a matter of who happens to be modest. It is built into how the place works: a designed condition, not a collection of well-mannered individuals — which is what good judgement looks like when it has been made part of the system rather than left to character.

A Reflective Question

The last time you were confident about something important and turned out to be wrong — did you catch it yourself, early, or did the world catch it for you, late? And if it was the latter: was it because the evidence wasn’t there to be seen, or because, having decided, you had quietly stopped looking for it?