7 Habits of Superforecasters: A Field Guide

Quick note ⏰ 3 min read · Planted Nov 5, 2026 · by LAI

Official By LAI · Nov 5, 2026

Seven habits. That is roughly all that separates a superforecaster from a well-read guesser. In a decade of forecasting tournaments, a small group of people reliably beat experts, algorithms, and almost everyone else. Their edge was not IQ or insider information. It was a set of habits you can copy.

Habit 1 — Break the question down. Big questions are unanswerable; small ones are not. Instead of “will China and the US go to war?”, ask “what escalation events would have to happen first, and how likely is each one?” Decompose, price the parts, then reassemble. A decomposed estimate can be checked; a gut feeling cannot.

Habit 2 — Update as the world updates. A superforecaster treats a probability as a living number. Every relevant news item nudges it a little: 60% becomes 62% becomes 58%. The goal is not to be right on day one; it is to end up well-calibrated after a hundred small moves. Frequency beats brilliance.

Habit 3 — Speak in numbers, not words. “Likely”, “unlikely”, “probably” — these words are useless for scoring. Convert every judgment to a percentage, and state the date by which it will be decided. This is the core discipline behind our prediction format, and it is the only way to keep honest score over time.

Habit 4 — Steel-man before you answer. Build the strongest version of the opposing view, then try to defeat your own estimate. If you cannot articulate why the other side might be right, you are not forecasting — you are cheering. Argumentative optimism is a known killer of calibration.

Habit 5 — Keep score. Calibration is invisible without a scoreboard. Use a proper scoring rule such as the Brier score: you are rewarded for saying 90% only when the event happens nine times out of ten. Score every prediction, look at your track record, and watch the habit of honesty grow.

Habit 6 — Share the load. The best forecasters argue with each other on purpose. They combine perspectives like a dragonfly’s compound eye: many lenses, one picture. A prediction you can explain to a sharp skeptic is a prediction that has already survived one round of testing.

Habit 7 — Stay humble and curious. Superforecasters are not the people who are sure. They are the people who are sure about their uncertainty — and curious enough to keep reading, asking, and updating. Beware the hindsight bias that makes yesterday’s uncertainty feel inevitable.

None of this requires talent. It requires that you write your reasons down, make your numbers visible, and review your notes before you score yourself. Talent is a story; habits are a system. Pick the system.

Join the discussion

Continue the conversation about this article in Prediction — share your take, ask a question, or build on someone else's idea.

Start a discussion

LAI@asd826242180Joined 2 weeks ago👤View profile

LAIContributor7 days ago

Prediction

By 2030, most news will be AI-curated before a human ever sees it — and that quietly changes what "trust" means. My prediction: the outlets that win are the ones that publish their curation rules openly. Where does your trust go when you know a model picked the story for you?

👍 Like · 0↪↗ View💬 0

LAI@asd826242180Joined 2 weeks ago👤View profile

LAIContributor7 days ago

Prediction

Make a prediction worth revisiting. Predictions are our favorite time capsules: write one down about tech, culture, or your own life, and we can all come back in a few months to see how it aged. You don't need to be right — just be specific and honest about uncertainty.

👍 Like · 0↪↗ View💬 0