Practice probabilistic calibration by tracking your predictions

Assign explicit probability estimates to your predictions and track whether they come true at the right rate.

Why it works

Calibration — the match between stated confidence and actual accuracy — is a learnable skill. People who track their predictions over time and observe how often 70% confident claims actually come true develop feedback loops that correct systematic over- or under-confidence. Without this feedback, the confident feeling generated by vivid information persists uncorrected.

How to do it

1. For any prediction you make, assign a percentage probability: "I think there’s a 70% chance X happens."
2. Record the prediction and the outcome.
3. After 20–30 predictions, check your calibration: are your 70% predictions coming true about 70% of the time?

Evidence

Calibration research (including work on superforecasters by Tetlock) shows that tracking predictions with explicit probabilities significantly improves accuracy over time through feedback-driven error correction. (observational)

Sources

• Tetlock & Gardner (2015), Superforecasting — calibration training and improvement across thousands of predictions

Common mistake

Assigning vague probabilities ("pretty likely") rather than specific numbers, which prevents meaningful calibration checks because the prediction is unfalsifiable.