Best Practices · Learning

Calibration

Also called: Probability calibration, Confidence ranges, Brier score (related)

Stating forecasts as probabilities or ranges, keeping score and adjusting. Well-calibrated people are right about as often as their stated confidence says they should be.

Evidence grade: Strong. How we grade evidence

For example

" No one tracks how often those forecasts come true, so her confidence is never tested against results.

DECIDE BETTER® · Rule 1Set the bar firstMake your gut testify. Before you open the report or run the model, write down what you expect and the result that would change your mind. Say it to your team.

Spot it in your week

Three quick questions. Your answers stay on this page.

  1. Did you say you were sure about something this week without knowing how often you turn out right?
  2. In a meeting this week, did people state forecasts in vague words like "likely" with no way to check them later?
  3. Did you read a forecast this week that gave no range or probability?

Your DecisionDNA™ can tell you which biases you're most exposed to.

Find your DecisionDNA™

Sources

Related entries

Get the countermeasure

Get the countermeasure and the decisions where this bites, free in the Volley app.

See the Volley app