Cognitive Biases · Learning

Overfitting

Tuning a model or rule so closely to past data that it fits the noise, so it explains the past well and predicts the future poorly.

Evidence grade: Statistical. How we grade evidence

For example

A team builds a scoring formula with dozens of tweaks until it matches last year's grant outcomes almost perfectly. Applied to this year's applicants, it ranks them badly.

DECIDE BETTER® · Rule 3Data answers your questionEvery data set answers somebody's question. Make sure it's yours, or you end up precisely wrong and only generally right.

Spot it in your week

Three quick questions. Your answers stay on this page.

  1. This week, did you adjust a formula or forecast until it matched past results almost exactly?
  2. In a discussion this week, did anyone praise a model mainly because it fit last year's numbers so well?
  3. Did you read a plan this week that leaned on a rule built entirely from past cases?

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