Two doctors in the same city can give different diagnoses to identical patients. Two judges can hand down different sentences for similar crimes. Even we ourselves can make different decisions in the same situation depending on whether it is morning or afternoon, whether we are tired, or whether lunchtime is approaching. There is not always an objective reason to explain these differences. Irrelevant factors often influence our judgment. This variability is known as noise.

Noise is a problem because it affects decisions that should be consistent. We tend to think that our judgments are mainly based on the facts, but they also depend on who makes the decision, when they make it, and under what conditions. It is not the same for a doctor to make the same mistake consistently, which would be a bias, as it is for two doctors to reach different conclusions about the same case. Bias can be corrected once we identify it. Noise is harder to detect because it appears as a seemingly normal difference between decisions. And that is precisely why it deserves more attention: often we do not need better arguments, but systems that reduce the influence of factors that should never have been part of the decision.