Revise belief in proportion to how much more likely the evidence is under one hypothesis than another.
The practical form: ask how surprising the evidence would be if you were wrong. Evidence that both hypotheses predict equally well should move nothing, however dramatic it feels.
In practice
- A test comes back positive; the useful question is how often it comes back positive for people who do not have the condition.
- An outage coincides with a deployment, but deployments happen daily and outages are rare, so the coincidence is weaker evidence than it feels.