Evaluate evidence by its likelihood ratio, not by how it makes you feel

Ask how much more likely this evidence would be if you’re right versus if you’re wrong.

Why it works

Not all evidence is equal; its strength is determined by its likelihood ratio: how much more (or less) probable the evidence is in the hypothesis-true world versus the hypothesis-false world. Evidence that is equally likely under both hypotheses carries zero information. This framing cuts through the common mistake of treating "evidence I like" as strong evidence and "evidence I dislike" as weak.

How to do it

1. State both hypotheses explicitly: H₁ (what you believe) and H₂ (the alternative).
2. Ask: "Would I see this evidence more often if H₁ were true, or if H₂ were true?"
3. If H₁ makes the evidence much more probable, it is strong evidence for H₁.
4. If H₂ also predicts the evidence easily, the evidence barely updates you.

Evidence

The likelihood ratio is the formal measure of evidential strength in Bayesian statistics and signal detection theory. Its use as a practical reasoning heuristic is supported by calibration research showing it outperforms intuitive evidence weighting. (mechanistic)

Applying likelihood ratios verbally is an approximation of the formal method; the formal Bayesian apparatus requires quantified probabilities that real-world beliefs often lack.

Common mistake

Treating evidence as strong just because it exists — "there is evidence for X" — without asking how much more common that evidence would be in a world where X is true versus false.