Baserate

Base rate, sometimes written as baserate, refers to the overall probability of an event occurring in a population before considering any individual evidence. It describes how common a condition, attribute, or outcome is within a defined group and acts as a prior probability in Bayesian reasoning. Base rates are independent of specific cases and are determined by population characteristics such as prevalence, incidence, or base frequencies.

A related concept is the base-rate fallacy, where people ignore base rates and focus on information about a particular case. For example, hearing that a medical test is 90% accurate may be misleading if the condition being tested is rare in the population; without considering the base rate, one might overestimate the likelihood that a positive result indicates actual disease.

In practical terms, the relationship between base rate and evidence is formalized by Bayes’ rule. If A denotes having a condition and B denotes a positive test result, then the posterior probability is P(A|B) = [P(B|A)P(A)] / P(B), where P(B) = P(B|A)P(A) + P(B|not A)P(not A). This shows how base rate (P(A)) and test characteristics (P(B|A), P(B|not A)) combine to determine the probability that a person has the condition given a positive result.

Base rates are central in fields such as medicine, epidemiology, risk assessment, marketing, and machine learning. They guide interpretation of diagnostic information, forecasting, and decision-making, and help prevent misinterpretation that arises when base rates are ignored.