The binomial distribution models the number of successes in a fixed number of independent trials, each with the same probability of success. It applies when each trial has two outcomes and the success probability remains constant. section 2.3.6
The probability of exactly k successes is the number of ways to arrange them, multiplied by the probability of any such sequence. Here, n is the trial count and p is the success probability. section 2.3.6
The model combines the multiplication rule for one sequence with combinatorial counting across sequences that yield the same success count. It therefore evaluates counts, not the order of outcomes, and supplies a likelihood for comparing assumed probabilities with observations. section 2.3.6 section 2.3.4
Estimating a machine’s defect rate
Inspecting 10 parts and finding 2 defects gives a binomial likelihood for the unknown defect probability. Combined with a Beta prior, this likelihood updates the estimate to a Beta posterior. section 7.4.2






