What is Bayes theorem in simple terms?

Bayes theorem is a way to update a probability when new evidence is observed. You start with a prior probability, combine it with how likely the evidence is under different possibilities, and end with a posterior probability. In plain language, it answers: now that I saw this evidence, how likely is the original event?

What is a prior probability?

A prior probability is the probability before the new evidence is considered. In a medical-style example, the prior may be the base rate of a condition before the test result is known. Bayes theorem shows why this starting probability matters.

What is a false positive in Bayes theorem?

A false positive means the evidence appears even when event A is not true. In the calculator, this is P(B|not A). If false positives are common, the posterior probability P(A|B) can be much lower than people expect, especially when A is rare.

Why is P(B) in the denominator?

P(B) is the total probability of seeing the evidence. It includes all ways B can happen, not just the route through A. Dividing by P(B) turns the numerator into a properly scaled conditional probability.

Is Bayes theorem only used in medical testing?

No. Medical testing is a common classroom example, but Bayes theorem is also used in spam filtering, risk analysis, diagnostics, data science, legal reasoning, machine learning, and any situation where beliefs are updated after evidence.

Education

Bayes Theorem Student Calculator

Use Bayes theorem to update a probability after evidence appears. This student-focused calculator shows the numerator, denominator, posterior probability, and the meaning of each part.

What Bayes theorem calculates

Bayes theorem updates a probability after new evidence appears. It connects the prior probability P(A), the likelihood P(B|A), and the overall probability of the evidence P(B). OpenStax describes Bayes’ theorem as a way to revise probability estimates in its exact probability theory section from Principles of Data Science.

This calculator uses the common two-condition version students see in introductory probability: event A happens or does not happen, and evidence B is observed. That setup is useful for test-result problems, classification examples, false-positive examples, and many classroom word problems.

Bayes theorem formula

P(A|B) = [P(B|A) × P(A)] ÷ P(B) For two cases: P(B) = P(B|A)P(A) + P(B|not A)P(not A)

The denominator matters because evidence B can happen in more than one way. A positive test result, for example, may come from a true positive or a false positive. Bayes theorem compares those routes instead of assuming the test accuracy alone is the final answer.

Why Bayes theorem feels counterintuitive

Bayes problems often surprise students because a strong likelihood does not always mean a high posterior probability. If the original event is rare, false positives can still make up a large share of positive results. This is why the calculator shows the numerator and the total evidence probability separately.

Frequently asked questions

Bayes theorem is a way to update a probability when new evidence is observed. You start with a prior probability, combine it with how likely the evidence is under different possibilities, and end with a posterior probability. In plain language, it answers: now that I saw this evidence, how likely is the original event?
A prior probability is the probability before the new evidence is considered. In a medical-style example, the prior may be the base rate of a condition before the test result is known. Bayes theorem shows why this starting probability matters.
A false positive means the evidence appears even when event A is not true. In the calculator, this is P(B|not A). If false positives are common, the posterior probability P(A|B) can be much lower than people expect, especially when A is rare.
P(B) is the total probability of seeing the evidence. It includes all ways B can happen, not just the route through A. Dividing by P(B) turns the numerator into a properly scaled conditional probability.
No. Medical testing is a common classroom example, but Bayes theorem is also used in spam filtering, risk analysis, diagnostics, data science, legal reasoning, machine learning, and any situation where beliefs are updated after evidence.