Null Hypothesis

The null hypothesis (written H₀) is the default, sceptical claim that a test tries to knock down — usually “no effect,” “no difference,” or “only chance is at work.” For an A/B test it might be “the new button has the same conversion rate as the old one.”

Everything in a classic test is computed assuming the null is true; the whole machinery asks how compatible the data are with that boring world. The null is deliberately specific (a single number, like difference = 0) because you can only compute probabilities against a precise reference.