A lesson from Yale's Introduction to Medical Software course traces the origin of the t-distribution to William Sealy Gosset, a chemist at the Guinness Brewery who needed a way to reason from small sample sizes and published under the pseudonym Student. The instructor explains why the distribution's fatter tails matter for judging extreme values, then moves into how it functions in clinical trial design today: power analysis, the tradeoff between Type I and Type II errors, and how researchers set sample sizes before running a study. The lecture closes with a caution about how statistical significance and p-values get misread in practice, including the problem of p-hacking, where researchers rerun tests until they find a result worth publishing. It is a compact, applied statistics lesson aimed at students who need to evaluate medical research rather than just compute p-values.