Bayes' theorem in the form that actually matters day to day: why a 99%-accurate test on a rare condition is still usually wrong, why doctors and juries fall for the same reasoning error, and how the same math runs underneath spam filters and alert triage. Priors, likelihoods, and posteriors made concrete with natural-frequency arithmetic instead of symbol-pushing.
Probability
-
Bayes' Theorem for Engineers: The Base-Rate Trap That Fools Doctors, Juries, and Alert Dashboards -
Bayesian Statistics for Engineers Bayesian and frequentist statistics answer different questions, and engineers benefit from knowing which question they are actually asking. This post walks through priors, conjugate posteriors, a worked Bayesian A/B test, MCMC at a working level, and where Bayesian reasoning genuinely beats classical inference.
-
The Kelly Criterion: Position Sizing as an Engineering Problem John Kelly's 1956 formula for optimal bet sizing is not finance mysticism — it is applied information theory. This post derives the criterion from first principles, explores fractional Kelly, quantifies sensitivity to estimation error, and shows why treating position sizing as an engineering problem changes how you think about risk.