Subtle... I like the ones that exploit exponential effects or the birthday paradox. For example: Men are <10% taller than women on average. But if you look at

Subtle... I like the ones that exploit exponential effects or the birthday paradox. For example: Men are <10% taller than women on average. But if you look at a largish group, the probability that the tallest person is a man is ~99%. With a bit of writing skill, that probability can be used to make the difference seem like 20%, 40% or even more, whatever the writer wants.