The Sugar Pill That Works

The Sugar Pill That Actually Works · Open board

From the Mind Blown track. It draws the case people usually make about placebos: give someone a pill with nothing in it, and something in their body actually changes. That case is mostly right and mostly misleading at once, and this lesson is about untangling the two.

Before you start. You don’t need any medical training or statistics for this one. You just need to be willing to watch one number get taken apart, piece by piece, until only the real part is left.

This is a lesson about how trials are designed and read, not medical advice - nothing here says what any person should do about a symptom.

The idea in 30 seconds

Most of what gets called “the placebo effect” isn’t caused by the placebo at all.
A single group, however large, can never prove an effect - you need a control group to know what would have happened anyway.
People measured at their worst look better later purely by chance, the same way an unusually high dice roll doesn’t repeat. That’s regression to the mean.
Conditions that naturally fade, and patients being polite to kind researchers, both add fake “improvement” with no pill involved.
What’s left over after subtracting all that is real, but small, and shows up in what people report feeling, not in what a machine measures.
A control group isn’t there to subtract a powerful placebo effect. It’s there to subtract everything that isn’t the drug - and the real placebo effect turns out to be the smallest thing in that pile.

Read the story · Skip to the numbers