Hot Water Can Freeze Faster Than Cold

The one drawing in this stack that argues with its own headline, in its own handwriting: “no single winner,” “many careful experiments DON’T see it at all,” “genuinely unsettled.” That hedge is not a hedge for effect. It is the actual answer, and this lesson is about why an honest answer can look like that.

Before you start. All you need is a thermometer and the patience to compare two numbers. The one idea this story leans on hardest is that a single reading on a thermometer never tells you the whole story of a cup of water : two cups at the same temperature can still differ in how much water they hold, how it is moving, and what is dissolved in it (Cold Is a Lie - fridge, AC, heat pump makes the same point about temperature in general).

The idea in 30 seconds

The Mpemba effect is the claim that a hotter sample of water can sometimes reach a frozen state before a cooler one, first reported by a Tanzanian schoolboy in 1963 and published with physicist Denis Osborne in 1969.
“The hot cup has to pass through the cold cup’s starting temperature” is a real argument, but only if the cup’s condition never changes except its temperature - and a real cup’s condition does change.
Evaporation removes water, so there is less of it left to cool and to freeze.
Convection stirs the cup on its own, which speeds up its heat loss and also means one thermometer reading does not describe the whole cup.
Supercooling means “frozen” is not a fixed instant at zero degrees, so two cups can disagree about which finish line they are even racing to.
The most careful controlled test done so far, in 2016, found no effect at all, and traced most reported sightings of it to exactly where the thermometer happened to sit.

Read the story · Skip to the numbers