Does hot water really freeze faster than cold water?

Here is a question that has outlasted empires. Put a cup of hot water and a cup of cold water in the freezer. Which freezes first? Common sense says the cold one, obviously. But for more than two thousand years, clever people have claimed the opposite, and scientists today still cannot agree on who is right.

The short answer

It is disputed. The claim that hot water can freeze faster than cold is known as the Mpemba effect. Some experiments seem to show it; others find nothing at all once measurements are done carefully. For plain water in a freezer, there is no scientific consensus that it is real.

The real story

The idea is old. Aristotle, Francis Bacon and René Descartes are all reported to have noticed that warm water might cool or freeze more quickly than cold. But the modern name comes from a schoolboy in Tanzania.

In the 1960s, Erasto Mpemba was making ice cream at Magamba Secondary School. Freezer space was scarce, so instead of letting his hot milk and sugar mix cool down like his classmates did, he put it straight in. Around an hour and a half later his batch had frozen, while theirs were still a thick slush. When he asked his physics teacher why, he was told he must have been confused, because that could not happen.

Mpemba did not let it go. He later teamed up with physicist Denis Osborne, who helped him run proper experiments. The pair published their results in 1969 in the journal Physics Education, and the effect has carried his name ever since.

Then the arguments began. In 2012 the Royal Society of Chemistry ran a public competition inviting people to explain the effect, reviving interest among scientists and the public alike. In 2016, Henry Burridge and Paul Linden of the University of Cambridge took a hard look at the evidence in the journal Scientific Reports. They found that results were extremely sensitive to tiny details, such as exactly where the thermometer sits in the cup. Shifting it by around a centimetre could make the effect seem to appear. Their conclusion, delivered with a touch of melancholy, was that there was no evidence for meaningful observations of the effect. Notably, they said Mpemba and Osborne’s original results were the only ones they surveyed showing a difference too large to blame on measurement error.

The story does not end there. In 2020, a team including physicist John Bechhoefer published a study in Nature showing a strong Mpemba-like effect in a very different system: tiny glass beads moved around by lasers. Theorists have shown that, in principle, some hotter systems can find shortcuts to cooling. Whether any of that applies to a cup of water in your kitchen freezer remains an open question.

Why people get it wrong

The claim is so counterintuitive that it feels like it must be a clever scientific secret, and it has a great origin story attached. Lists of amazing facts tend to present it as proven, skipping the decades of disagreement. Meanwhile, sceptics often dismiss it as impossible, which is not quite right either. The honest answer sits in the uncomfortable middle: maybe, sometimes, under conditions nobody has fully pinned down.

“I’ve heard this one argued under more desks than I can count. Still no winner. Honestly, mate, I’d just buy a bag of ice.”