Why does hot water sometimes freeze faster than cold water?
Sometimes it does and sometimes it does not — and after sixty years of attempts, nobody has produced an experiment that reliably shows it or a mechanism everyone accepts.
Simple intuition
The plain reason, in everyday words
The claim sounds impossible: to freeze, hot water must first cool to the temperature the cold water started at, so it seems it can never arrive first. That argument would be watertight if temperature were the only thing that changed on the way down. It is not. Water that has been heated has lost some of its mass to evaporation, has lost dissolved gases, is moving inside the container because warm water circulates, and may have melted the frost under its container so that it sits in better contact with the freezer shelf. Any of those could plausibly speed up the second half of the journey. So the effect is not obviously impossible. What it has never done is turn up reliably when someone sets out to measure it carefully, which is a different and more interesting problem.
It is a well-established fact of physics.
It is a well-established anecdote. Careful controlled attempts have largely failed to reproduce it, and the most thorough recent study concluded that the apparent effect tracked measurement position rather than starting temperature.
It is obviously impossible, because hot water must pass through cold on the way.
That argument assumes the two samples are identical once they reach the same temperature, and they are not — heating changes mass, dissolved gas content and internal motion. The claim is not ruled out by logic, it is simply unsupported by measurement.
Evaporation explains it.
Evaporation is real and does remove mass from the hot sample, but the magnitude is too small for the larger reported effects, and the effect has been reported in sealed containers where evaporation cannot operate.
Because the water version is doubtful, the whole idea is nonsense.
A rigorous version exists for systems relaxing toward equilibrium and has been demonstrated in controlled experiments with colloidal particles. What is doubtful is specifically the claim about water in a freezer.
It is a genuinely useful example of what an open question looks like from the inside. There is a real observation, a memorable story, several plausible mechanisms, and still no answer — and the honest response is to say so rather than to pick the most satisfying explanation. It also shows how much of an experimental result can live in the apparatus rather than the phenomenon: moving a thermometer a few centimetres changed the conclusion.
Who worked it out
Aristotle recorded that people wishing to cool water quickly began by standing it in the sun, and both Francis Bacon and René Descartes later described something similar.
What problem forced it
The modern investigation began in 1963 when Erasto Mpemba, then a secondary school student in Tanzania, observed it while making ice cream and pressed the question against considerable discouragement; he and the physicist Denis Osborne published in 1969.
How it changed since
Decades of proposed mechanisms followed without resolution. A 2012 competition produced many explanations and no consensus, a 2016 study failed to reproduce the effect under controlled conditions, and from 2017 the theoretical idea found firmer ground in the study of how systems relax toward equilibrium.
Supercooling and why water resists freezing
The strongest surviving candidate mechanism, and interesting on its own terms.
What makes a result reproducible
This case is a compact lesson in how measurement choices can manufacture an effect.
Written for Curio rather than collected from a forum — it is part of the curated corpus that ships with the platform. The references it draws on are listed under Sources.