Why Does the Shower Curtain Attack You?

This is part six of Lunch Break Physics, the series that has argued a suction cup isn’t sucking, a fridge doesn’t make cold, the sky isn’t blue because it reflects the sea, ice floats because water freezes into a roomy cage, and planes fly by throwing air downward. Every episode so far has ended with a clean answer. This one doesn’t, and that turns out to be the interesting part. Standard physics below; sources at the bottom; corrections very welcome in the comments.

The story you were told

Ask around and you will hear two confident explanations. The first is static electricity: the curtain builds up a charge and clings to your wet leg. The second is that hot air rises, and as it climbs out over the rail it drags the curtain up and in.

Both are wrong, and both are easy to rule out at home. If it were static, a cold shower would not do it, and yet a cold shower attacks you just as reliably as a hot one. If it were purely hot air rising, the same cold shower would leave the curtain hanging still. It doesn’t. Whatever is going on survives having the heat taken away, which quietly kills the two answers most people reach for first.

Watch it happen

Turn the water up and watch three things move together: the spray drags air down with it, the pressure inside the stall drops, and the curtain leans in. Turn it back off and the curtain hangs straight again.

One honest caveat is built into the picture. It shows the outcome everyone agrees on, lower pressure inside pulling the curtain in, but it does not claim to show why the pressure drops. As you will see in a moment, that is the part still being argued over, so the demo deliberately stays out of the fight.

The actual physics, or the part we agree on

Here is the one thing every explanation shares. When the shower runs, the air inside the curtain ends up at a slightly lower pressure than the still air of the bathroom. Air always pushes from high pressure toward low, so the bathroom leans on the outside of the curtain harder than the shower air leans back on the inside, and the curtain moves in.

If that shape of argument feels familiar, it should. Back in part one a suction cup stuck to the wall for exactly this reason: higher pressure outside, lower pressure in the sealed pocket, and the atmosphere pressing the cup flat. This is the same family. A difference in air pressure across a thin barrier does the work. The twist is that in the suction cup I made the difference on purpose, by squeezing the air out and sealing the gap, so the cause was never in doubt. In the shower, the running water makes the difference for us, and how it manages that is where the agreement ends.

Where it stops being settled

This is the first time in the series I have to stop and tell you the textbook has a blank page. There are at least three serious candidates for what lowers the pressure, and they genuinely disagree.

The Bernoulli explanation is the popular one. The spray drags air along with it, moving air has lower pressure than still air, so the fast-moving air on the inside face of the curtain is at lower pressure and the curtain gets pushed toward it. It sounds tidy, and it is the answer you will find most often. The trouble is that the one person who actually simulated the whole thing says it doesn’t hold up: Bernoulli’s principle balances pressure against acceleration in a clean fluid, and a shower is full of water droplets that change the accounting [1].

The buoyancy or chimney explanation says the warm shower heats the air, the warm air is less dense and rises out over the rail, and cooler bathroom air pushes in low to replace it. This one has the same fatal problem as the folk version: it predicts that a cold shower should be safe, and cold showers still pull the curtain in [1].

The vortex explanation is the most interesting, and it comes with a good story. In 2001 David Schmidt, a mechanical engineer at the University of Massachusetts Amherst, ran a proper computer simulation of a shower on his home PC, in the evenings, over two weeks, splitting a model bathroom into fifty thousand cells and solving the equations of fluid motion for thirty seconds of shower time [1]. He found that the spray drives a horizontal vortex, a slow sideways whirlpool of air with a low-pressure core, and it is that core that tugs the curtain.

Why would a whirl of air have a low-pressure heart at all? For the same reason a tornado or a dust devil does. Any air circling an axis is being forced off a straight line, and turning a corner needs a force pulling inward, toward the centre. In a free-standing vortex the only thing that can supply that inward pull is the air pressure itself, so the pressure has to be lower at the axis than out at the rim — a genuine low-pressure hole running down the core of the spin [4]. Schmidt’s shower vortex is a sideways version of that, fed continuously by the spray so it never spins down, and the curtain simply happens to hang right alongside its low-pressure centre. He described it as a bit like a sideways dust devil, using fluid-simulation software normally aimed at jet engines [3], and the work earned him a 2001 Ig Nobel Prize [2].

So which is it? The honest answer is that we don’t know for certain. Schmidt’s vortex is the best-supported candidate, but it is one simulation of one model bathroom, and reference works still list the shower-curtain effect among the unsolved problems in physics, with the Bernoulli, buoyancy, vortex, and even boundary-layer ideas all still on the table [2].

A little bit of math

You can see why the effect is so feeble with one line. The force on the curtain is the pressure difference times the area it acts on:

Force  =  (pressure difference)  ×  (curtain area)

The pressure difference here is only a few pascals, call it 5 Pa, against the roughly 101,000 Pa of the atmosphere. A curtain is about 0.7 metres wide and 1.8 metres tall, so roughly 1.3 square metres. That gives a force of about 5 × 1.3 ≈ 6 newtons, the weight of a couple of apples, spread over the whole sheet. That is plenty to swing a light plastic film and nowhere near enough to move a heavy vinyl curtain, which is the single most useful thing this number tells you.

The same trick, elsewhere

A moving fluid dropping the pressure alongside a surface, and something nearby getting pushed toward it, shows up all over the place once you know to look. These are genuine members of the family, not the disputed part.

• Two trucks passing on the motorway. As they draw level, the air squeezed between them speeds up and its pressure drops, and each vehicle feels a nudge toward the other. Motorcyclists and cyclists feel it far more sharply, which is why a fast overtake can tug you sideways.
• Ships sailing close and parallel. The same low-pressure channel opens up in the water between two ships running side by side, and they can be drawn together hard enough to be a real navigational hazard; it is a standard piece of seamanship to keep clear of it.
• A perfume atomiser or an old scent bottle. Squeeze the bulb, blow a fast jet of air across the top of the little tube, the pressure there drops, and the liquid is pushed up the tube and into the jet. Same physics, in a bottle.

The common thread is a moving stream of fluid, a drop in pressure beside it, and a nearby object pushed toward the low-pressure side.

False friends

These look like they belong in the same story, and they don’t. This is where a careful reader should poke hardest, especially since our own mechanism is unsettled.

• Static cling. The very myth we started with. Static electricity is real and it genuinely can make a dry curtain stick to a dry leg, but it runs on charge, not air pressure, and it doesn’t need the shower running. Different force entirely.
• The suction cup from part one. This is the family resemblance that fools people. It is the same pressure-difference outcome, but the cause is completely different: the suction cup holds a sealed, static low-pressure pocket that you created by pushing the air out, while the shower’s low pressure is made continuously by moving water and moving air. Same family, different cause, which is exactly the trap this section exists to flag.
• The curtain sticking to a wet wall. Sometimes the curtain clings to a tiled wall it happens to touch. That is a water film and surface tension gluing two wet surfaces together, adhesion, not a pressure difference across the curtain.
• A draught from the open bathroom door. If the door is open and the house has a breeze, bulk airflow can shove the curtain around. That is just wind, an actual current of air pushing on the sheet, not the shower quietly lowering the pressure on one side.

The tell for a false friend is to ask whether a moving fluid is lowering the pressure next to a surface. Static cling, a wet-wall stick, and a hallway draught all say no.

Fun consequences

So the next time the curtain lunges at you, you can say two true things with total confidence: the bathroom air is pushing it in because the pressure inside dropped, and nobody can tell you for certain what dropped it. For a phenomenon this everyday, that second half is oddly wonderful. Physics still has a blank page taped to the bathroom wall.

Lunch Break Physics runs Tuesdays at noon. Last week: how planes generate lift. This week’s episode is the first one where the textbook shrugs, and I think that’s the best kind. Next Tuesday: how noise-cancelling headphones work, where the answer is that you make it quieter by adding more sound. Got an everyday-physics puzzle you’d like poked at? The comments are open.

Sources

• [1] Why Does the Shower Curtain Move Toward the Water? — David Schmidt, Scientific American (2001) — the researcher’s own account: he argues the Bernoulli and buoyancy explanations both fail (buoyancy predicts a cold shower would be safe, and it isn’t), and his 50,000-cell CFD simulation instead finds the spray drives a low-pressure horizontal vortex that pulls the curtain in; the force is weak, so only light curtains are affected and weighting the hem fixes it.
• [2] Shower-curtain effect — Wikipedia — catalogues the effect under “unsolved problems in physics” and lays out the competing hypotheses (buoyancy/stack effect, Bernoulli, Schmidt’s horizontal vortex, and the Coandă boundary-layer effect) with no settled conclusion; notes Schmidt received the 2001 Ig Nobel Prize in Physics for the partial solution.
• [3] Run rings around a curtain — The Guardian (2001) — reports Schmidt’s use of computational fluid-dynamics software normally used to model jet engines to find the vortex spinning at the bottom of the shower as the cause of the curtain’s behaviour.
• [4] Tornado: physical characteristics — Encyclopædia Britannica — a vortex core is “a roughly cylindrical area of lower atmospheric pressure” bounded by the fastest circulating winds: the air turning around the axis requires an inward force, supplied by a pressure that is lower at the centre than at the rim, so a sustained vortex carries a low-pressure core.

About the Author

Stefan Christoph is a Principal Solutions Architect at AWS, focused on agentic AI, media & entertainment, and helping builders move from demo to production. He writes about AI architecture, developer productivity, and the future of software.

This is a personal blog. Opinions expressed here are my own and do not represent the views or positions of my employer.

Learn more →

• Physics
• LunchBreakPhysics
• Science
• Curiosity