In 2014, a group of researchers in Australia set out to test what many people now suspect: that the ice bath placebo effect explains most of what cold water immersion does for recovery. They designed a clever trial, published the results, and the finding was striking. A fake recovery treatment worked just as well as actual cold water for perceived recovery and muscle strength. The honest answer is: partly, yes. Belief and expectation drive a meaningful share of the recovery people report after cold water immersion.
But that finding, striking as it is, tested one layer of recovery and was silent about the others. The same body of evidence that shows perception matters also shows that cold water triggers measurable neurochemical and neurological changes that have nothing to do with what you believe. And a separate line of research demonstrates that perceived recovery the thing sceptics dismiss as “just placebo” reliably predicts actual physical performance afterward.
The study everyone cites only tested one part of it.
What the Broatch study actually found
In 2014, James Broatch, a postdoctoral researcher at Victoria University, led a randomised controlled trial published in Medicine & Science in Sports & Exercise. It remains one of the best-designed investigations of whether cold water immersion outperforms belief alone for exercise recovery.
Broatch and his team recruited 30 recreationally active men and had them complete a high-intensity cycling protocol designed to induce muscle damage. Afterward, the participants were split into three groups. One group sat in cold water at 10°C for 15 minutes standard CWI protocol. A second group sat in thermoneutral water at 34°C, but with a twist: they were told a “recovery oil” had been added to the water and given a professional-looking brochure explaining its performance benefits. In reality, the oil was a skin cleanser. It did nothing. A third group sat in the same thermoneutral water with no story attached.
Over the next 48 hours, the researchers tracked perceived recovery, muscle soreness, and isometric strength. Both the cold water group and the fake-oil group reported nearly identical recovery outcomes. Both felt significantly better than the plain warm water group. On muscle strength measures, the pattern held. The placebo condition matched genuine cold water immersion.
What this implied was hard to dismiss: a large portion of what people experience as recovery from CWI appears to be driven by their expectation that it will work. Writing in The Conversation, Broatch framed the finding not as a debunking but as a revelation about how powerful belief is in the recovery process. Even the researcher behind the study didn’t conclude that ice baths are useless. He concluded that we had underestimated the machinery of expectation.
Subsequent work supports this. A 2025 study in the European Journal of Applied Physiology by Romare and colleagues tested national-level soccer players and found CWI was no more effective than a sham laser placebo. The authors wrote that “recovery is not solely a physiological process.”
What the study didn’t measure
Broatch’s design was smart, but it only measured specific things: subjective recovery ratings, muscle soreness, isometric force. It didn’t measure what happens inside the body when cold water hits skin, regardless of what anyone believes about recovery oils.
A 2000 study in the European Journal of Applied Physiology by Šrámek and colleagues immersed participants in water at 14°C for one hour and measured the biochemical aftermath. Plasma norepinephrine increased by 530 percent. Norepinephrine is involved in attention, arousal, mood regulation, and the body’s acute stress response. The increase was dose-dependent, reproducible, and measured from blood draws, not questionnaires. No belief system produces a fivefold spike in a catecholamine.
Šrámek’s team also recorded a 250 percent increase in plasma dopamine, though this deserves a caveat: peripheral dopamine measured in blood doesn’t translate directly to dopamine activity in the brain. The norepinephrine data is the more defensible claim, and it alone establishes that cold water immersion produces objective physiological changes independent of expectation.
Then there is the neural evidence. A 2023 fMRI study published in Biology by Dr Ala Yankouskaya, a senior lecturer in psychology at Bournemouth University scanned participants’ brains before and after cold water immersion. Scans showed altered connectivity between regions involved in attention control, emotion regulation, and self-regulation — structural changes in how the brain’s networks communicated, visible on imaging, not dependent on self-report.
“These are the parts of the brain that control our emotions, and help us stay attentive and make decisions,” Yankouskaya explained. “So when the participants told us that they felt more alert, excited and generally better after their cold bath, we expected to see changes to the connectivity between those parts.”
Broatch tested one channel of recovery and found it was strongly influenced by belief. Yankouskaya and Šrámek tested different channels and found changes that belief cannot explain. The neurochemistry happens whether or not someone hands you a brochure. The brain rewires whether or not you think the water contains a magic oil. The leap from “expectation drives perceived recovery” to “cold water does nothing real” requires ignoring every objective measurement that sits outside that single experiment’s scope.
When “placebo” stops being a dismissal
Even if we accept the Broatch finding at full strength there is a follow-up question most commentary skips: does perceived recovery actually matter?
A 2013 study in the British Journal of Sports Medicine by Cook and Beaven investigated exactly this. They examined the relationship between athletes’ subjective ratings of their recovery intervention and their subsequent physical performance. When integrated with physiological factors, perceived recovery predicted actual sprint performance with a correlation of r=0.77. Athletes who believed they had recovered well went on to perform better — not just feel better, but move faster, measured by sprint times.
This reframes the entire placebo conversation. If perceived recovery reliably predicts subsequent performance, then perception may not be a confound to be controlled for — it could be a mechanism. Calling something “just placebo” implies it is inert, a trick that dissolves under scrutiny. When the trick appears to produce measurable performance outcomes, the word “just” stops doing useful work.
And the pattern holds. A 2018 trial with professional soccer players by Bouzid and colleagues found that CWI improved physical performance recovery without reducing creatine kinase, a standard marker of muscle damage. Tissue didn’t recover faster by measurable blood markers, but their bodies performed as if it had.
Recovery, as measured by how a person subsequently moves and performs, appears to be a composite outcome. Some of it is tissue repair. Some of it is neurochemical. And some of it is perceptual — shaped by what you believe, what you expect, and how the experience felt. Dismissing one channel because it involves belief is like dismissing sleep because part of its benefit is psychological. The outcome is the outcome.
Why environment is not a footnote
If recovery operates on these simultaneous layers then the practical question becomes: what shapes the perceptual and ritual layers?
Environment. Framing. Commitment. Preparing for cold water, stepping in, breathing through the shock response, and stepping out creates a psychological event distinct from both the neurochemistry and the passive expectation. Stepping in becomes an act of agency. A boundary between one state and another. And the context in which it happens shapes the belief that the Broatch study proved so powerful.
A clinical tub in a fluorescent room and a considered, well-designed bath in a calm space may trigger identical vasoconstriction. They are unlikely to create identical experiences. Expectation differs. So does the sense of commitment, and the ritual quality. According to the very research sceptics cite, those differences may matter for outcomes.
A simple phrase captures the design philosophy behind the ice baths we build: the engineering disappears, the ritual remains. Temperature accuracy, insulation, water management all of it exist so the experience itself takes centre stage. If perception is a mechanism of recovery, then everything that shapes perception is functional, not decorative.
This logic extends to hospitality. How someone is guided into their first cold water immersion affects their expectation, their confidence, and therefore their recovery. Broatch’s fake recovery oil worked because it came with a professional brochure and a credible story. Good hospitality programming does something functionally similar: it frames the experience with competence and care, which primes expectation, which shapes outcome. Whether you’re stepping into your own ice bath at home or being guided through one at a hotel, the introduction is not separate from the effect.
At the W Hotel Brisbane, the ice baths installed see more than 50 plunges a day from repeat guests. That is not clinical proof. But it is behavioural evidence that whatever the mechanism; physiological, perceptual, ritualistic, the outcome is consistent enough for people to come back.
The better question
The debate over whether ice baths are “real” or “placebo” assumes those are opposites. The evidence says they are not. Cold water triggers a neurochemical cascade that is measurable and independent of belief. It also triggers a perceptual shift that is shaped by expectation and context. And the act of immersion itself carries psychological weight that neither blood work nor questionnaires fully capture.
Broatch’s finding is real. Belief matters enormously. But the conclusion most people draw from it – that ice baths are therefore fake – mistakes one layer of recovery for the whole thing.
The better question was never “is it placebo?” It was “what if recovery has always been more than one thing, and we just didn’t have a framework for it?” Because if perception is a genuine mechanism, designing for it is not a concession to placebo. It is taking recovery seriously in its entirety.
