There’s no single correct protocol for either modality — only sourced ranges, each labeled by how firm the evidence actually is. For post-exercise soreness, cold-water immersion studies cluster around 11–15°C for 11–15 minutes; novices should start far shorter, around 1–3 minutes. Traditional Finnish sauna runs roughly 80–100°C, with one long-running cohort averaging about 14 minutes a session. Contrast therapy is typically heat-then-cold for 2–4 rounds — a common convention, not a tested protocol. And the popular “11 minutes a week” cold target traces to a podcast naming, not a study finding — a plausible extrapolation, not a proven dose.
Why we give ranges, not a protocol
Search any of the practical questions this page answers — how long, how hot, how cold, how often — and the top results are mostly brand and retailer blogs, each presenting one exact protocol as if it were the single correct answer: a specific number of minutes at a specific temperature, repeated a specific number of times, with little or no indication of where the number came from or how firm the evidence behind it actually is. Some of those pages cite a real study; most just state a number. The ranges below come from the same handful of studies and cross-vendor conventions that those pages draw from — the difference here is that every figure carries a tier, so you can tell a controlled finding from a common practice before you build a routine around it.
Two tiers do almost all the work on this page: Literature, meaning the figure comes from a named, verifiable study, and Convention, meaning it’s a cross-source practitioner consensus with no controlled trial behind it. A third tier, Manufacturer, appears exactly once below, for infrared cabin temperature — that one is an equipment spec, not a health dose, and we say so plainly rather than let a product spec sheet masquerade as a protocol.
One more rule threads through this entire page, not just this section: for the same nominal strength of evidence, cold-plunge figures are read one tier more skeptical than sauna figures. That’s a standing rule for this hub, not a judgment call made fresh here — cold-water immersion carries a materially higher, well-documented acute cardiovascular risk profile than heat exposure does (see our contraindications guide, linked below), so a “Literature” label on a cold-side figure is read with an extra measure of caution built in, even when the citation itself is solid. We’re naming the rule here so it isn’t an invisible thumb on the scale.
| Modality | Parameter | Range | Tier | Anchor study |
|---|---|---|---|---|
| Cold plunge | Temperature (soreness) | 11–15°C | Literature | Machado et al. 2016 |
| Cold plunge | Duration (soreness) | 11–15 min | Literature | Machado et al. 2016 |
| Cold plunge | Novice starting duration | 1–3 min | Convention | Cold-water-safety guidance |
| Cold plunge | Frequency | 2–3× / week | Literature-derived | Søberg et al. 2021 |
| Cold shower | Daily duration | 30–90 sec | Literature | Buijze et al. 2016 |
| Sauna | Temperature (traditional) | ≈80–100°C | Convention | Finnish Sauna Society |
| Sauna | Session length (cohort avg.) | ≈14 min | Literature, observational | Laukkanen et al. 2015 |
| Sauna, infrared | Cabin temperature | ≈45–60°C | Manufacturer | Equipment spec, not a dose |
| Contrast | Heat → cold → rest | 15–20 / 2–4 / ≈10 min | Convention | Cross-brand protocol pages |
| Contrast | Rounds & sequence | 2–4×, ending cold | Convention | Cross-brand protocol pages |
Treat this table as the spine of the page — every section below unpacks one row, with the mechanism and the caveats a table can’t carry.
Two side-by-side charts. Left chart, cold plunge: horizontal axis water temperature from 5 to 25 degrees Celsius; vertical axis duration from 0 to 65 minutes. A shaded band, tagged Literature, spans 11 to 15 degrees Celsius by 11 to 15 minutes, from Machado et al. 2016 — the dose combination most consistently tied to reduced muscle soreness after exercise. A wider, lower, dashed band, tagged Convention, spans roughly 8 to 22 degrees Celsius by 1 to 3 minutes, marking the novice starting duration recommended by cold-water-safety guidance rather than a study. A single point at 14 degrees Celsius and 60 minutes, tagged Literature, marks Šrámek et al. 2000's one-hour research physiology exposure — not a recommended session length. A single point at 20 degrees Celsius and 5 minutes, tagged Literature, marks Yankouskaya et al. 2023's short immersion used to study acute mood effects. Right chart, sauna: horizontal axis cabin or water temperature from 40 to 110 degrees Celsius; vertical axis duration from 0 to 25 minutes. A dashed band, tagged Convention, spans 80 to 100 degrees Celsius across the full duration range, marking the traditional Finnish sauna temperature range. A separate dashed band, tagged Manufacturer, spans 45 to 60 degrees Celsius across the full duration range, marking the temperature ceiling of infrared cabins as an equipment specification, not a health dose. Two connected points at 79 degrees Celsius mark the Laukkanen et al. 2015 KIHD cohort: the average session length of about 14 minutes, and the roughly 19-minute threshold associated with lower sudden cardiac death risk in that same observational cohort — tagged Literature, observational.
Cold: how cold, how long
Temperature bands
The most consistently cited dose in the cold-plunge literature comes from Machado and colleagues’ 2016 systematic review and meta-analysis: cold-water immersion between 11–15°C for 11–15 minutes produced the best results for reducing post-exercise muscle soreness, though the underlying trials were rated low quality and the effect itself was described as modest — slightly better than passive recovery, not dramatically so Literature — Machado et al. 2016. A 2025 network meta-analysis corroborates the band, finding medium-duration, medium-temperature immersion (10–15 minutes at 11–15°C) most effective for soreness specifically, while colder water in the 5–10°C range performed better on other markers like creatine kinase and jump-height recovery Literature — Frontiers in Physiology, 2025.
Two other frequently cited temperatures come from studies with a different purpose entirely, and confusing them with a session recommendation is a real, common mistake. Šrámek and colleagues’ 2000 physiology study used a single one-hour head-out immersion at 14°C to measure hormonal and metabolic responses in ten healthy young men — a research exposure, not anyone’s suggested plunge length Literature — Šrámek et al. 2000. Yankouskaya and colleagues’ 2023 study used a much shorter five-minute immersion at 20°C to study acute mood and brain-connectivity effects in 33 participants Literature — Yankouskaya et al. 2023. Neither was designed to answer “how long should my plunge be,” and neither should be read as if it were.
Duration
Trial durations for soreness relief mostly sit in the 5–15 minute range Literature, with Machado’s 11–15 minute figure the single most defensible anchor inside that range. None of that is a starting point. Novices are conventionally directed toward roughly 1–3 minutes for a first cold exposure Convention — cold-water-safety guidance — well short of the studied soreness-relief range — and to build up gradually rather than starting at a study-length dose on day one. The full mechanics of why a slow start matters, and the cold-shock physiology behind it, live in our contraindications guide rather than here; see who should get medical clearance before plunging at all, and how to enter safely.
Position A — the studied sweet spot holds that 11–15°C for 11–15 minutes is the combination with the most direct trial support for soreness relief, so that’s the range worth working toward Literature — Machado et al. 2016. Position B — colder means shorter, a popular framing in podcast and brand-blog protocols, holds that a colder plunge needs less time to produce the same stimulus, so pushing temperature down and pulling duration in is a reasonable trade. What’s genuinely true of both: Machado’s own data shows a real dose-response relationship between temperature and time, so the intuition behind Position B isn’t wrong. What it isn’t is untested: the 2025 network meta-analysis cited above ranked colder 5–10°C doses against the 11–15°C band directly, and the colder water came second for soreness while beating that band on creatine kinase and jump-height recovery Literature — Frontiers in Physiology, 2025. Colder isn’t unstudied for soreness — it’s studied, and on that specific outcome it lost.
Frequency
Søberg and colleagues’ 2021 protocol — the study behind the “Søberg protocol” framing covered below — describes winter swimmers exposing themselves to cold water roughly 2–3 times per week Literature-derived — Søberg et al. 2021. The study itself compared seven winter-swimming men against eight non-swimming controls, all male, and the authors explicitly flag the small, single-sex sample and note it doesn’t support a causal claim about metabolic disease or weight — it’s evidence about cold-adaptation physiology, not a frequency prescription with outcome data behind it.
Separately, Buijze and colleagues’ 2016 randomized controlled trial — a much larger and more rigorous study — randomized 3,018 adults evenly into four groups: a 30-, 60-, or 90-second cold finish to a hot shower, or no cold finish at all, daily for 30 days. It found a 29% reduction in sickness-absence days in the cold-shower arms versus the control group, though no significant effect on the number of illness days themselves Literature — Buijze et al. 2016. That’s a cold-shower protocol, not a full-body plunge, and it measured a different outcome (work-absence, not soreness or mood) — worth naming as a genuinely different dataset rather than folding it into the same number as Søberg’s frequency figure.
The “11 minutes a week” claim, traced
This is the single most repeated cold-exposure number on the internet, and it deserves the full paper trail rather than a citation. Søberg’s own 2021 peer-reviewed study never states an 11-minute weekly total — it describes exposure frequency only, as “2–3 times per week,” in a small, all-male cohort Literature — Søberg et al. 2021. The specific “11 minutes” figure was popularized by neuroscientist Andrew Huberman on the Huberman Lab podcast, discussing that research with Søberg herself, and it was Huberman who named the framing the “Søberg Principle”:
“11 minutes total per week of uncomfortably cold, but safe cold, exposure.”Andrew Huberman, Huberman Lab podcast
This isn’t a case of a number drifting away from a source that disowns it. Søberg has since repeated and endorsed the 11-minute figure herself in interviews and her own book — alongside a companion 57-minutes-a-week heat figure — and has publicly credited Huberman with naming the “Søberg Principle” Literature-derived, provenance confirmed. So the chain is real and traceable: a peer-reviewed paper about exposure frequency, in a small cohort, became a podcast-named weekly-minutes target, which the paper’s own author later adopted in her public communication. What it did not become, at any point in that chain, is a weekly-minutes figure that was itself tested against an outcome in a controlled trial.
Treating “11 minutes a week” as a dose a trial has validated is the most common version of this mistake, and it’s an easy one to make given how confidently the figure circulates. Read it for what the paper trail actually supports: a plausible extrapolation from real cold-adaptation research, given a memorable name by a podcast host, later adopted by the researcher herself — not a target with its own outcome data behind it. For the separate question of whether cold exposure right after strength training helps or hurts muscle growth — a genuinely different, better-evidenced finding — see our full breakdown of cold plunge and muscle recovery.
Sauna: how hot, how long
Traditional Finnish sauna convention runs roughly 80–100°C, with most bathers settling closer to 80°C in practice Convention — Finnish Sauna Society. The largest dataset behind sauna-session norms comes from the KIHD cohort: Laukkanen and colleagues followed 2,315 Finnish men aged 42–60 for a median of 20.7 years, and their sessions averaged roughly 79°C for about 14 minutes Literature, observational — Laukkanen et al. 2015.
Two figures from that same cohort are easy to conflate and worth separating cleanly. Men who used the sauna 4–7 times a week had roughly 40% lower all-cause mortality than men who used it once a week — that’s a broad, all-cause comparison. A separate, narrower comparison in the same cohort looked specifically at sudden cardiac death: the hazard ratio was 0.78 for 2–3 sessions a week and 0.37 for 4–7 sessions a week, each versus once weekly, and sessions longer than 19 minutes carried a hazard ratio of 0.48 versus sessions under 11 minutes Literature, observational — Laukkanen et al. 2015. Both figures come from the same paper; they are not the same comparison, and neither is proof of cause — see the callout below.
Position A — the association is striking and worth taking seriously. A 20.7-year, 2,315-person cohort with a dose-response pattern (more sessions, longer sessions, lower hazard) is a genuinely large, well-conducted piece of observational evidence, and it’s the best sauna-and-mortality data this category has Literature, observational — Laukkanen et al. 2015. Position B — an association isn’t a controlled trial. The KIHD cohort is self-selected, drawn from a country with a deep cultural sauna tradition, and “healthy-user bias” — people healthy enough to sauna often being healthier to begin with — is a known, unresolved confounder in exactly this kind of data. Both are true at once: the finding is real and worth citing; it is not the same claim as “sauna causes lower mortality,” and this page won’t round the one into the other.
Infrared saunas run at meaningfully lower cabin temperatures — roughly 45–60°C — because the panels heat the body directly rather than heating the surrounding air the way a traditional heater does Manufacturer. Read that figure as exactly what it is: the temperature range the equipment operates at, not a dose anyone has studied for a specific outcome. A spec sheet is not a protocol, and this page won’t treat it as one.
The Finnish cohort literature is deep enough to deserve its own dedicated treatment beyond session length and temperature — a fuller look at what it does and doesn’t support is planned as a future companion piece to this hub, rather than compressed into a few paragraphs here.
Contrast: sequencing
Search “contrast therapy protocol” and you’ll find near-unanimous agreement on a sequence: heat for 15–20 minutes, cold for 2–4 minutes, rest around 10 minutes, repeat for 2–4 rounds, and end on cold. That agreement is real — it’s a genuine cross-brand convention Convention — but it is convention, not a sequence any trial tested end to end. What the underlying research actually looked at is narrower and less exciting than the protocol pages suggest.
| What’s commonly asserted | What Bieuzen et al. 2013 actually tested |
|---|---|
| A specific sequence: heat 15–20 min → cold 2–4 min → rest ≈10 min, repeated 2–4 times, always ending cold. | Contrast water therapy as a category — alternating hot and cold water immersion — compared against passive rest. Not this specific minute-by-minute sequence. |
| Ending on cold specifically primes daytime alertness; ending on warm aids sleep. | Sequencing and end-temperature effects on alertness or sleep were not outcomes the review measured. |
| Presented as a validated recovery protocol. | A modest reduction in exercise-induced muscle soreness versus passive rest; no superiority over other recovery methods; all 18 included trials flagged high risk of bias Literature — Bieuzen, Bleakley & Costello 2013. |
Two horizontal rows of four boxes each, connected by arrows, sharing the same column positions to invite direct comparison. Top row, labeled "what protocols commonly assert," dashed borders throughout: Heat, 15 to 20 minutes; arrow to Cold, 2 to 4 minutes; arrow to Rest, about 10 minutes; arrow to Repeat 2 to 4 times, ending on cold. Bottom row, labeled "what Bieuzen et al. 2013 actually tested," solid borders throughout: Contrast water therapy (CWT); arrow to "versus passive rest: soreness reduced, modest"; arrow to "no superiority versus other recovery methods"; arrow to "18 trials, high risk of bias." A legend clarifies that dashed boxes represent convention-level, commonly asserted claims, while solid boxes represent literature actually reported in Bieuzen, Bleakley and Costello's 2013 systematic review. The two rows occupy the same four column positions but contain different kinds of claims, visually underscoring that the popular protocol's structure does not match what was actually studied.
Position A — end on cold. The framing most associated with the “Søberg Principle” discussion holds that ending a contrast session cold primes alertness and energy for the rest of the day. Position B — end on warm. Other practitioner guidance holds the opposite for an evening session: ending warm supports wind-down and sleep. What’s true of both: neither is an RCT outcome — both are convention-level, physiologically plausible recommendations rather than tested findings Convention. If you plunge in the morning, ending cold is the more commonly recommended default; if you’re winding down for the night, ending warm is the more commonly recommended one. Either way, remember that core temperature keeps falling for a while after you exit cold water — “afterdrop” — so dry off and get into warm, dry clothing promptly regardless of which temperature you end on Convention — Tipton et al. 2017.
Before you start: the safety envelope
None of the ranges on this page are a substitute for the screening question. This is informational content, not medical advice, and every figure above should be read with that frame around it. If you have a known arrhythmia, severe aortic stenosis, a recent ICD shock, ischemic heart disease, are pregnant, take blood-pressure or thermoregulation medication, or have uncontrolled epilepsy, start with our sauna and cold plunge contraindications guide before you build a protocol from anything on this page — that article covers the screening question and the mechanism behind each caution; this one covers the dial-in once you’re cleared.
FAQ
How long should a cold plunge be?
For post-exercise soreness relief, the most consistently studied combination is 11–15°C for 11–15 minutes Literature — Machado et al. 2016. That’s not a starting point: novices are conventionally directed toward roughly 1–3 minutes for a first exposure Convention, building up gradually rather than jumping straight to the studied range. A frequently cited one-hour exposure at 14°C comes from a physiology research study, not a session recommendation Literature — Šrámek et al. 2000.
Is the “11 minutes a week” cold plunge target real?
It’s a real, traceable figure — but not a proven dose. Søberg’s own 2021 peer-reviewed study describes exposure frequency only, as 2–3 times a week, in a small, all-male cohort Literature — Søberg et al. 2021. The specific “11 minutes” weekly total was popularized by Andrew Huberman on the Huberman Lab podcast, who named it the “Søberg Principle,” and Søberg has since repeated the figure herself in public interviews and her book Literature-derived, provenance confirmed. Treat it as a plausible extrapolation from real research, not a weekly target that’s itself been tested against an outcome.
How hot should a sauna be?
Traditional Finnish convention runs roughly 80–100°C, with most bathers settling closer to 80°C Convention — Finnish Sauna Society. The largest cohort behind sauna-session norms, the KIHD study, averaged sessions of roughly 79°C for about 14 minutes across 2,315 Finnish men Literature, observational — Laukkanen et al. 2015. Infrared cabins run much cooler — roughly 45–60°C — because they heat the body directly rather than the air, and that figure is an equipment spec, not a health dose Manufacturer.
What’s the right contrast therapy (hot/cold) sequence?
There isn’t one tested end to end. The commonly asserted pattern — heat 15–20 minutes, cold 2–4 minutes, rest about 10 minutes, 2–4 rounds, ending cold — is a cross-brand convention, not a trial result Convention. The most-cited systematic review in this space, Bieuzen and colleagues’ 2013 analysis of contrast water therapy, found a modest reduction in exercise-induced soreness versus passive rest, no superiority over other recovery methods, and flagged all 18 included trials as high risk of bias Literature — Bieuzen, Bleakley & Costello 2013.