Information you can trust. All content on Veohtu is meticulously researched and written by Thomas Solomon, PhD. He does not sell supplements, recovery products, or ad space, and he has no sponsorships, brand affiliations, or ambassador roles. Everything you read reflects his independent views, shaped solely by peer-reviewed scientific evidence — and that will never change.
The Recovery Magic Tool from Thomas Solomon, PhD.
What are the best recovery methods for runners, OCR athletes, and endurance training?
Optimal recovery and performance is achieved with a well-planned and monitored training load combined with good nutrition, sleep, and rest. However, many athletes “do” their recovery using other “recovery” modalities, aka “magic”. This free tool is an up-to-date summary of all known scientific evidence for the most popular lines of recovery magic. I’ve designed this resource for scientists, practitioners, coaches, and athletes to help inform their choices for using additional recovery approaches that can supplement good nutrition, sleep, and rest. I will keep it up-to-date as new evidence emerges, and it can be used in combination with my Sports Supplements Tool.
Is this your first time using this tool? 
If yes, I strongly recommend reading the intro section below (900 words; 5 min read) because it contains important information about how you can learn to make informed decisions about choosing your approach to recovery. However, if you’ve already read the intro below, click the arrow to jump down to the tool.
“It definitely works!”
I've lost count of how many times I’ve heard social media influencers spew the “I've used this and it was awesome” narrative. Regrettably, such phrases are also spouted from the mouths of “reputable” athletes, coaches, and other practitioners including scientists, nutritionists, psychologists, physiologists, and medical doctors on podcast, radio, and TV interviews when talking about a new pill, potion, or device. Frustratingly, these folks rarely say what “it” works for, what “it” is being compared to, or whether using “it” made them objectively faster, stronger, or healthier. When you hear such narratives, think to yourself:
Were appropriate baseline measurements made?
What were the baseline measurements?
When were the baseline measurements followed up?
Did the baseline measurements actually change after the intervention period?
Did the person make any other lifestyle changes during the intervention period?
Is the person being paid or sponsored to promote the product? And, so on…
Because many endurance athletes indulge in a smorgasbord of “recovery” pills, potions, and devices, I want to bring clarity where there is obscurity in the often snake-oil-doused world of recovery magic. So, to help you understand whether “it” actually does improve recovery, I’ve dug into all known scientific evidence on this topic and created a free resource to help inform your decisions.
High-quality robust evidence comes from studies with a randomised controlled trial (RCT)The “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. design. But, “cherry-picking” a study to confirm a bias is not a valid pursuit for informing practice. A systematic reviewA systematic review answers a specific research question by systematically collating all known experimental evidence, which is collected according to pre-specified eligibility criteria. A systematic review helps inform decisions, guidelines, and policy. examines all the “cherries” in a standardised way and, when the studies included in a systematic review are of high enough quality, a meta-analysisA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all studies of that treatment. of all the available data can be completed. This calculates an overall effect sizeAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful that effect is. Common effect size measures include Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. along with a 95% confidence interval (CI)A measure of uncertainty used in Frequentist statistics. The 95% confidence interval is a plausible range of values within which the true value (e.g., the true treatment effect) would be found 95% of the time if the data was repeatedly collected in different samples of people. If this range of values (the confidence interval) crosses zero, there is little confidence in the effect. (the range of values the true effect size is likely to be found if the intervention is repeated) and a heterogeneity score or I-squared (I2) valueI-squared (I2) is a statistic used in meta-analysis to quantify the percentage of variation across studies that is due to heterogeneity rather than chance. A low I2 value (25-50%) indicates that the findings are relatively consistent between studies, whereas a high I2 value (>75%) suggests considerable variability among the study results, which may affect the reliability of the overall conclusions. (how variable the effect is). In simple words, a meta-analysis analyses all the “cherries” simultaneously to produce a useable effect size based on all available scientific evidence, enabling good decisions to be made.
So, when I say that “I’ve dug into all known scientific evidence”, I mean that I’ve read all known meta-analyses on PubMed and summarised the evidence in this free resource: the Recovery Magic Tool. I aim to keep each topic up-to-date when scientific advances are made. This will help inform your decisions when choosing from the multitude of recovery modalities available. But, before diving in, always remember that “feeling” ready to go is different from actually being ready to unleash your maximal potential. And, before making any decisions, always conduct a cost-benefit analysis, where “cost” includes a combination of financial costs, time costs, moral costs, risk of contamination, potential performance impairment, and harm to health. For example:
If there is no benefit, there is no point in using magic.
If there is a benefit and no (or little) cost, use it; you’d be foolish not to.
If the cost outweighs the benefit, do not proceed.
OK… You’re now ready for some science.
The essential and non-negotiable tools in your toolbox:
Plus, all the other magic:
What is sleep?
Sleep isn’t magic; it’s biology doing essential maintenance. It’s mandatory for health and a core tool in your recovery kit. If you want the full deep dive, I’ve written about it here:
→ “Sleep: a five-letter word to supercharge your recovery.”
→ “Sleep will supercharge your recovery. But what about napping and your chronotype?”
What is the scientific evidence on the effect of sleep on recovery and athletic performance?
Getting sufficient daily sleep is an essential and non-negotiable part of recovery. For an in-depth overview of the evidence, please read my articles on sleep and napping. The short version:
Cutting sleep reduces recovery from training, slows adaptation, and knocks performance.
Extending sleep tends to boosts several facets of recovery and performance.
You can “top up” nightly sleep with a daytime nap. Naps make many people feel refreshed, but — as after a full night of sleep — you might feel a short spell of grogginess (sleep inertiaSleep inertia is that groggy, heavy-headed feeling you can get right after waking—usually lasting minutes to an hour—while your brain and body fully “boot up.”) when you wake.
When you already sleep well at night, a 30 to 60 minute afternoon nap can reduce post-exercise fatigue and improve cognitive and physical performance, especially when testing happens at least 1 hour after waking from the nap (see Mesas et al. 2023).
Napping might also help after sleep loss (athletes included), but the current evidence base is small and of low to moderate quality (see Lastella et al. 2021, Souabni et al. 2021, Mesas et al. 2023 & Boukhris et al. 2023). It’s still unclear whether naps can fully restore performance after sleep deprivation.
Because the direct effects of napping on recovery are not fully nailed down, your best bet is to optimise sleep hygiene and protect your nightly sleep. See“Sleep: a five-letter word to supercharge your recovery” for practical tips.
Can sleep enhance athletic recovery and performance?
Getting more sleep is likely to speed up recovery and help restore performance after training, whereas losing sleep is likely to hurt it.
The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. looks small to medium.
When your nightly sleep is already solid, a 30 to 60 minute afternoon nap may further reduce fatigue and may improve performance, with moderate to large effects in some tests..
Whether naps can fully fix the damage from sleep deprivation remains uncertain. We need more high-quality randomized controlled trials.
Effects seem similar in trained athletes and untrained folks. Sex differences are unclear because most studies include more males than females.
Keep in mind: designs and effects between studies have moderate to high variability (heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate.), many studies have a moderate to high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported. (often due to a lack of blindingBlinding is when people in a study don’t know which treatment they’re getting. It stops expectations or beliefs (from patients or researchers) from skewing the results. “Single-blind” means participants don’t know; “double-blind” means participants and researchers don’t know; “triple-blind” means that the participants, researchers, and data analysts are kept in the dark. The goal is simple: fair tests and trustworthy findings.), and there is possible publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is moderateA moderate quality of evidence means that, in general, studies in this field have some limitations. This could be due to somewhat inconsistent effects between studies, a moderate range of effect sizes between studies, and/or a moderate risk of bias (caused by a small to medium number of studies, small to medium numbers of participants, poorly described randomization processes, some missing data, some inappropriate methods/statistics). When the quality of evidence is moderate, there is some doubt and only moderate confidence in the overall effect of an intervention, and future studies could change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials. at best. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: getting more sleep doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, get your zzz’s on.
How to use this: Aim for a consistent 7–9 hours of sleep per night. If nights are solid, trial a 30–60 minute nap 1–3× per week and schedule it so you’ve got at least 60 minutes before any hard sessions/races.
Full list of meta-analyses examining sleep for recovery.
Here are the meta-analyses I've summarised above:
Effects of Acute Sleep Deprivation on Sporting Performance in Athletes: A Comprehensive Systematic Review and Meta-Analysis Mingjun Gong, Min Sun, Yaqi Sun, Lijuan Jin, Shen Li. Nat Sci Sleep. 2024
The Impact of Sleep Interventions on Athletic Performance: A Systematic Review. Lúcio A. Cunha, Júlio A. Costa, Elisa A. Marques, João Brito, Michele Lastella, Pedro Figueiredo. Sports Medicine Open. 2023
Dawn of a New Dawn: Advances in Sleep Health to Optimize Performance. Alice D. LaGoy, Andrew G. Kubala, Sean Deering, Anne Germain, Rachel R. Markwald. . 2023
The Impact of Daytime Napping Following Normal Night-Time Sleep on Physical Performance: A Systematic Review, Meta-analysis and Meta-regression. Omar Boukhris, Khaled Trabelsi, Haresh Suppiah, Achraf Ammar, Cain C. T. Clark, Haitham Jahrami, Hamdi Chtourou, Matthew Driller. Sports Med. 2023
Is daytime napping an effective strategy to improve sport-related cognitive and physical performance and reduce fatigue? A systematic review and meta-analysis of randomised controlled trials. Arthur Eumann Mesas, Sergio Núñez de Arenas-Arroyo, Vicente Martinez-Vizcaino, Miriam Garrido-Miguel, Ruben Fernández-Rodríguez, Bruno Bizzozero-Peroni, Ana I Torres-Costoso. Br J Sports Med. 2023
A systematic review of effects of daytime napping strategies on sports performance in physically active individuals with and without partial-sleep deprivation. Priya Sirohi, Moazzam Hussain Khan, Saurabh Sharma, Shibili Nuhmani, Wafa Hashem Al Muslem, Turki Abualait. PeerJ. 2022
How much does sleep deprivation impair endurance performance? A systematic review and meta-analysis. Thiago Ribeiro Lopes, Hugo Maxwell Pereira, Lia Rita Azeredo Bittencourt, & Bruno Moreira Silva. Eur J Sports Sci. 2022
The Impact of Dietary Factors on the Sleep of Athletically Trained Populations: A Systematic Review. Jackson Barnard, Spencer Roberts, Michele Lastella, Brad Aisbett, and Dominique Condo. Nutrients. 2022
The influence of blue light on sleep, performance and wellbeing in young adults: A systematic review. Marcia Ines Silvani, Robert Werder, and Claudio Perret. Front Physiol. 2022
How Sleep Affects Recovery and Performance in Basketball: A Systematic Review. Javier Ochoa-Lácar, Meeta Singh, Stephen P. Bird, Jonathan Charest, Thomas Huyghe, and Julio Calleja-González. Brain Sci. 2022
Effects of Acute Sleep Loss on Physical Performance: A Systematic and Meta-Analytical Review. Jonathan Craven, Danielle McCartney, Ben Desbrow, Surendran Sabapathy, Phillip Bellinger, Llion Roberts, Christopher Irwin. Sports Med. 2022
Sleep interventions for performance, mood and sleep outcomes in athletes: A systematic review and meta-analysis. Kate Gwyther, Simon M Rice, Rosemary Purcell, Courtney C Walton. Psychology of Sport and Exercise. 2021
The effects of evening high-intensity exercise on sleep in healthy adults: A systematic review and meta-analysis. Emmanuel Frimpong, Melodee Mograss, Tehila Zvionow, Thien Thanh Dang-Vu. Sleep Med Rev. 2021
Effects of High-Intensity Interval Training on Sleep: A Systematic Review and Meta-Analysis. Leizi Min, Dizhi Wang, Yanwei You, Yingyao Fu, Xindong Ma. Int J Environ Res Public Health. 2021
To Nap or Not to Nap? A Systematic Review Evaluating Napping Behavior in Athletes and the Impact on Various Measures of Athletic Performance. Lastella M, Halson SL, Vitale JA, Memon AR, Vincent GE. Nat Sci Sleep. 2021
Benefits of Daytime Napping Opportunity on Physical and Cognitive Performances in Physically Active Participants: A Systematic Review. Souabni M, Hammouda O, Romdhani M, Trabelsi K, Ammar A, Driss T. Sports Med. 2021
Sleep disruption in medicine students and its relationship with impaired academic performance: A systematic review and meta-analysis. Hernan A Seoane, Leandra Moschetto, Francisco Orliacq, Josefina Orliac, Ezequiel Serrano, María Inés Cazenave, Daniel E Vigo, Santiago Perez-Lloret. Sleep Med Rev. 2020
Sleep Interventions Designed to Improve Athletic Performance and Recovery: A Systematic Review of Current Approaches. Bonnar D, Bartel K, Kakoschke N, Lang C. Sports Med. 2018
Inadequate sleep and muscle strength: Implications for resistance training. Olivia E Knowles, Eric J Drinkwater, Charles S Urwin, Séverine Lamon, Brad Aisbett. J Sci Med Sport. 2018
What is nutrition?
Nutrition is food. Not magic. Real food. It’s non-negotiable — without it, you will literally grind to a halt. For runners, trail folks, ultra humans, and OCR (obstacle course racing) athletes, food is a core tool in the recovery toolbox.
I’ve written deep-dive reviews that walk through what to eat and when, with summaries of current systematic reviews and meta-analyses:
→ “Recovery nutrition starts with a healthy eating pattern.”
→ “The “post-exercise nutrition window” is your between-session period for nutrient optimisation.”
→ “Eating carbohydrate replenishes muscle glycogen.”
→ “Eating protein supports muscle protein synthesis to repair and build tissue.”
→ “Carbohydrate plus protein is your recovery superpower.”
I also cover training nutrition — fat oxidation, “fueling for the work required”, and the long saga of low-carb vs. high-carb — here:
→ “What are fat oxidation rates and why do they matter?”
→ “Increasing fat burning during exercise — acute manipulations of carbohydrate availability.”
→ “Does a low-carb, high-fat diet offer a performance advantage to an endurance athlete?”
→ “Understanding the “carb wars” using context and perspective.”
What is the scientific evidence on the effect of nutrition on recovery athletic performance?
Maintaining a healthy eating pattern and enough energy intake supports health, recovery, and consistency.
Timing and dose matter in the space between sessions: carbohydrates help replenish muscle glycogen and protein supports muscle repair and rebuilding.
Combining carbohydrate with protein between your sessions can add a small extra boost for recovery.
But, there’s no need to stress. Eating a wide variety of colorful foods across and within all food groups, spreading meals through the day, and including carbs, fats, and protein at each meal or snack will keep most people on track.
Plus, the “urgency” to refuel depends on the time since you last ate and the time until your next workout. If your next workout is days away, rapid post-exercise carbs are not urgent — regular meals will restore glycogen. If your pre-workout meal had plenty of protein, immediate protein after is also less urgent.
Can nutrition enhance athletic recovery and performance?
Nutrition is very likely to reduce muscle soreness after exercise and improve the restoration of performance.
Low-carb approaches can impair running economy and performance, while higher carbohydrate intake supports recovery between sessions and regular carb intake during exercise can improve performance. For muscle-damaging work, protein supplements slightly improve the recovery of performance, and higher daily protein supports gains in strength and muscle size — with benefits flattening around 1.6 grams per kilogram per day.
Effect sizesA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. range from small to large, depending on the outcome measured.
For carbs and protein, effects look similar in males and females, but females are under-represented in studies, so more research is needed.
Carb effects appear similar in trained athletes and untrained folks. Protein effects may be larger in trained athletes — probably because more training makes protein more relevant to performance and recovery.
Keep in mind: there is moderate to high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, and some risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is moderateA moderate quality of evidence means that, in general, studies in this field have some limitations. This could be due to somewhat inconsistent effects between studies, a moderate range of effect sizes between studies, and/or a moderate risk of bias (caused by a small to medium number of studies, small to medium numbers of participants, poorly described randomization processes, some missing data, some inappropriate methods/statistics). When the quality of evidence is moderate, there is some doubt and only moderate confidence in the overall effect of an intervention, and future studies could change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
How to use this: Aim to eat enough overall energy each day; build meals around carbohydrate-rich staples plus a protein source; include fats you enjoy; and adjust urgency to eat again based on the time until your next session. If your next hard session is soon, prioritize a carb-focused meal or snack after training and include a decent protein source; if the next session is far away, just return to your regular meal pattern. For daily protein, target a total intake near 1.6 grams per kilogram of body weight per day across several meals — more is not always better.
Full list of meta-analyses examining nutrition for recovery.
Here are the meta-analyses I've summarised above:
The effect of protein intake on athletic performance: a systematic review and meta-analysis. Zhao et al. Front Nutr. (2024)
The ergogenic effects of acute carbohydrate feeding on endurance performance: a systematic review, meta-analysis and meta-regression. Ramos-Campo DJ, Clemente-Suárez VJ, Cupeiro R, Benítez-Muñoz JA, Andreu Caravaca L, Rubio-Arias JÁ. Crit Rev Food Sci Nutr. 2023
Effect of Soy Protein Supplementation on Muscle Adaptations, Metabolic and Antioxidant Status, Hormonal Response, and Exercise Performance of Active Individuals and Athletes: A Systematic Review of Randomised Controlled Trials. Zare R, Devrim-Lanpir A, Guazzotti S, Ali Redha A, Prokopidis K, Spadaccini D, Cannataro R, Cione E, Henselmans M, Aragon AA. . 2023
Effects of Timing and Types of Protein Supplementation on Improving Muscle Mass, Strength, and Physical Performance in Adults Undergoing Resistance Training: A Network Meta-Analysis. Zhou HH, Liao Y, Zhou X, Peng Z, Xu S, Shi S, Liu L, Hao L, Yang W. Int J Sport Nutr Exerc Metab. 2023
The impact of dietary protein supplementation on recovery from resistance exercise-induced muscle damage: A systematic review with meta-analysis. Alice G. Pearson, Karen Hind & Lindsay S. Macnaughton. Eur J Clin Nutr. 2022
The Ergogenic Effects of Acute Carbohydrate Feeding on Resistance Exercise Performance: A Systematic Review and Meta-analysis. Andrew King, Eric Helms, Caryn Zinn, Ivan Jukic. Sports Med. 2022
Systematic review and meta-analysis of protein intake to support muscle mass and function in healthy adults. Everson A Nunes, Lauren Colenso-Semple, Sean R McKellar, Thomas Yau, Muhammad Usman Ali, Donna Fitzpatrick-Lewis, Diana Sherifali, Claire Gaudichon, Daniel Tomé, Philip J Atherton, Maria Camprubi Robles, Sandra Naranjo-Modad, Michelle Braun, Francesco Landi, Stuart M Phillips. J Cachexia Sarcopenia Muscle. 2022
Muscle Protein Synthesis Responses Following Aerobic-Based Exercise or High-Intensity Interval Training with or Without Protein Ingestion: A Systematic Review.. Reza Bagheri, Isabelle Robinson, Sajjad Moradi, Jessica Purcell, Elita Schwab, Tharindie Silva, Brooke Baker & Donny M. Camera. Sports Medicine. 2022
Effects of the ketogenic diet on performance and body composition in athletes and trained adults: a systematic review and Bayesian multivariate multilevel meta-analysis and meta-regression. Ana Clara C Koerich, Fernando Klitzke Borszcz, Arthur Thives Mello, Ricardo Dantas de Lucas, Fernanda Hansen. Crit Rev Food Sci Nutr. 2022
The Effect of a Ketogenic Low-Carbohydrate, High-Fat Diet on Aerobic Capacity and Exercise Performance in Endurance Athletes: A Systematic Review and Meta-Analysis.
Cao J, Lei S, Wang X, Cheng S.
Nutrients. 2021
A systematic review and meta-analysis: Effects of protein hydrolysate supplementation on fat-free mass and strength in resistance-trained individuals.
Shen M, Zhang W, Wu G, Zhu L, Qi X, Zhang H.
Crit Rev Food Sci Nutr. 2021
Pre-Sleep Casein Supplementation, Metabolism, and Appetite: A Systematic Review.
Dela Cruz J, and Kahan D.
Nutrients. 2021
Performance effects of periodized carbohydrate restriction in endurance trained athletes – a systematic review and meta-analysis.
Gejl KD, Nybo L.
J Int Soc Sports Nutr. 2021
Does Protein Supplementation Support Adaptations to Arduous Concurrent Exercise Training? A Systematic Review and Meta-Analysis with Military Based Applications.
Chapman S, Chung HC, Rawcliffe AJ, Izard R, Smith L, Roberts JD.
Nutrients. 2021
Protein supplementation increases adaptations to endurance training: A systematic review and meta-analysis.
Lin YN, Tseng TT, Knuiman P, Chan WP, Wu SH, Tsai CL, Hsu CY.
Clin Nutr. 2020
Coingestion of Carbohydrate and Protein on Muscle Glycogen Synthesis after Exercise: A Meta-analysis
Margolis LM, Allen JT, Hatch-McChesney A, Pasiakos SM.
Med Sci Sports Exerc. 2020
Effects of pre-sleep protein consumption on muscle-related outcomes - A systematic review.
Reis CEG, Loureiro LMR, Roschel H, da Costa THM.
J Sci Med Sport. 2020
A review of the ketogenic diet for endurance athletes: performance enhancer or placebo effect?
Bailey CP, Hennessy E.
J Int Soc Sports Nutr . 2020
Fruit supplementation reduces indices of exercise-induced muscle damage: a systematic review and meta-analysis.
Doma K, Gahreman D, Connor J. Eur J Sport Sci. 2020
The Effect of Ingesting Carbohydrate and Proteins on Athletic Performance: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.
Kloby Nielsen LL, Tandrup Lambert MN, Jeppesen PB.
Nutrients. 2020
The Role of Muscle Mass Gain Following Protein Supplementation Plus Exercise Therapy in Older Adults with Sarcopenia and Frailty Risks: A Systematic Review and Meta-Regression Analysis of Randomized Trials.
Liao CD, Chen HC, Huang SW, Liou TH.
Nutrients. 2019
Impact of cow's milk intake on exercise performance and recovery of muscle function: a systematic review.
Alcantara JMA, Sanchez-Delgado G, Martinez-Tellez B, Labayen I, Ruiz JR.
J Int Soc Sports Nutr. 2019
Chocolate milk for recovery from exercise: a systematic review and meta-analysis of controlled clinical trials.
Amiri M, Ghiasvand R, Kaviani M, Forbes SC, Salehi-Abargouei A. Eur J Clin Nutr. 2019
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis. Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
The Effect of Whey Protein Supplementation on the Temporal Recovery of Muscle Function Following Resistance Training: A Systematic Review and Meta-Analysis. Davies RW, Carson BP, Jakeman PM.
Nutrients. 2018
Post-exercise Ingestion of Carbohydrate, Protein and Water: A Systematic Review and Meta-analysis for Effects on Subsequent Athletic Performance.
McCartney D, Desbrow B, Irwin C.
Sports Med. 2018
Supplementation Strategies to Reduce Muscle Damage and Improve Recovery Following Exercise in Females: A Systematic Review.
Köhne JL, Ormsbee MJ, McKune AJ.
Sports (Basel). 2016
Effects of protein supplements on muscle damage, soreness and recovery of muscle function and physical performance: a systematic review.
Pasiakos SM, Lieberman HR, McLellan TM.
Sports Med. 2014
Effects of protein in combination with carbohydrate supplements on acute or repeat endurance exercise performance: a systematic review.
McLellan TM, Pasiakos SM, Lieberman HR.
Sports Med. 2014
Protein supplementation augments the adaptive response of skeletal muscle to resistance-type exercise training: a meta-analysis.
Cermak NM, Res PT, de Groot LC, Saris WH, van Loon LJ.
Am J Clin Nutr. 2012
Effects of ingesting protein in combination with carbohydrate during exercise on endurance performance: a systematic review with meta-analysis.
Stearns RL, Emmanuel H, Volek JS, Casa DJ.
J Strength Cond Res. 2010
What is rest?
Rest is the space between the work. It’s the minutes between intervals, the hours you sleep between days, the easy days between hard ones, and the lighter weeks inside a training block. That space is where your body resets its internal balance (homeostasis), repairs, and adapts so you can show up ready for the next bout. No mystery, no magic — just physiology taking its time..
Because rest is non-negotiable, smart training balances short-term stress with longer-term load — learn all about that at veohtu.com/trainingload and veohtu.com/trimp. Planning, doing, and reviewing your training keeps that balance in check, don’t forget your headspace. Mental load from decision fatigue, life stress, or endlessly second-guessing your choices can also sap performance — read all about that at veohtu.com/centralfatigue.
What is the scientific evidence on the effect of rest on recovery and athletic performance?
Tapering — reducing training load in the final days before a race — can improve performance. Multiple lines of evidence back this up (see Ramirez-Campillo et al. 2021, Vachon et al. 2020, Grivas et al. 2018, & Bosquet et al. 2007).
Keeping a healthy balance between recent (acute) load and longer-term (chronic) load often means adding more easy time between hard days — and this balance is linked with better performance (see Coyne et al. 2022, Inoue et al. 2022, Lima-Alves et al. 2022, Fox et al. 2018, McLaren et al. 2017, Booth et al. 2017, & Jaspers et al. 2016).
Mental fatigue is a state of feeling tired or flat after long periods of effort or heavy thinking. It can come from anything that burdens your mind, such as training decisions, food choices, recovery planning — the whole circus.
Evidence shows that mentally demanding tasks can hurt cognitive skills and sport-specific skills, raise how hard exercise feels (RPE), and nudge down strength and endurance (see Martin et al. 2019, Habay et al. 2021, Sun et al. 2021, Clemente et al. 2021, & Cao et al. 2022, Alix-Fages et al. 2022, Brown et al. 2019, & McMorris et al. 2018). However, most studies in this field are small and further high-quality randomised controlled trials are needed to be more certain of the effects (Holgado et al. 2023).
Can rest enhance athletic recovery and performance?
Physical rest — temporarily reducing your training load, which before a race is called tapering — is likely to improve performance and speed up the recovery of muscle soreness following exercise. .
Managing the balance between recent (acute) and longer-term (chronic) training load — which often means more rest between hard days — can also improve performance.
On the flipside, mental fatigue likely harms performance. Therefore, resting both your body and your mind is part of smart recovery and the path to optimal performance.
These effect sizesA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. are small.
And, the effect of rest and training load management appear to be similar between males and females; however, male participants are predominant and females are vastly underrepresented among studies in this field.
Plus, the effect of mental fatigue appears to be similar between trained athletes and untrained folks, but this comparison has not been studied in detail and further work is needed.
Keep in mind: there is a very small number of studies and high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies. Therefore, there is a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported. and the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
How to use this: Purposefully plan your rest. Keep easy time between hard sessions; protect your sleep before key workouts and races; and in race week, reduce total training load (taper) to arrive fresh. Remove avoidable mental clutter — make simple plans, decide once, and give your brain some peace and quiet in the hours before you want to unleash the beast. None of this needs to be fancy; it just needs to be done.
Full list of meta-analyses examining rest for recovery.
Here are the meta-analyses I've summarised above:
The counteractive effects of interventions addressing mental fatigue on sport-specific performance among athletes: A systematic review with a meta-analysis. Sun et al. J Sports Sci. (2024)
Assessing the Evidential Value of Mental Fatigue and Exercise Research. Holgado D, Mesquida C, Román-Caballero R. Sports Med. 2023
The Current State of Subjective Training Load Monitoring: Follow-Up and Future Directions. Joseph O. C. Coyne, Aaron J. Coutts, Robert U. Newton & G. Gregory Haff. Sports Med Open. 2022
Internal Training Load Perceived by Athletes and Planned by Coaches: A Systematic Review and Meta-Analysis. Allan Inoue, Priscila dos Santos Bunn, Everton Crivoi do Carmo, Eduardo Lattari & Elirez Bezerra da Silva. Sports Med Open. 2022
The relationship between internal and external loads as a tool to monitor physical fitness status of team sport athletes: a systematic review. Adriano Lima-Alves, João Gustavo Claudino, Daniel Boullosa, Crislaine Rangel Couto, Francisco Teixeira-Coelho, and Eduardo M. Pimenta. Biol Sport. 2022
Mental Fatigue and Basketball Performance: A Systematic Review. Shudian Cao, Soh Kim Geok, Samsilah Roslan, He Sun, Soh Kim Lam and Shaowen Qian. Front Psychol. 2022
Effects of Mental Fatigue on Strength Endurance: A Systematic Review and Meta-Analysis. Carlos Alix-Fages, Jozo Grgic, Pablo Jiménez-Martínez, Eneko Baz-Valle, Carlos Balsalobre-Fernández. Motor Control. 2022
Does mental fatigue affect skilled performance in athletes? A systematic review. He Sun, Kim Geok Soh, Samsilah Roslan, Mohd Rozilee Wazir Norjali Wazir, Kim Lam Soh. PLoS One. 2021
Mental Fatigue and Sport‑Specifc Psychomotor Performance: A Systematic Review. Jelle Habay, Jeroen Van Cutsem, Jo Verschueren, Sander De Bock, Matthias Proost, Jonas De Wachter, Bruno Tassignon, Romain Meeusen, Bart Roelands. Sports Med. 2021
Effects of Mental Fatigue in Total Running Distance and Tactical Behavior During Small-Sided Games: A Systematic Review With a Meta-Analysis in Youth and Young Adult's Soccer Players. Filipe Manuel Clemente, Rodrigo Ramirez-Campillo, Daniel Castillo, Javier Raya-González, Ana Filipa Silva, José Afonso, Hugo Sarmento, Thomas Rosemann, Beat Knechtle. Front. Psychol. 2021
Tapering strategies applied to plyometric jump training: a systematic review with meta-analysis of randomized-controlled trials. Rodrigo Ramirez-Campillo, Lucas A Pereira, David C Andrade, Guillermo Mendez-Rebolledo, Carlos I De La Fuente, Mauricio Castro-Sepulveda, Felipe Garcia-Pinillos, Tomás T Freitas, Irineu Loturco. Journal. 2021
Effects of tapering on neuromuscular and metabolic fitness in team sports: a systematic review and meta-analysis. Adrien Vachon, Nicolas Berryman, Iñigo Mujika, Jean-Baptiste Paquet, Denis Arvisais & Laurent Bosquet. European Journal of Sport Science. 2020
Mental Fatigue Might Be Not So Bad for Exercise Performance After All: A Systematic Review and Bias-Sensitive Meta-Analysis. Darías Holgado, Daniel Sanabria, José C. Perales, and Miguel A. Vadillo. J Cogn. 2020 (NOTE: this article has a correction.)
Physiological Factors Which Influence Cognitive Performance in Military Personnel. Kristy Martin, Julien Périard and David B. Pyne. Journal of the Human Factors and Ergonomics Society. 2019
Effects of Prior Cognitive Exertion on Physical Performance: A Systematic Review and Meta-analysis. Brown DMY, Graham JD, Innes KI, Harris S, Flemington A, Bray SR. Sports Med. 2020
The Association Between Training Load and Performance in Team Sports: A Systematic Review. Jordan L. Fox, Robert Stanton, Charli Sargent, Sally‑Anne Wintour, Aaron T. Scanlan. Sports Med. 2018
The Effects of Tapering on Performance in Elite Endurance Runners: A Systematic Review. Gerasimos V. Grivas. International Journal of Sports Science. 2018
Cognitive fatigue effects on physical performance: A systematic review and meta-analysis. McMorris T, Barwood M, Hale BJ, Dicks M, Corbett J. Physiol Behav. 2018
The Relationships Between Internal and External Measures of Training Load and Intensity in Team Sports: A Meta-Analysis. Shaun J. McLaren, Tom W. Macpherson, Aaron J. Coutts, Christopher Hurst, Iain R. Spears, Matthew Weston. Sports Med. 2017
The effect of training loads on performance measures and injury characteristics in rugby league players. A systematic review. Mark Booth, Rhonda Orr, Stephen Cobley. International Journal of Sports Physiology and Performance. 2017
Relationships Between Training Load Indicators and Training Outcomes in Professional Soccer. Arne Jaspers, Michel S. Brink, Steven G. M. Probst, Wouter G. P. Frencken, Werner F. Helsen. Sports Med. 2016
Effects of Tapering on Performance. A Meta-Analysis. Laurent Bosquet, Jonathan Montpetit, Denis Arvisais, Iñigo Mujika. Med Sci Sports Exerc. 2007
What is active recovery?
The paradox of “active recovery” is somewhat confusing since recovery — that is the restoration of homeostasis and performance — takes time and rest. Easy effort sessions in between your carefully planned days of homeostatic mayhem help you recover in a sense by allowing adequate time to pass between hard days. However, easy effort sessions in themselves are sessions that disrupt homeostasis and require time to restore things back to normal.
The world is full of gadgets and pills and potions that claim recovery benefits by “lowering lactate”, “increasing blood flow”, “clearing toxic metabolites”, “suppressing muscle soreness”, and “boosting feelings of recovery”. But it’s best not to neglect the most powerful stimulus of all — movement. Going for a gentle stroll after a session is a form of active recovery. Sitting down and resting is a form of passive recovery.
A 2018 narrative review from Bas Van Hooren and Jonathan Peake suggested that while many athletes perform an active “cool-down” including 5- to 15-min of low- to moderate-intensity exercises within an hour of completing a session or race, it may only be effective for restoring performance when the next bout is up to 20-min later. For subsequent bouts that are longer than 4 hours later, an active cool-down is likely not effective for boosting sports performance later that day or in the following days. However, a narrative review, albeit credible, does not use a standardised systematic approach to evaluating the literature. Furthermore, Van Hooren and Peake aimed to discuss the effects of a cool-down on later-in-the-day performance, not between-session active recovery. So, what do the systematic approaches say?
What is the scientific evidence on the effect of active recovery (cool down) on recovery and athletic performance?
Note that some researchers have referred to modalities like massage, foam rolling, cold exposure, etc as “active” recovery. For this post, “active recovery” refers to moving one’s ass with light activities like walking, jogging, cycling, etc.
Between sessions, active recovery (light activity) can have a large effect on reducing post-exercise feelings of delayed onset muscle soreness (DOMS), especially in the short-term (for up to 6 to 24 hours) but is less effective at lowering the perceived feelings of fatigue (see Dupuy et al. 2018).
However, the optimal intensity and duration for active recovery sessions on days between heavier sessions is unknown at this time. From a running perspective, most endurance coaches agree that low-to-moderate intensity exercise is favourable on days in between harder work.
Within running sprint interval sessions, active recovery (moving) during rest intervals tends to blunt the restoration of running sprint performance and elevate feelings of perceived effort when compared to passive recovery, i.e. sitting or lying down (see Perrier-Melo et al. 2021 & Madueno et al. 2019). However, a systematic reviewA systematic review answers a specific research question by systematically collating all known experimental evidence, which is collected according to pre-specified eligibility criteria. A systematic review helps inform decisions, guidelines, and policy. by Zouhal et al. 2024 found that either active or passive recovery regimes during interval training can improve fitness in trained and untrained adults suggesting that the type of within-session recovery might not influence training outcomes.
Within days of repeated, prolonged high-intensity efforts, including running, cycling, and soccer, an active recovery of between 6 to 10 minutes seems to more favourably restore high-intensity performance for a repeated effort within hours, when compared to passive rest or active rest of less than 6 or greater than 10 minutes (see Ortiz et al. 2019). This might be relevant when competing in multiple competitions or races on a single day. However, due to variability between studies, more high-quality studies are needed before a firm conclusion can be made.
One caveat to this field it that among the studies selected in the various systematic reviews, the definition of “active recovery” is vague (see Ortiz et al. 2019 for an overview). Sometimes studies refer to walking, jogging, cycling, or swimming at a low to moderate intensity as “active recovery”, while some studies refer to the use of devices or modalities that are designed to facilitate recovery as “active recovery”. Consequently, the quality of evidence in this field is poor.
Since we have known for many moons that the best way to increase muscle contraction, blood flow, and cardiac output is to move (see Joyner and Casey 2015), contracting your muscles with gentle walking/cycling/swimming etc or even some light jogging will increase your heart rate and massively increase blood flow to your muscles, elevating venous return — no device can mimic the magnitude of the exercise-induced change in these physiological variables.
A feeling that you are not recovering and/or a feeling that your performance level is down is sometimes induced by a lack of rest in between sessions. Being active as a form of “recovery” is not always the best course of action and indeed the phrase “active recovery” is somewhat paradoxical. If you feel that you need rest — passive rest (no exercise) — then don’t move, just sit down and do something relaxing.
Can active recovery enhance athletic recovery and performance?
Active recovery (moving) is likely to improve the restoration of performance and the recovery of muscle soreness following high-intensity exercise. However, passive recovery (sitting or lying down) between intervals is likely to be preferable for maximising power output during a sprint interval session. That said, the type of within-session recovery (active vs. passive) does not seem to influence gains in fitness during training interventions.
The effect sizesA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. are unclear due to an insufficient number of studies and no meta-analysisA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
Because of a lack of research, it is unclear how active recovery compares between trained athletes and untrained folks, and between males and females.
Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and a small number of studies. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. remains lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials. until additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are published and a meta-analysisA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment. can be completed.
How to use this: INSERT.
Full list of meta-analyses examining active recovery.
Here are the meta-analyses I've summarised above:
Effects of Passive or Active Recovery Regimes Applied During Long-Term Interval Training on Physical Fitness in Healthy Trained and Untrained Individuals: A Systematic Review. Zouhal et al. Sports Med Open. (2024)
Effect of active versus passive recovery on performance-related outcome during high-intensity interval exercise. Perrier-Melo RJ, D'Amorim I, Meireles Santos T, Caldas Costa E, Rodrigues Barbosa R, DA Cunha Costa M. J Sports Med Phys Fitness. 2021
The Use of Acute Exercise Interventions as Game Day Priming Strategies to Improve Physical Performance and Athlete Readiness in Team-Sport Athletes: A Systematic Review. Mason B, McKune A, Pumpa K,Ball N. Sports Med. 2020
A Systematic Review on the Effectiveness of Active Recovery Interventions on Athletic Performance of Professional-, Collegiate-, and Competitive-Level Adult Athletes. Ortiz RO Jr, Sinclair Elder AJ, Elder CL, Dawes JJ. J Strength Cond Res. 2019
A systematic review examining the physiological, perceptual, and performance effects of active and passive recovery modes applied between repeated-sprints. Madueno MC, Guy JH, Dalbo VJ, Scanlan AT. J Sports Med Phys Fitness. 2019
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis. Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
Contrast water therapy and exercise induced muscle damage: a systematic review and meta-analysis. Bieuzen F, Bleakley CM, Costello JT. PLoS One. 2013 (NOTE: yes, I know this doesn’t look like an active recovery review but some of the studies selected used active recovery as a comparator to contrast water therapy.)
What is stretching?
Stretching to increase flexibility is a mainstay in sports that require feats of bendiness, like gymnastics. Following her many years of gymnastics training and competition in her youth, 20 years on, my wife could still easily market herself as a human slinky. However, gaining extreme levels of bendiness is not necessarily a useful pursuit for a runner. Nonetheless, stretching exercises have existed since the dawn of textbooks and are regularly recommended for the purported aim of “preventing injury”, “reducing muscle soreness”, and “enhancing performance”.
Recent thoughts have focused on how performance is affected by static stretching (i.e. holding a stretch for 30–60 seconds). In previously untrained folks, some studies show that pre-lifting static stretching may reduce the number of repetitions achieved and the total volume of training while blunting resistance training-induced gains in muscle mass. But, other studies show that performance is differentially affected by varying types of pre-session stretching protocols (e.g. static, ballistic, vs. proprioceptive neuromuscular facilitation), while further studies find that static stretching of hamstrings (i.e. the antagonist muscle) between sets of squats can increase biceps femoris (hamstring) muscle thickness. Then there is the timing of stretching: some work shows that between-set quadriceps stretching can reduce neuromuscular performance of knee extension exercise in resistance-trained men. But, in previously untrained folks, between-set stretching may help facilitate increases in muscle strength and hypertrophy during the early phase of a resistance training programme. Meanwhile, other work shows that static stretching on the days between lifting sessions may augment strength gains during resistance training in novice lifters.
Confused? You should be!
This rather complex scene has received attention from several narrative reviews. One such review by James Nuzzo was even so bold to propose, “The Case for Retiring Flexibility as a Major Component of Physical Fitness”. Nuzzo argues that flexibility can be maintained or improved by other exercise modalities besides stretching — strength training, for example — that also cause robust health benefits. But, increased flexibility is not always the goal of stretching. For example, some folks use stretching to prevent injuries — some narrative reviews conclude that there’s some support that pre-exercise stretching might help prevent muscle strains but that there’s little support for stretching in overall injury prevention (see McHugh et al. 2010 & Chaabene et al. 2019).
So, some patterns are emerging but, while individual studies and narrative reviews are interesting and thought-provoking, we can do better and turn to the highest level of evidence — systematic reviews of randomised controlled trials. So, what do they say…?
What is the scientific evidence on the effect of stretching on recovery and athletic performance?
Note that this post examines the role of stretching in recovery and performance, not rehab from injury or surgery.
A single bout of stretching or regular stretching is very useful for increasing flexibility, which is important in some sports (see Ingram et al. 2024, Behm et al. 2021, Cayco et al. 2019, Medeiros et al. 2018, & Medeiros et al. 2016).
While having a high level of flexibility in certain joints is useful for hurdlers and steeplechase racers, it’s unlikely to be useful to runners. Accordingly, the current evidence shows that increased flexibility (or joint range of motion) is not associated with lower injury risk in runners (see Sanfillippo et al. 2021, Hamstra-Wright et al. 2021, & Hamstra-Wright et al. 2014) and that stretching does not prevent injuries in runners (see Alexander et al. 2019, Baxter et al. 2017, & Yeung et al. 2011). That said, the body of evidence is generally of low quality and shows large variability between studies, so further high-quality studies are needed to bolster the evidence.
If increased flexibility (or joint range of motion) is your goal, as an alternative to stretching you might consider strength training because it not only increases flexibility (see Alizadeh et al. 2023, Kay et al. 2022, Vetter et al. 2022, Afonso et al. 2021) but also improves running economyThe rate of energy expenditure (measured in kiloJoules [KJ], kilocalories [kcal] or oxygen consumption [V̇O2]) per kilogram body mass (kg) per unit of distance i.e. per 1 kilometer traveled. A runner with a lower energy cost per kilometer has a higher economy than a runner with a higher energy cost. (see O’Sullivan et al. 2019, Denadai et al. 2016, & Balsalobre-Fernández et al. 2016) and running performance (see Cuthbert et al. 2021, Trowel et al. 2019, Lum et al. 2019, Alcaraz-Ibañez et al. 2018, Blagrove et al. 2017, Beattie et al. 2014, & Yamamoto et al. 2008). Furthermore, strength training is associated with a reduced risk of chronic disease (see Coleman et al. 2022, Momma et al. 2021, & Yanghui et al. 2019).
Stretching either before or after exercise (or both) does not produce clinically important reductions in exercise-induced delayed-onset muscle soreness (see Afonso et al. 2021, & Herbert et al. 2011).
Furthermore, stretching either before or after exercise does not enhance the restoration of muscle strength performance (see Afonso et al. 2021). There is currently insufficient data to make conclusions about the effect of stretching on the recovery of endurance performance.
So, it’s rather clear that stretching is unlikely to help improve your recovery. But, the murkiness emerges when we look at the effect of stretching on performance:
Immediate pre-exercise static stretching has a trivial to small but meaningful effect to impair performance in strength, power, speed, and agility exercises (see Simic et al. 2012, Kallerud et al. 2013, & Peck et al. 2014). However, this negative effect is short-lived (e.g. a few minutes; see Peck et al. 2014 & Chaabene et al. 2019). Narrative reviews by Chaabene et al. 2019 and Rubini et al. 2007 also conclude that pre-exercise stretching probably reduces muscle strength, but that the variety of stretching protocols makes it difficult to make firm conclusions. Meanwhile, due to insufficient evidence, the effect of pre-exercise static stretching on endurance performance is less clear and more studies are needed to make firm conclusions (see Peck et al. 2014).
Dynamic stretching exercises are a suitable alternative to static stretching as part of a pre-exercise routine (e.g. the warm-up). Dynamic stretching exercises do not impair performance (see Kallerud et al. 2013, Peck et al. 2014, & Opplert et al. 2018) and their inclusion in a pre-exercise warm-up is of benefit to subsequent performance, especially for strength- and power-dominant activities (see Peck et al. 2014, & Opplert et al. 2018).
Furthermore, the negative effect of pre-exercise static stretching (during a warm-up, for example) can likely be overcome if the stretches are short or if dynamic exercises follow them (see Simic et al. 2012, & Peck et al. 2014).
Therefore, since the detrimental effect of pre-exercise static stretching likely only persists for a few minutes and is overridden by subsequent dynamic exercises, a static stretch embedded into a dynamic warm-up routine is unlikely to harm your performance. So, if you fancy a quick quad stretch on the start line because it feels good, don’t stress over whether it will ruin your race.
In contrast to the acute detrimental effects of static stretching, regular static stretching (i.e. flexibility training) alongside regular training does not appear to blunt strength training-induced gains in muscle strength (see Thomas et al. 2022). However, strength training-induced gains in muscle strength are blunted if static stretching is performed immediately prior to the strength session (see Thomas et al. 2022). So, if you choose to engage in regular stretching and don’t want to hinder your strength gains, it would be sensible to place the two modalities of exercise at different times of the day.
Furthermore, regular static stretching (flexibility training) does not affect jumping or sprinting performance (see Warneke et al. 2024).
Interestingly, regular static stretching (flexibility training) can have a trivial to small effect on increasing muscle strength and power in people not engaging in strength training (see Warneke et al. 2024, Arntz et al. 2023 & Thomas et al. 2022). This concept is known as stretch-mediated hypertrophy and strength. However, the quality of evidence is low and these effects are far more evident in sedentary/untrained folks and in older-aged people than in trained athletes (see Arntz et al. 2023). Furthermore, there is currently insufficient evidence to make conclusions about the effect of regular stretching on changes in endurance performance.
Overall, it is important to note that the inconsistent description of stretch procedures hinders the quality of evidence in this field. For example, many studies do not clearly state the stretch duration or frequency, or whether the stretch is static or dynamic including the specific type of stretch (e.g. static, ballistic, proprioceptive neuromuscular facilitation, etc). Further studies with greater methodological clarity would bolster the evidence in this field.
Can stretching enhance athletic recovery and performance?
Stretching is unlikely to improve the restoration of performance or the recovery of muscle soreness after exercise, but dynamic stretching as part of a pre-exercise warm-up routine is likely to enhance performance.
Meanwhile pre-exercise static stretching is likely to impair performance, particularly during strength or power-type exercise. The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is trivial to small, but likely clinically significantReflects how meaningful a change is to a person’s health or performance. A small change can be statistically significant but not clinically significant; but, if a change is big enough to matter to people in real life, then it is clinically significant (and doesn’t just tick a statistical significance box).. But, this issue can be prevented if the static stretch is brief or if dynamic stretching exercises are completed after the static stretches.
That said, regular static stretching (flexibility training) alongside strength training is unlikely to impair gains in strength performance unless the static stretch is performed immediately before the strength session.
These effects appear to be similar between trained athletes and untrained folks, and between males and females; however, most studies are in young trained males so further research is needed to better understand these comparisons.
Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a small number of small studies, and a moderate risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
How to use this: INSERT.
Full list of meta-analyses examining stretching for recovery.
Here are the meta-analyses I've summarised above:
Effects of Chronic Static Stretching on Maximal Strength and Muscle Hypertrophy: A Systematic Review and Meta-Analysis with Meta-Regression. Warneke et al. Sports Med Open. (2024)
Effects of chronic static stretching interventions on jumping and sprinting performance-a systematic review with multilevel meta-analysis. Warneke et al. Front Physiol. (2024)
Examining the Influence of Warm-Up Static and Dynamic Stretching, as well as Post-Activation Potentiation Effects, on the Acute Enhancement of Gymnastic Performance: A Systematic Review with Meta-Analysis. Yu et al. J Sports Sci Med. (2024)
Chronic Effects of Static Stretching Exercises on Muscle Strength and Power in Healthy Individuals Across the Lifespan: A Systematic Review with Multi-level Meta-analysis. Fabian Arntz, Adrian Markov, David G. Behm, Martin Behrens, Yassine Negra, Masatoshi Nakamura, Jason Moran & Helmi Chaabene. Sports Med. 2023
Does Stretching Training Influence Muscular Strength? A Systematic Review With Meta-Analysis and Meta-Regression. Ewan Thomas, Salvatore Ficarra, João Pedro Nunes, Antonio Paoli, Marianna Bellafiore, Antonio Palma, Antonino Bianco. J Strength Cond Res. 2022
A Comparison of the Effects of Foam Rolling and Stretching on Physical Performance. A Systematic Review and Meta-Analysis. Andreas Konrad, Markus Tilp, Masatoshi Nakamura. Front Physiol. 2021
The Effectiveness of Post-exercise Stretching in Short-Term and Delayed Recovery of Strength, Range of Motion and Delayed Onset Muscle Soreness: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Afonso J, Clemente FM, Nakamura FY, Morouço P, Sarmento H, Inman RA, Ramirez-Campillo R. Front Physiol. 2021
The Effectiveness of Post-exercise Stretching in Short-Term and Delayed Recovery of Strength, Range of Motion and Delayed Onset Muscle Soreness: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Afonso J, Clemente FM, Nakamura FY, Morouço P, Sarmento H, Inman RA, Ramirez-Campillo R. Front Physiol. 2021
Do alterations in muscle strength, flexibility, range of motion, and alignment predict lower extremity injury in runners: a systematic review. Christopher SM, McCullough J, Snodgrass SJ, Cook C. Arch Physiother. 2019
Acute Effects of Static Stretching on Muscle Strength and Power: An Attempt to Clarify Previous Caveats. Chaabene H, Behm DG, Negra Y, Granacher U. Front Physiol. 2019
Acute Effects of Dynamic Stretching on Muscle Flexibility and Performance: An Analysis of the Current Literature. Opplert J, Babault N. Sports Med. 2018
Influence of chronic stretching on muscle performance: Systematic review. Medeiros DM, Lima CS. Hum Mov Sci. 2017
The effects of stretching on performance. Peck E, Chomko G, Gaz DV, Farrell AM. Curr Sports Med Rep. 2014
Effects of stretching on performances involving stretch-shortening cycles. Kallerud H, Gleeson N. Sports Med. 2013
Does pre-exercise static stretching inhibit maximal muscular performance? A meta-analytical review. Simic L, Sarabon N, Markovic G. Scand J Med Sci Sports. 2013
Stretching to prevent or reduce muscle soreness after exercise. Herbert RD, de Noronha M, Kamper SJ Cochrane Database of Systematic Reviews. 2011
What is cold water immersion?
CrossFit athletes plunging their arms into ice water after crushing Murph... Mo Farah lowering himself into an ice-cold water tub after a lengthy interval workout... Social media “influencers” heading for their weekly cryo-chamber visit... Wim Hof climbing mountains in the winter in his underpants... These images are all visually stimulating and even quite exciting. But, why do they do it?
Plunging into ice-cold water does two things: First, you yell some expletive like, “By the beard of Zeus”. Second, your autonomic nervous system immediately goes nuts — cold water immersion induces shivering (aka muscle contraction-induced heat production) plus a robust increase in epinephrine (adrenaline), heart rate, and breathing rate — just like exercise. I.e. the opposite of rest 
. But, cold water immersion after a session lowers core body temperature back to normal. After the initial shock of entering the low temperature, the parasympathetic nervous system is activated to help “calm shiz down” (the physiological responses of cold water immersion have been systematically reviewed in the British Journal of Sports Medicine). For some folks, cold water immersion also lowers the feeling of fatigue and lessens the feeling of muscle soreness — it “feels” like recovery. Subjective feelings of recovery (less muscle soreness) using cold water immersion have been confirmed in runners recovering from a half-marathon (although objective measures including creatine kinase, testosterone, and muscle function were not improved). Some folks even find cold water immersion exhilarating (I am in this camp) because of the literal adrenaline rush. Meanwhile, others hate it. I even know athletes who hate cold water plunges yet still jump in because they think they’re supposed to!?
With regards to the restoration of performance, there are several proposed mechanisms and many experimental studies examining the acute and chronic effects of cold water immersion on both strength and endurance exercise modalities. When you delve into the individual studies, you find a mixed bag of outcomes but the adaptive biochemical and functional responses have been very well summarised in a 2018 narrative review in Sports Medicine. Fortunately, there are a sufficient number of high-quality experimental studies that can be systematically reviewed and meta-analysed to help you learn how to effectively use such cold immersion in practice.
What is the scientific evidence on the effect of cold water immersion, ice baths, and cryotherapy on recovery and athletic performance?
Cold water immersion after exercise can reduce markers of muscle damage (e.g. creatine kinase) and feelings of muscle soreness and fatigue, when compared to passive recovery, aka rest (see Feng et al. 2024, Xiao et al. 2023, Moore et al. 2022, Wang et al. 2021, Machado et al. 2016 & Hohenauer et al. 2015).
Cold water immersion after exercise can also help restore muscle power performance (e.g. sprinting, jump height), particularly after high-intensity exercise (see Moore et al. 2022, & Poppendieck et al. 2013). However, cold water immersion after exercise does not affect the recovery of muscle strength or endurance performance.
To achieve these effects, the current recommended cold water immersion/ice bath protocol is to immerse the body up to the neck in cold water at a temperature of between 10 and 15°C for 5 to 15 minutes (see Machado et al. 2016). Also see Halson (2011) for a narrative of the practicalities of cold water immersion.
A study by Stephens et al. 2017 found that the drop in core temperature when immersed in cold water is faster in people with lower body fat percentage; therefore, you should consider your body composition when designing your cold water immersion strategy to ensure you stay safe and don’t develop hypothermia.
Unlike cold water immersion, simply icing the local area of sore muscle does not reduce feelings of muscle soreness (DOMS) or improve the restoration of performance following muscle-damaging exercise (see Nogueira et al. 2019).
Similar to the effect of post-exercise cold water immersion, contrast water therapy (intermittent cold and hot water immersion) after exercise can also reduce feelings of muscle soreness when compared to passive recovery (see Bieuzen et al. 2013).
Due to the lack of precise methodological details provided in many studies (e.g. temperature, duration, type of cold exposure etc.), the lack of blinding, and the combination of cold therapy with other “recovery” approaches in many studies, the current quality of evidence in this field is low and high-quality randomised controlled trials are urgently needed to bolster the evidence.
Furthermore, there is insufficient evidence (too few studies) to determine whether whole‐body cryotherapy — i.e., single or repeated exposures to extremely cold dry air below -100°C in a specialised chamber — reduces self‐reported muscle soreness, improves feelings of recovery, or restores performance after exercise (see Feng et al. 2024, Costello et al. 2015). Some evidence shows that whole‐body cryotherapy can lower some inflammation markers (e.g., C-reactive protein) elevated by post-exercise fatigue but it is unclear whether this is beneficial and there are very few studies that have investigated this phenomenom (Feng et al. 2024).
When different cold therapy modalities are compared, there is currently little difference between their effects (see Azevedo et al. 2022). However, there are very few studies making such comparisons, and the phrase “cryotherapy” (which I refer to as cold dry air) is also used in some studies to mean cold water immersion, which can get confusing.
However, THE MOST IMPORTANT THING to know about cold water immersion is the difference between the effects of an acute (one) exposure and chronic (regular) exposures:
— After resistance exercise, regular cold water immersion may reduce training adaptations. For example, cold water immersion can blunt molecular signalling (Roberts et al. 2015) and suppress muscle protein synthesis (Fuchs et al. 2019) after a session. These effects were reviewed by Broatch et al. 2018. Furthermore, the 2021 systematic review from Malta et al. found that regular cold water immersion during resistance training blunts muscle mass and strength gains, including 1-rep max, maximum strength, strength endurance, and power.
— After endurance exercise, the inhibitory effect of cold therapy is less apparent. For example, some evidence indicates that one session of cold water immersion after an endurance exercise bout may actually increase molecular signals associated with mitochondrial biogenesis (see Broatch et al. 2018). But, there is little to no effect of regular cold therapy on training-induced changes in molecular signalling (Broatch et al. 2018) or endurance performance, including time-trial performance and V̇O2maxYour maximal rate of oxygen consumption; a measure of cardiorespiratory fitness and maximal aerobic power, which contributes to endurance performance. (see Malta et al. 2021). All that said, there are only a few studies addressing these concepts and further high-quality randomised controlled trials are needed to bolster the current evidence.
— Therefore, after your hard effort or lifting session, sure, go take a dip in that cold lake (if ambient conditions permit and adequate warm and dry post-dip facilities are available). But, it’s probably not a good idea to do it every day or you might prevent the improvements you are sweating so hard to achieve!
Can cold water immersion or ice baths enhance athletic recovery and performance?
Cold water immersion (an ice bath) is likely to improve the recovery of muscle soreness and help restore muscle power performance after high-intensity exercise. There are too few studies on cryotherapy (cold dry air) to fully understand its role in recovery or performance.
The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is moderate to large for muscle soreness, and small to moderate for performance.
The effect appears to be similar between trained athletes and untrained folks, and between males and females; however, comparative analyses are few in number, and further research is needed in females because they are underrepresented among studies in this field.
Importantly, there is a moderate-sized detrimental effectA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. of daily cold water immersion (or ice baths) on training-induced gains in muscle strength, strength endurance, and power. So, daily cold water immersion is probably not a good idea for athletes interested in improving their strength performance.
Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in effects between studies, a moderate to high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported., and possible publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
How to use this: INSERT.
Full list of meta-analyses examining cold water immersion, ice baths, and cryotherapy for recovery.
Here are the meta-analyses I've summarised above:
The effects of hydrotherapy and cryotherapy on recovery from acute post-exercise induced muscle damage-a network meta-analysis. Chen et al. BMC Musculoskelet Disord. (2024)
A evidence-based approach to selecting post-exercise cryostimulation techniques for improving exercise performance and fatigue recovery: A systematic review and meta-analysis. Feng et al. Heliyon. (2024)
Effects of cold water immersion after exercise on fatigue recovery and exercise performance--meta analysis. Feiyan Xiao, Anastasiia V. Kabachkova, Lu Jiao, Huan Zhao, Leonid V. Kapilevich. Front Physiol. 2023
Effects of Cold-Water Immersion Compared with Other Recovery Modalities on Athletic Performance Following Acute Strenuous Exercise in Physically Active Participants: A Systematic Review, Meta-Analysis, and Meta-Regression. Emma Moore, Joel T Fuller, Clint R Bellenger, Siena Saunders, Shona L Halson, James R Broatch, Jonathan D Buckley. Sports Med. 2023
The effect of cold water immersion on the recovery of physical performance revisited: A systematic review with meta-analysis. Hui Cheng Choo, Marcus Lee, Vincent Yeo, Wayne Poon, Mohammed Ihsan. Sports Sci. 2022
Different Cryotherapy Modalities Demonstrate Similar Effects on Muscle Performance, Soreness, and Damage in Healthy Individuals and Athletes: A Systematic Review with Meta analysis. Klaus Porto Azevedo, Júlia Aguillar Ivo Bastos, Ivo Vieira de Sousa Neto, Carlos Marcelo Pastre, Joao Luiz Quagliotti Durigan. J Clin Med. 2022
Impact of Cold-Water Immersion Compared with Passive Recovery Following a Single Bout of Strenuous Exercise on Athletic Performance in Physically Active Participants: A Systematic Review with Meta-analysis and Meta-regression. Emma Moore, Joel T. Fuller, Jonathan D. Buckley, Siena Saunders, Shona L. Halson, James R. Broatch & Clint R. Bellenger. Sports Med. 2022
Comparison between cryotherapy and photobiomodulation in muscle recovery: a systematic review and meta-analysis. João Vitor Ferlito, Marcos Vinicius Ferlito, Ernesto Cesar Pinto Leal-Junior , Shaiane Silva Tomazoni, Thiago De Marchi. Lasers Med Sci. 2021
Heat and cold therapy reduce pain in patients with delayed onset muscle soreness: A systematic review and meta-analysis of 32 randomized controlled trials. Yutan Wang, Sijun Li, Yuanyuan Zhang, Yanru Chen, Fanghong Yan, Lin Han, Yuxia Ma. Phys Ther Sport. 2021
The Effects of Regular Cold-Water Immersion Use on Training-Induced Changes in Strength and Endurance Performance: A Systematic Review with Meta-Analysis. Malta ES, Dutra YM, Broatch JR, Bishop DJ, Zagatto AM. Sports Med. 2021
The effects of cryotherapy on athletes' muscle strength, flexibility, and neuromuscular control: A systematic review of the literature. Kalli K, Fousekis K. J Bodyw Mov Ther. 2020
Effects of local cryotherapy for recovery of delayed onset muscle soreness and strength following exercise-induced muscle damage: systematic review and meta-analysis. Nogueira NM, Felappi CJ, Lima CS, Medeiros DM. Sport Sciences for Health. 2019
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis. Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
Can Water Temperature and Immersion Time Influence the Effect of Cold Water Immersion on Muscle Soreness? A Systematic Review and Meta-Analysis. Machado AF, Ferreira PH, Micheletti JK, de Almeida AC, Lemes ÍR, Vanderlei FM, Netto Junior J, Pastre CM. Sports Med. 2016
The Effect of Post-Exercise Cryotherapy on Recovery Characteristics: A Systematic Review and Meta-Analysis. Hohenauer E, Taeymans J, Baeyens JP, Clarys P, Clijsen R. PLoS One. 2015
Whole-body cryotherapy (extreme cold air exposure) for preventing and treating muscle soreness after exercise in adults. Costello JT, Baker PR, Minett GM, Bieuzen F, Stewart IB, Bleakley C. Cochrane Database Syst Rev. 2015
Cooling and performance recovery of trained athletes: a meta-analytical review. Poppendieck W, Faude O, Wegmann M, Meyer T. Int J Sports Physiol Perform. 2013
Contrast water therapy and exercise induced muscle damage: a systematic review and meta-analysis. Bieuzen F, Bleakley CM, Costello JT. PLoS One. 2013
Cold water immersion and recovery from strenuous exercise: a meta-analysis. Leeder J, Gissane C, van Someren K, Gregson W, Howatson G.
Br J Sports Med. 2012
Evidence of the physiotherapeutic interventions used currently after exercise-induced muscle damage: systematic review and meta-analysis.
Torres R, Ribeiro F, Alberto Duarte J, Cabri JM.
Phys Ther Sport. 2012
What is hot water immersion?
Many folks love a good soak in a warm bath. During the closing miles of a long winter run, I always daydream of how high I can stack the bubbles when I get home… Lying in warm water can be truly relaxing.
As a four-eyed ginger youngster in the 80s, I played football and lived for 3 pm every Saturday. When the FA Cup rolled around each year, I was enamoured by the post-match drinking celebrations in the dressing rooms — everyone looked so happy. But there is another memory firmly etched in my neurons… the massive bathtubs. It was totally normal (and occasionally televised) for 11 men to get bollock-naked and take a bath after the game, simultaneously, in the same filthy tub. I asked everyone why they did it; my coaches, my parents, and my teammates. In return, I was treated to all sorts of wild answers, my favourite response being an Irish-accented, “they just love it”, from my Grandad. My tiny eight-year-old brain concluded that they just had to get clean before dinner and this was the most efficient method. But, was the seemingly inadvertent immersion in the steamy H2O having beneficial effects besides impromptu Saturday afternoon man-porn? Well, let’s take a look at the evidence...
What is the scientific evidence on the effect of hot water immersion (a hot bath) on recovery and athletic performance?
Please note that I am talking about recovery from your sessions and races. None of the studies I refer to here is relevant to the recovery from injury, illness, or disease, nor are they relevant to heat acclimation (you can go deep on that in my heat acclimation article at veohtu.com/heat-3-howtoheatacclimate).
Hot water immersion studies have typically immersed subjects for between 15 and 25 minutes with water between 35 and 45°C.
Experimental hot water immersion studies have typically compared hot water immersion to cold water immersion or contrast water therapy (hot and cold), or both. Some studies have also made comparisons to passive recovery (rest), active recovery (light jogging/spinning), or other means of magic.
Although one systematic review of post-exercise “heat therapy” found a benefit for reducing muscle soreness, it combined studies examining hot water immersion with studies examining local heat packs, infra-red phototherapy, and sauna. This review also noted the lack of high-quality studies and that more work is needed.
There is currently no systematic review specifically examining the effect of hot water immersion on recovery. The current evidence (here, here, here, and here) shows that hot water immersion is probably not beneficial for lessening exercise-induced delayed onset muscle soreness (DOMS) and/or feelings of fatigue when compared to cold water immersion, contrast water therapy, or passive recovery. Although there is no systematic review of this topic, a 2018 narrative review from McGorm et al. nicely summarises these sentiments.
Very recent work, published in 2020, found that hot-water immersion during recovery from a single bout of resistance exercise did not further increase muscle protein synthesis rates or augment the postprandial incorporation of dietary amino acids into muscle in young people. Thus, the sentiments floating around that post-lifting heat therapy causes more hypertrophy are not looking credible — it is probably the vasodilation caused by the post-session heat that simply makes muscles appear to be massive.
So, diving into your warm bath after a session is not going to have any magic effect on your recovery or restoration of performance but it probably won’t hurt. But do keep in mind that your core temperature rises when you work out and needs to return to normal for you to recover and stay healthy. For this reason, be sensible and use warm baths when you fancy some R&R, fancy getting warm after a chilly session, or if you are using a heat acclimation strategy for an upcoming race.
What is the scientific evidence on the effect of contrast water therapy (intermittent hot and cold) on recovery and athletic performance?
Contrast water therapy studies have typically used intermittent immersions in cold (10 to 15°C) and hot (35 to 45°C) at 1:1 to 3:1 time duration ratios.
There is currently insufficient research to determine optimal immersion durations, water temperatures, or time ratios (in the case of contrast therapy).
The systematic review by Bieuzen et al. (2013) found that contrast water therapy lessened exercise-induced muscle damage and feelings of soreness when compared to passive recovery (rest) and hot water immersion, but not when compared to other recovery modalities like active recovery or cold water immersion.
Although the review from Bieuzen et al. focussed on muscle damage, the authors also conducted a sub-group analysis of performance finding that contrast water immersion may slightly improve recovery of muscle strength but not muscle power when compared to passive recovery or hot water immersion.
However, Bieuzen et al. (2013) concluded that “Overall the study quality in this review was low. The majority of studies had a high risk of bias making the validity of most of the results uncertain.”.
Can hot water immersion (a hot bath) or contrast water therapy enhance athletic recovery and performance?
Post-exercise hot water immersion is unlikely to improve recovery from exercise.
On the other hand, contrast water therapy (intermittent hot and cold) is somewhat likely to alleviate exercise-induced muscle soreness with a moderate effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients..
Furthermore, regular post-exercise hot water immersion is likely to induced heat acclimation, which could potentially improve performance in the heat (go deep on heat acclimation at veohtu.com/heatacclimation).
These effects appear to be similar between trained athletes and untrained folks, and between males and females; however, because of sparse research such comparisons are uncertain and further research is needed in females because they are underrepresented among studies in this field.
Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported., and possible publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: neither post-exercise hot water immersion nor post-exercise contrast water therapy appear to have a detrimental effect on recovery or performance. So, if you like them and believe they work for you, give em a whirl. However, if you choose such approaches, it would be sensible to drink to thirst to prevent heat-induced dehydration so you can prevent dehydration-induced impairments in your recovery and performance. Go deep on hydration at veohtu.com/hydration. Also, be aware that because increased core temperature can impair performance, pre-exercise heat exposure is probably best to avoid. Read all about that at veohtu.com/heat.
How to use this: INSERT.
Full list of meta-analyses examining hot water immersion (hot baths) and contrast water therapy for recovery.
Here are the meta-analyses I've summarised above:
Effects of Post-Exercise Heat Exposure on Acute Recovery and Training-Induced Performance Adaptations: A Systematic Review. Ahokas et al. Sports Med Open. 2025
The effect of post-exercise heat exposure (passive heat acclimation) on endurance exercise performance: a systematic review and meta-analysis. Solomon and Laye. BMC Sports Sci Med Rehabil. 2025
Heat and cold therapy reduce pain in patients with delayed onset muscle soreness: A systematic review and meta-analysis of 32 randomized controlled trials. Yutan Wang, Sijun Li, Yuanyuan Zhang, Yanru Chen, Fanghong Yan, Lin Han, Yuxia Ma. Phys Ther Sport. 2021
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis.
Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
Contrast water therapy and exercise induced muscle damage: a systematic review and meta-analysis.
Bieuzen F, Bleakley CM, Costello JT.
PLoS One. 2013
What is sauna?
A sauna is usually composed of a wooden shack with burning embers that increase the temperature to around 60–80°C, sometimes evaporating water into the atmosphere — a “wet” sauna in which the humidity is greater than 50%) — or not — a “dry” sauna in which the humidity is less than 20%. In a wet sauna, you’ll likely feel very sweaty; in a dry sauna, you probably won’t because the dry “low humidity” air sucks that sweat straight off your skin. Many folks feel relaxed, rested, refreshed, and re-invigorated when they walk out of a sauna. Many also do not. I sit in the latter camp. A sauna gives me a feeling of excess fatigue. The only thing I enjoy about sweating my balls off in a hot box is inhaling the hot humid air, which my asthma-infested lungs find very soothing.
The list of proposed benefits of regular sauna treatment is as long as a piece of string. Some popular podcasters and “influencers” regularly promote the health benefits of regular sauna (e.g. here), and indeed some experimental studies find that regular sauna treatment can lower stress, anxiety, inflammation, etc etc (reviewed here and here). For example, one of the largest studies, the Kuopio Ischemic Heart Disease Risk Factor Study, which surveyed ~3000 middle-aged Finnish men and women over 20-years, found an association between sauna use and a reduced risk of hypertension (high blood pressure), stroke, cardiovascular-related mortality (death), and dementia (inc. Alzheimer's disease), even when controlling for age, socioeconomic status, diet, and physical activity. Some evidence, mostly from animal models, also shows that passive heat exposure (e.g. sauna) without exercise might help maintain muscle mass and/or force development (see here and here; however, further high quality randomised controlled trials are needed to confirm this effect in humans.
Yes, the Finns swear by it and I have been privileged to have visited pine-cabin saunas with many a Finn in various parts of their great country — it is indeed true that many apartment buildings even have a sauna in the basement. And, nothing beats giving a scientific talk in front of university academics, several of whom have experienced your full monty in a sauna just hours previously!
The associative role of sauna in the prevention of chronic disease certainly sounds impressive, but this line of evidence is derived from associations found in prospective long-term cohort studies where diet, activity, and sauna use are documented with questionnaire-based recall rather than being prescribed using a randomised controlled trial design (which is needed to provide confidence to a direct causal effect). Importantly, if you avoid dehydration and heat stress, there are also no adverse effects of a sauna.
But, what about using a sauna to aid the recovery of your performance?
What is the scientific evidence on the effect of sauna on recovery and athletic performance?
Be mindful that this article is about the recovery from and adaptations to exercise that restore your performance. This article is not about the recovery from injury or illness nor is it relevant to the use of a sauna as an adjunct for the maintenance of health.
Please also note that this article is not about using a sauna for helping with heat acclimation (you can go deep on that in my heat acclimation article at veohtu.com/heatacclimation).
When you sit in a sauna, your heart rate, breathing rate, sweat rate, oxygen consumption (VO2), cardiac output, and blood pressure will all increase — similar physiological effects that occur during exercise. But whether regular trips to the sauna can be a replacement for regular exercise or whether regular trips to the sauna have the same health benefits as regular exercise, is unknown — be wary of any journalist or former cell biologists-turned public health gurus telling you otherwise (e.g. here and here). Yes, there’s a long list of physiological things that change when you expose yourself to an unusually hot environment, but these acute changes are not necessarily signs of adaptation and improvements, they are simply indications that your body is trying to maintain homeostasis under the stress/stimulus of the heat. Interestingly, a bacterial/viral infection will also increase your heart rate, breathing rate, sweat rate, oxygen consumption (VO2), cardiac output, and blood pressure, so just because something has similar acute effects to exercise doesn’t mean it is healthy or causing the same long-term health adaptations.
Since sitting in a sauna (or any hot environment) increases your sweat rate, it is important to be aware of your feelings of thirst and to quench it to stay hydrated. After all, dehydration will impair your recovery from, and adaptations to, your training. It is also important to avoid heat stress. So, exit the (hot) building if you feel like poop. (To go deep on hydration, check out my series on hydration at veohtu.com/hydration-1-whatweknow.)
Most of the sauna research on exercise-related recovery has been done on rodents, the results of which cannot be confidently extrapolated to humans.
There are very few human studies examining the effect of sauna on recovery, and there is currently no systematic review of the literature or a meta-analysis.
Some studies have found that a post-exercise sauna impairs swimming time trial performance the following day (in 30 trained swimmers and triathletes) and that maximal voluntary contraction (MVC) force, a marker of muscle strength, is reduced following a sauna in untrained folks. Therefore, it is probably unwise to take a sauna immediately before, or even the day before, a maximal effort (session or race). Another study of 10 healthy men found that a far-infrared sauna after a ~40-min maximal-effort endurance session may restore neuromuscular performance (countermovement jump height). But, larger, high-quality randomised controlled trials are needed to fully understand these effects.
If you feel relaxed, rested, refreshed, and reinvigorated when you leave a sauna, then go for it. If you do not, then the sauna is not something you need to add to your recovery toolbox. That said, using a post-exercise sauna as part of a heat acclimation strategy can help improve performance in the heat (read all about that in my heat acclimation article at veohtu.com/heat-3-howtoheatacclimate).
Can sauna enhance athletic recovery and performance?
Due to a lack of research, it is currently unclear how post-exercise sauna affects recovery from exercise.
That said, regular post-exercise heat exposure using either hot water immersion or sauna is likely to induce heat acclimation, which could potentially improve performance in the heat (go deep on heat acclimation at veohtu.com/heatacclimation). However, the current evidence shows that the effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is trivial to small.
.
Due to a lack of research, it is unclear how effects compare between trained athletes and untrained folks, and between males and females.
Keep in mind: because of a lack of research, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is currently lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to understand potential effects and to conduct ameta-analysisA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: post-exercise sauna doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm. Also be aware that because increased core temperature can impair performance, pre-exercise heat exposure is probably best to avoid. Read all about that at veohtu.com/heat.
Also remember that drinking to thirst to prevent sauna-induced dehydration will prevent dehydration-induced impairments in your recovery and performance. So, when in a sauna, always be mindful of your feelings of thirst, and always leave the sauna if you feel unwell. Go deep on hydration at veohtu.com/hydration.
How to use this: INSERT.
Full list of meta-analyses examining sauna for recovery.
Here are the meta-analyses I've summarised above:
There are currently no meta-analyses of sauna for recovery from exercise. However, there are some analyses examining the role of post-exercise heat exposure (via a hoit bath or sauna) on heat acclimation and performance:
Effects of Post-Exercise Heat Exposure on Acute Recovery and Training-Induced Performance Adaptations: A Systematic Review. Ahokas et al. Sports Med Open. 2025
The effect of post-exercise heat exposure (passive heat acclimation) on endurance exercise performance: a systematic review and meta-analysis. Solomon and Laye. BMC Sports Sci Med Rehabil. 2025
What is infrared phototherapy/photobiomodulation therapy (PBMT)?
The local application of heat to an area of tissue that needs recovery has been all the rage since Mr Miyagi suppressed the feeling of pain in Daniel LaRusso leg before he headed back out into the ring to unleash his “Crane-kick”... That classic story.
“Old school” methods using hot water-perfused garments have found that heat-treatment of the thigh (~52°C) for 90-mins 5-days/week for 8-weeks may increase knee extensor strength in untrained young healthy adults. Similarly, just 5 daily 90-min sessions of thigh heat treatment (54–55°C) following eccentric knee extensor exercise (300 bilateral maximal eccentric contractions), has been found to hasten the recovery of fatigue resistance (but not strength) of the knee extensor in untrained young healthy adults. Interesting.
Mr Miyagi’s use of local heat to dull LaRusso’s leg pain clearly helped him “recover” mid-match to destroy Johnny Lawrence at the All-Valley Karate Championships. However, things have advanced since the ole friction-based hand-rub application of heat technique. In this new-fangled world of tech, infrared phototherapy has emerged from the burning embers. But, infra-red phototherapy is like a spy in that it goes by a variety of identities, including photobiomodulation, diathermy, infra-red low-level laser therapy, light-emitting diode (LED) therapy, etc etc. Whatever its name, it involves firing a beam of light of an infra-red (or near-infrared) wavelength — which we humans can’t see — at the muscle of interest.
A narrative review of the literature concluded that there is a potential benefit for infra-red phototherapy in sports performance, particularly for muscle strength/hypertrophy, which might raise its ethical use in line with current WADA rules. Indeed as the number of studies grows, the evidence shows support for a role in boosting the hypertrophic or strength stimulus of weight lifting in untrained men and elderly women. An identical-twin case study also documented the use of infra-red phototherapy for boosting strength and reducing muscle damage. Another case-study found potential endurance-enhancing benefits in one elite runner, but this finding was not confirmed in a small randomised controlled trial.
All the above sounds rather impressive but these studies are more relevant to performance rather than the recovery of performance. Furthermore, the studies are simply cherry-picked from the literature. We can do better than that…
What is the scientific evidence on the effect of infrared phototherapy/photobiomodulation therapy on recovery and athletic performance?
Despite a potential effect of pre-exercise infra-red phototherapy on lowering post-exercise circulating levels of creatine kinase (CK), a biochemical marker of muscle damage (see Li et al. 2024, Machado et al. 2020, Luo et al. 2022, & Nampo et al. 2016), infra-red phototherapy is unlikely to decrease muscle soreness or pain caused by exercise (see Nampo et al. 2016). However, there are few studies, and the large between-study variability of the effects of phototherapy prevents firm conclusions at this time. There is also currently no data on the recovery of performance following exercise.
Using infra-red phototherapy before resistance exercise does not improve muscle strength (Bezerra et al. 2023), but can extend the time-to-fatigue (reps-to-failure), increase total reps, and reduce the decline in maximal voluntary isometric contraction during resistance exercise (see Li et al. 2024, Dutra et al. 2022, Luo et al. 2022, & Leal-Junior et al. 2013). Phototherapy before resistance exercise may also increase the recovery of feelings of fatigue after a session (Bezerra et al. 2023).
Using infra-red phototherapy before endurance exercise might extend the time-to-exhaustion during cycling (see Dutra et al. 2022 and Luo et al. 2022), but there are very few studies on endurance performance (including running performance), and the effects on time trial performance needs further research to make firm conclusions. Furthermore, the current evidence does not show a beneficial effect of infra-red phototherapy alone or combined with a training program on improving running performance measured using either time-trial or time-to-exhaustion tests (see do Nascimento et al. 2024 a).
The beneficial effects on performance are predominantly found when administering phototherapy before exercise, rather than after (see Dutra et al. 2022, Luo et al. 2022, Leal-Junior et al. 2013, & Borsa et al. 2013).
As found in the systematic review by Leal-Junior et al., the largest and most consistent results are found when: (i) red or infrared wavelengths are used, (ii) phototherapy is used before exercise, and (iii) light provides between 50–200 milliwatts at a dose of 5–6 joules per spot. These specifications are worth checking if you choose to opt for a phototherapy device.
It is important to note that, currently, there are only a handful of studies on this topic, many of which are published by the same research group. Additional independent studies are therefore needed to bolster the evidence base.
Another important application was found by Hafen et al. 2019 where infra-red heat treatment of the quadriceps muscle (which raised muscle temp 4.2±0.29°C above normal body temperature) for 2 hours daily for 10-days reduced muscle atrophy in the immobilized leg of physically-active and healthy adults. This shows promise for the use of infra-red phototherapy in maintaining muscle mass during prolonged bed rest (injury, illness, hospitalisation) or old age, but more studies are needed to make firm conclusions.
Furthermore, phototherapy might outperform cryotherapy for improving relief from exercise-induced muscle soreness (DOMS) and damage (see Ferlito et al. 2021), but far more research is needed to bolster that claim.
Importantly, there are no documented adverse effects of phototherapy so it’s unlikely to harm your recovery.
Can infrared phototherapy/photobiomodulation therapy enhance athletic recovery and performance?
Pre-exercise infrared phototherapy/photobiomodulation therapy (PBMT) is likely to improve feelings of muscle soreness and reduce fatigue-induced drops in muscle strength, but due to the lack of research its effect on the recovery of performance is unclear.
Pre-exercise infrared phototherapy/PBMT is also likely to improve performance during muscular endurance tasks, such as reps-to-failure and total reps; however, muscle strength does not appear to improve.
Some evidence suggests that pre-exercise infrared phototherapy/PBMT might improve endurance time-to-exhaustion during cycling, but the evidence is unclear for running.
Some evidence suggests that infrared phototherapy/PBMT might be useful for maintaining muscle mass during prolonged bed rest (injury, illness, hospitalisation).
The effect sizesA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. are generally small, and appear to be similar between trained athletes and untrained folks, and between males and females. However, few studies have directly examined such comparisons.
Keep in mind: there are few studies on this topic and many have been published by the same lab. Furthermore, there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in effect sizes between studies, a moderate to high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported., and possible publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
Furthermore, further research is needed to determine the optimal dose and light wavelength of infrared phototherapy/PBMT.
The nice part: Infrared phototherapy/PBMT doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
How to use this: INSERT.
Full list of meta-analyses examining infrared phototherapy/photobiomodulation therapy for recovery.
Here are the meta-analyses I've summarised above:
Effects of photobiomodulation, intermittent pneumatic compression and neuromuscular electrical stimulation on muscle recovery: Systematic review with meta-analysis. Canez et al. J Bodyw Mov Ther. 2025
Photobiomodulation therapy as an adjunct to resistance exercises on muscle metrics, functional balance, functional capacity, and physical performance among older adults: A systematic scoping review. Kumar et al. Lasers Med Sci. (2024)
Can pre-exercise photobiomodulation improve muscle endurance and promote recovery from muscle strength and injuries in people with different activity levels? A meta-analysis of randomized controlled trials. Li et al. Lasers Med Sci. (2024)
A Meta-Analysis of Randomized Controlled Trials on the Effects of Photobiomodulation Therapy on Running Performance. Nascimento et al. Int J Exerc Sci. (2024)
Effects of photobiomodulation therapy on the functional performance of healthy individuals: a systematic review with meta-analysis. Bezerra LO, de Macedo LES, da Silva MLA, de Oliveira JMP, de Morais Gouveia GP, de Andrade PR, Micussi MTABC. Lasers Med Sci. 2023
Deconstructing the Ergogenic Effects of Photobiomodulation: A Systematic Review and Meta-analysis of its Efficacy in Improving Mode-Specific Exercise Performance in Humans. Yago M Dutra, Elvis S Malta, Amanda S Elias, James R Broatch, Alessandro M Zagatto. Sports Med. 2022
Effects of Low-Level Laser Therapy on Muscular Performance and Soreness Recovery in Athletes: A Meta-analysis of Randomized Controlled Trials. Wun-Ting Luo, Chieh-Jui Lee, Ka-Wai Tam, and Tsai-Wei Huang. Sports Health. 2022
Comparison between cryotherapy and photobiomodulation in muscle recovery: a systematic review and meta-analysis. João Vitor Ferlito, Marcos Vinicius Ferlito, Ernesto Cesar Pinto Leal-Junior , Shaiane Silva Tomazoni, Thiago De Marchi. Lasers Med Sci. 2021
Phototherapy on Management of Creatine Kinase Activity in General Versus Localized Exercise: A Systematic Review and Meta-Analysis.
Machado AF, Micheletti JK, Lopes JSS, Vanderlei FM, Leal-Junior ECP, Netto Junior J, Pastre CM. Clin J Sport Med. 2020
Effect of phototherapy (low-level laser therapy and light-emitting diode therapy) on exercise performance and markers of exercise recovery: a systematic review with meta-analysis.
Leal-Junior EC, Vanin AA, Miranda EF, de Carvalho Pde T, Dal Corso S, Bjordal JM.
Lasers Med Sci. 2015
Effect of low-level phototherapy on delayed onset muscle soreness: a systematic review and meta-analysis.
Nampo FK, Cavalheri V, de Paula Ramos S, Camargo EA.
Lasers Med Sci. 2015
Does phototherapy enhance skeletal muscle contractile function and postexercise recovery? A systematic review.
Borsa PA, Larkin KA, True JM.
J Athl Train. 2013
What is compression clothing?
Compression garments were initially developed to help prevent deep-vein thrombosis and to relieve the symptoms of peripheral vascular disease, a clinical condition where cardiac output is too low for blood to efficiently return to the heart. In the context of recovery, compression clothes are marketed to “reduce swelling”, “increase blood flow” (assist venous return), and “prevent blood pooling” in the limbs.
Wearing compression clothing like stockings, tights, and sleeves can feel like a massive hug — most of us love a good hug. Many folks also swear they “feel” more recovered when they wear their compression stockings after a good workout. Many of the athletes I have worked with echo these sentiments. From a scientific perspective, in recreational basketball players, compression clothing has been shown to improve the feeling of recovery but not performance. While, in rugby players, the recovery of strength following fatiguing exercise has been found to be greater in those who wore custom-fitted compression garments. Other experimental studies have found that compression clothing may reduce muscle vibrations during running. But, such findings have not always been reproduced and such interventions are typically rather short (the previous example used 3-minute runs at 8, 10, and 12 kph). Furthermore, whether reducing muscle vibration extrapolates to better recovery, performance, and/or injury prevention is unclear.
These study examples are simply cherry-picked anecdotes and, as a healthy and smart athlete, you may be wondering why you need something to cause blood to flow when you thought your heart and vascular system were already doing a fine job. “Medical grade” compression has indeed been shown to promote blood flow and tissue oxygenation in patients with vascular disease, but what does the science say about the role of compression clothing in the recovery of athletic performance?
What is the scientific evidence on the effect of compression clothing on recovery and athletic performance?
Wearing compression garments after exercise may have moderate to large effects on increasing heart rate and stroke volume (the amount of blood the heart pumps per beat) during the recovery from exercise (see Fig. 3 in Lee et al. 2022), but these effects are driven by very large effects in 1 or 2 studies and most studies only find trivial to small effects.
Wearing compression clothing can have a small effect on increasing venous limb blood flow (blood flow in veins of the calf and thigh) at rest as well as during and in the recovery from exercise (although not immediately after), while the effects on arterial limb blood flow are less clear (see O'Riordan et al. 2023).
Wearing compression clothes after exercise can have small to moderate effects on reducing blood markers of muscle damage (creatine kinase, CK, and lactate dehydrogenase, LDH), lowering feelings of delayed onset muscle soreness (DOMS), and promoting the feeling of recovery (see Hill et al. 2014, Marqués-Jiménez et al. 2016, Engel et al. 2016, Dupuy et al. 2018, and Mota et al. 2020).
While Brown et al. 2017 concluded that compression improves the recovery of muscle strength, their meta-analysis inappropriately pooled diverse outcomes (e.g. recovery of running time trial performance, time-to-exhaustion, strength, power, vertical jump height, etc) and multiple post-exercise time points as a singular marker of recovery, therefore increasing the likelihood of finding a beneficial effect size when there may not be one. When an appropriate analysis is made, the conclusion is that wearing compression garments during or after exercise is unlikely to help recover muscle strength or power (János Négyesi et al. 2022).
Brown et al. 2017 also concluded that compression improves next day cycling performance but the analysis has the same issues as described earlier. Wearing compression garments during or after endurance exercise (e.g. running) is unlikely to improve the recovery of performance, but this needs further investigation.
Compression clothes worn during exercise are unlikely to improve performance, including vertical jump height, V̇O2maxYour maximal rate of oxygen consumption; a measure of cardiorespiratory fitness and maximal aerobic power, which contributes to endurance performance., blood lactate, RPE, etc (see da Silva et al. 2018 and Mota et al. 2020). That said, one systematic review (Engel et al. 2016) assessed the effects of wearing compression clothing specifically during running, finding a trivial but possibly meaningful effect on time-to-exhaustion and running economyThe rate of energy expenditure (measured in kiloJoules [KJ], kilocalories [kcal] or oxygen consumption [V̇O2]) per kilogram body mass (kg) per unit of distance i.e. per 1 kilometer traveled. A runner with a lower energy cost per kilometer has a higher economy than a runner with a higher energy cost.; however, there was no meaningful effect on running time trial performance (during ½-marathon, 15 km trail running, 5/10 km runs, or 400 m sprints; in agreement with da Silva et al. 2018).
But…
Few studies have examined this topic using a randomised controlled design, and the summary effect sizes are often driven by a huge effect size found in a single study. There is also large variability in the study designs and the measured outcomes between the studies included in these systematic reviews, meaning that more standardised methodological approaches are warranted so that more accurate conclusions can be made. Furthermore, the majority of experimental studies have not measured the compression pressure exerted by the garments and simply report the levels indicated by the manufacturer, typically 15 to 35 mmHg; therefore, optimal compression pressures or durations are currently unknown.
Since we have known for many moons that the best way to increase muscle contraction, blood flow, and cardiac output is to move, it is unlikely that the purported role of compression to “improve blood flow” is of any use since no device or tool can simulate the magnitude of change in these physiological parameters that is seen during exercise.
If you are healthy and do not have vascular disease, blood does not pool in your limbs, it flows very well, and the return of venous blood to the heart works just fine, even at rest. Contracting lower limb muscles with gentle walking/cycling/swimming etc or even some light jogging will increase your heart rate and massively increase blood flow in your limbs, elevating venous return — a compression garment will not match the magnitude of that effect.
Compression clothing is not detrimental to post-exercise recovery and, excitingly, appears to lessen feelings of soreness and may facilitate the recovery of strength performance. Since compression clothes often feel like a big hug, that “psychologically soothing” feeling will also be relaxing and restful; two additional ingredients that might get you ready to go again.
Can compression clothing enhance athletic recovery and performance?
Wearing compression clothing after exercise is likely to improve feelings of recovery and muscle soreness after exercise.
The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is small to medium and appears to be similar between trained athletes and untrained folks, and between males and females, but few studies have directly made these comparisons.
Wearing compression clothing after exercise is unlikely to improve the recovery of muscle strength/power or endurance performance.
Wearing compression clothing during exercise is unlikely to improve strength or endurance performance.
Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported., and possible publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
Furthermore, additional research is needed to determine the optimal compression pressure and duration.
The nice part: compression clothing doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
How to use this: INSERT.
Full list of meta-analyses examining compression clothing for recovery.
Here are the meta-analyses I've summarised above:
Do Sports Compression Garments Alter Measures of Peripheral Blood Flow? A Systematic Review with Meta-Analysis. Shane F O'Riordan, David J Bishop, Shona L Halson, James R Broatch. Sports Med. 2023
Wearing Compression Garment Enhances Central Hemodynamics? A Systematic Review and Meta-Analysis. Daniel C W Lee, Ajmol Ali, Sinead Sheridan, Derwin K C Chan, Stephen H S Wong. J Strength Cond Res. 2022
Putting the Squeeze on Compression Garments: Current Evidence and Recommendations for Future Research: A Systematic Scoping Review. Jonathon Weakley, James Broatch, Shane O’Riordan, Matthew Morrison, Nirav Maniar & Shona L. Halson. Sports Med. 2022
Can Compression Garments Reduce the Deleterious Effects of Physical Exercise on Muscle Strength? A Systematic Review and Meta-Analyses. János Négyesi, Tibor Hortobágyi, Jessica Hill, Urs Granacher, Ryoichi Nagatomi. Sports Med. 2022
Effects of Wearing Compression Stockings on Exercise Performance and Associated Indicators: A Systematic Review. Mota GR, Simim MAM, Dos Santos IA, Sasaki JE, Marocolo M. Open Access J Sports Med. 2020
Association of Lower Limb Compression Garments During High-Intensity Exercise with Performance and Physiological Responses: A Systematic Review and Meta-analysis. César Augusto da Silva, Lucas Helal, Roberto Pacheco da Silva, Karlyse Claudino Belli, Daniel Umpierre & Ricardo Stein. Sports Med. 2018
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis.
Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
Compression Garments and Recovery from Exercise: A Meta-Analysis.
Brown F, Gissane C, Howatson G, van Someren K, Pedlar C, Hill J. Sports Med. 2017
Are compression garments effective for the recovery of exercise-induced muscle damage? A systematic review with meta-analysis. Marqués-Jiménez D, Calleja-González J, Arratibel I, Delextrat A, Terrados N. Physiol Behav. 2016
Is There Evidence that Runners can Benefit from Wearing Compression Clothing?
Engel F, Holmberg, H, Sperlich B.
Sports Med. 2016
Compression garments and recovery from exercise-induced muscle damage: a meta-analysis. Hill J, Howatson G, van Someren K, Leeder J, Pedlar C. Br J Sports Med. 2014
What is pneumatic compression?
Pneumatic compression — also known as external/intermittent pneumatic compression (IPC) or external counterpulsation (ECP) — is an extension of compression that was developed to assist venous return in patients with terrible cardiac output (e.g. folks recovering from a stroke, heart attack, or folks with peripheral vascular disease), and to promote shear stress (a stimulus caused by increased blood flow that promotes angiogenesis, the formation of new blood vessels) in patients who are unable to exercise. However, systematic reviews on stroke, angina, and heart failure have found a widespread low quality of evidenceA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials., preventing clear conclusions regarding the effectiveness of ECP for disease rehabilitation.
Pneumatic compression involves a series of pneumatic cuffs placed on the limbs that inflate/deflate sequentially to provide a peristalsis-like stimulus to blood vessels, helping blood return to the heart. I spent some time at the Cleveland Clinic in 2009 collecting pilot data using this tech in patients with diabetes; the feeling of wearing a pneumatic compression device is certainly unique and the patients found it fun, but the pilot project was aborted when the effects on blood glucose control were unimpressive.
Since pneumatic compression can moderately increase limb blood flow, it garners interest from athletes, coaches, and sports teams, and “recovery” centres often have some form of pneumatic compression device ready to massage your veins in a peristalsis-like manner. But, what does the evidence say?
What is the scientific evidence on the effect of pneumatic compression on recovery and athletic performance?
Until 2022 there was no systematic review or meta-analysis on this topic but some research had attempted to determine the effect of pneumatic compression on recovery…
Twenty minutes of pneumatic compression administered between two bouts on the same day caused a smaller decline in 1.2 km shuttle run performance and increased feelings of recovery in recreationally-active men.
Thirty minutes of pneumatic compression following a session better-restored cycling peak power output in 7 National Rugby League players, compared to rest only.
On the contrary, 30-minutes of pneumatic compression treatment following a session of S.H.I.T. (short high-intensity training) had no effect on the restoration of jump height, mean power during an 8-minute time trial, or feelings of fatigue, in elite triathletes or in team sports players.
Similarly, 30-minutes of pneumatic compression treatment following a plyometric session did not affect the restoration of jump height or feelings of fatigue, in recreationally active adults.
And, competitors in the 161-km Western States race who received a post-race 20-minute session of massage combined with pneumatic compression reported lower feelings of fatigue but their recovery of performance was not improved in a 400 m time-trial 3- and 5-days post-race, compared to competitors who simply rested post-race.
This all sounds promising, but we needed a meta-analysis of the evidence. In 2022, we got one…
Wiśniowski and colleagues (2022) Sports Med reported that post-exercise pneumatic compression reduces feelings of muscle soreness (with a small effect size) but with no effect on the restoration of jump height. However, the review identified only 12 studies with just 322 participants and reported a high risk of bias in several studies. In 2024, a meta-analysis of 17 studies including 319 participants (Maia et al. 2024) reported a trivial to moderate beneficial effect of post-exercise pneumatic compression on muscle pain and soreness and a highly variable effect on muscle damage markers. Moreover, protocols of 20 to 30 minutes using a compression pressure of about 80 mmHg seems to be the most used approach. However, high-quality studies are urgently needed to make firm conclusions.
Since we have known for many moons that the best way to increase blood flow and cardiac output is to move, it is unlikely that the purported role of pneumatic compression to “improve blood flow” is of any use to athletes since their vascular function is impeccable and no device or tool can simulate the magnitude of change in blood flow and cardiac output measured during exercise. Consequently, if an athlete wants to increase blood flow between sessions, the best thing they could do is go for a walk.
Currently, no adverse effects of pneumatic compression have been reported so it’s unlikely to harm your recovery. But, far more research is needed to make definitive conclusions.
Can pneumatic compression enhance athletic recovery and performance?
Pneumatic compression is likely to reduce feelings of muscle soreness after exercise.
The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is small.
Pneumatic compression is unlikely to improve performance or the recovery of performance after exercise.
The effects appear to be similar between trained athletes and untrained folks, and between males and females; however, further research is needed in females because they are underrepresented among studies in this field.
Keep in mind: there are very few studies, high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a moderate/high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported., and possible publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: pneumatic compression doesn't appear to hurt recovery. So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
How to use this: INSERT.
Full list of meta-analyses examining pneumatic compression for recovery.
Here are the meta-analyses I've summarised above:
Effects of photobiomodulation, intermittent pneumatic compression and neuromuscular electrical stimulation on muscle recovery: Systematic review with meta-analysis. Canez et al. J Bodyw Mov Ther. 2025
Effects of lower-limb intermittent pneumatic compression on sports recovery: A systematic review and meta-analysis. Maia et al. Biol Sport. (2024)
The Effect of Pressotherapy on Performance and Recovery in the Management of Delayed Onset Muscle Soreness: A Systematic Review and Meta-Analysis. Paweł Wiśniowski, Maciej Cieśliński, Martyna Jarocka, Przemysław Seweryn Kasiak, Bartłomiej Makaruk, Wojciech Pawliczek, Szczepan Wiecha. J Clin Med. 2022
What is massage?
The proposed effects of massage include relieving muscle tension, reducing soreness, and improving joint range of motion. The potential benefits of massage are proposed to derive from an increase in parasympathetic activity, skin and muscle temperature, and blood flow. You can read more about this in a narrative review by Weerapong et al. (2015). There is also a psychophysiological response to massage that helps to enhance mood and reduce feelings of fatigue through relaxation. I.e. massage provides the subjective sense of “going through recovery”. Subjective feelings of recovery (less muscle soreness) using massage have been confirmed in runners recovering from a half-marathon, but objective measures (creatine kinase, testosterone, and muscle function) were not improved.
Everyone can agree that a great massage is truly relaxing and even the simple act of seeing a masseur can be relaxing. But every massage is different… I am guessing you have, at some point, also walked away from a brutal far-from-relaxing “deep-tissue” massage from He-Man, who even informs you, “you’ll be sore for a while” as you hobble away. More soreness and being left feeling rubbish is the precise reason that the only type of massage I have opted for since my last brutal “He-Man” massage in 2003 is a relaxing one, ideally from my wife.
A sports masseur will tell you that a massage will “increase blood flow”, “reduce muscle soreness”, “prevent inflammation”, and “flush out lactate” (and maybe even some other enticing nouns coupled with a direction of change). But, what about the science?
Some work has shown that massages “relax” the autonomic nervous system by triggering the parasympathetic nervous system (see here and here). Some studies show small increases in limb blood flow during a massage but other studies do not, while some studies find that light exercise (hand grip and knee extensions) far more greatly increase blood flow than massage. Some studies find that massage can lower blood lactate and heart rate following exercise, but others find no effect compared to passive rest or even a lower effect compared to light activity (here and here). Massage has been also shown to reduce delayed onset muscle soreness (DOMS) in some studies but not others (see here, here, and here).
So, the effects of massage sound like a mixed bag. Some older narrative reviews (here, here and here) have delved into this area but we should take a more systematic approach than that; especially since my narrative may indeed be cherry-picked to confirm my biases. So…
What is the scientific evidence on the effect of massage on recovery and athletic performance?
Note that this post is not about injury prevention or rehab, it is about recovery from exercise.
Anecdotally, a massage can feel bloody good (in the right hands) and might confer psychological benefits associated with relaxation and the perception of “feeling” more recovered. The evidence shows that post-exercise massage can indeed have a moderate to large effect on lowering feelings of muscle soreness and fatigue (see Guo et al. 2017, Dupuy et al. 2018, Davis et al. 2021, and Dakić et al. 2023).
The effect of post-exercise massage on the recovery of performance is less clear and possibly dependent on the type of exercise and/or type of massage (see Poppendieck et al. 2016 and Davis et al. 2021). However, the most up-to-date meta-analyses (Davis et al. 2021 and Dakić et al. 2023) find that post-exercise massage has trivial and meaningless effects on sprint, jump, strength, and endurance performance. That said, the large variability in study designs and duration and type of massage indicates that more studies are needed to make firm conclusions.
Importantly, massage does not appear to impair the recovery of performance. So, if it feels good, you like it, and you find it relaxing, a post-session massage should do no harm.
An important thing to note is that it is not currently known whether massaging a painful muscle is a good or bad thing. Erring on the side of caution, it is probably wise to remember that causing more pain and soreness should never be a goal of your between-session recovery periods. So, perhaps minimise your visits to the classic “I’m going to focus on your knots and hot spots” type of massage from He-Man; instead, opt for an alternative recovery method, like rest.
Can massage enhance athletic recovery and performance?
Post-exercise massage is likely to improve feelings of muscle soreness and fatigue following intense exercise. The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is medium to large.
However, post-exercise massage is unlikely to improve the recovery of performance following exercise.
These effects appear to be similar between trained athletes and untrained folks, and between males and females; however, such comparisons have not been adequately studied to know for sure.
Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported. (especially because blindingBlinding is when people in a study don’t know which treatment they’re getting. It stops expectations or beliefs (from patients or researchers) from skewing the results. “Single-blind” means participants don’t know; “double-blind” means participants and researchers don’t know; “triple-blind” means that the participants, researchers, and data analysts are kept in the dark. The goal is simple: fair tests and trustworthy findings. is impossible), and publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research. is possible. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: Massage doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm. Furthermore, it’s unclear whether massaging a painful muscle is sensible or not; I would err on the side of caution and skip a massage if it is painful and allow the soreness to recover with good old-fashioned rest.
How to use this: INSERT.
Full list of meta-analyses examining massage for recovery.
Here are the meta-analyses I've summarised above:
The Effects of Massage Therapy on Sport and Exercise Performance: A Systematic Review. Dakić M, Toskić L, Ilić V, Đurić S, Dopsaj M, Šimenko J. Sports (Basel). 2023
Effect of sports massage on performance and recovery: a systematic review and meta-analysis. Davis HL, Alabed S, Chico TJA. BMJ Open Sport Exerc Med. 2020 — Note that this article has a correction (see here).
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis. Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
Massage Alleviates Delayed Onset Muscle Soreness after Strenuous Exercise: A Systematic Review and Meta-Analysis. Guo J, Li L, Gong Y, Zhu R, Xu J, Zou J, Chen X. Front Physiol. 2017
Massage and Performance Recovery: A Meta-Analytical Review. Poppendieck W, Wegmann M, Ferrauti A, Kellmann M, Pfeiffer M, Meyer T.
Sports Med. 2016
Immunological effects of massage after exercise: A systematic review. Tejero-Fernández V, Membrilla-Mesa M, Galiano-Castillo N, Arroyo-Morales M. Phys Ther Sport. 2015
Evidence of the physiotherapeutic interventions used currently after exercise-induced muscle damage: systematic review and meta-analysis. Torres R, Ribeiro F, Alberto Duarte J, Cabri JM. Phys Ther Sport. 2012
Does post-exercise massage treatment reduce delayed onset muscle soreness? A systematic review. Ernst E. Br J Sports Med. 1998
What is foam rolling?
Foam rolling is a self-inflicted myofascial release technique — some folks love it, others hate it. The repetitive rolling over a muscle group can be very soothing and relaxing but, like a massage, it can also be painful… which prompts the obvious thought: “is more pain a sensible goal when I am supposed to be recovering?”.
A huge range of foam rollers and roll-massaging “rolling pins” can be purchased, with varied sizes and surface textures. Proponents for foam rolling will tell you it “triggers mechanoreceptors”, “increases blood flow”, “boosts flexibility”, “releases endorphins”, and “alleviates fatigue”. But, its not currently understood what foam rolling does at the cellular and molecular level. Despite that, as you are probably aware, foam rolling is hugely popular — many runners own a roller, every gym hosts one, and some folks consider them a viable replacement for a masseur. Heck, you can even become “certified” in foam rolling. But, is foam rolling a useful adjunct for assisting your recovery?
What is the scientific evidence on the effect of foam rolling on recovery and athletic performance?
A single bout of foam rolling before a session increases joint range of motion (see Skinner et al. 2020, Hendricks et al. 2019 & Wiewelhove et al. 2019). This effect might be useful if increased flexibility is a key determinant of performance in your sport.
Regular foam rolling may also increase joint range of motion, especially in the knee and particularly with longer interventions (e.g. at least 4-weeks of regular rolling), but more high-quality studies are needed to confirm these effects (see Martínez-Aranda et al. 2024, Konrad et al. 2024, Konrad et al. 2022, Pagaduan et al. 2022 & Skinner et al. 2020).
Foam rolling before a session does not appear to improve performance ( Martínez-Aranda et al. 2024, Glänzel et al. 2022, Skinner et al. 2020, Hendricks et al. 2019, & Wiewelhove et al. 2019). Furthermore, the effect of foam rolling before a session on performance is similar to the effect of stretching before a session (Wilke et al. 2019 & Konrad et al. 2021). However, the type of stretching may influence this comparison. For example, one systematic review (Konrad et al. 2021) found that pre-exercise foam rolling may improve quadriceps strength when compared to pre-exercise static stretching but not dynamic stretching. However, more work is needed to confirm this nuance because this systematic review inappropriately pooled many diverse types of exercise (lifting, jumping, sprinting, etc) into a single performance outcome.
Combining foam rolling with stretching as part of a warm-up before a session has a small effect on increasing range-of-motion compared to doing nothing, but this combined approach has no additional benefit over either foam rolling or stretching alone (see Konrad et al. 2021). Combined foam rolling and stretching before exercise might also increase power (jump height) and sprint performance compared to stretching alone, but perhaps only when foam rolling is followed by stretching (see Konrad et al. 2021). However, there are few studies examining the order of warm-up modalities and this systematic review inappropriately pooled many diverse types of exercise (lifting, jumping, sprinting, etc) into a single performance outcome.
Regular foam rolling for several weeks does not improve performance (see Martínez-Aranda et al. 2024, Konrad et al. 2022, and Pagaduan et al. 2022), but studies have primarily examined muscle strength, power, and vertical jump height; studies on endurance performance are lacking.
While neither acute (1 session) or chronic (regular) foam rolling improves performance, it also does no harm (see Konrad et al. 2022, Pagaduan et al. 2022, Glänzel et al. 2022, Skinner et al. 2020, Hendricks et al. 2019, & Wiewelhove et al. 2019). That said, it is not known whether rolling muscles that are painful is a good or bad thing to do. Siding with caution, it is probably best to remember that causing more pain and soreness should never be a goal of your between-session recovery periods.
Foam rolling after a session might reduce the feelings of muscle soreness (Martínez-Aranda et al. 2024, Zhou et al. 2024, Skinner et al. 2020, Hendricks et al. 2019, & Wiewelhove et al. 2019) but is unlikely to do so after a session that causes exercise-induced muscle damage (Medeiros et al. 2023). However, due to the variability in study designs among studies and an insufficient number of well-designed and high-quality studies, it is currently difficult to make firm conclusions (Medeiros et al. 2023).
The current evidence indicates that approx. 90 seconds of foam rolling per muscle group might be the minimum duration that reduces feelings of muscle soreness (Hughes et al. 2019). But, because of the heterogeneity of methods used between studies, there is currently no consensus on the optimal foam rolling (or roller-massage) approach (e.g., treatment time, pressure, cadence, type of roller, etc.). This lack of consensus makes me wonder how one can become “certified” in foam rolling (aka, myofascial release).
Can foam rolling enhance athletic recovery and performance?
Foam rolling is likely to increase joint range of motion (flexibility). The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is medium to large.
Foam rolling is also likely to reduce feelings of muscle soreness. The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is small.
However, foam rolling is unlikely to improve performance or the recovery of performance after exercise.
These effects appear to be similar between trained athletes and untrained folks, and between males and females.
Keep in mind: There is a lack of consensus on the optimal foam rolling time, pressure, cadence, type of roller, etc. Furthermore, there is moderate heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in effects between studies, and a moderate risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is moderateA moderate quality of evidence means that, in general, studies in this field have some limitations. This could be due to somewhat inconsistent effects between studies, a moderate range of effect sizes between studies, and/or a moderate risk of bias (caused by a small to medium number of studies, small to medium numbers of participants, poorly described randomization processes, some missing data, some inappropriate methods/statistics). When the quality of evidence is moderate, there is some doubt and only moderate confidence in the overall effect of an intervention, and future studies could change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials. at best. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: Foam rolling doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm. Furthermore, it’s unclear whether foam rolling a painful muscle is sensible or not; I would err on the side of caution and skip a massage if it is painful and allow the soreness to recover with good old-fashioned rest.
How to use this: INSERT.
Full list of meta-analyses examining foam rolling for recovery.
Here are the meta-analyses I've summarised above:
Preventive effect of foam rolling on muscle soreness after exercise: A systematic review and meta-analysis. Zhou et al. J Bodyw Mov Ther. (2024)
Static Stretch Training versus Foam Rolling Training Effects on Range of Motion: A Systematic Review and Meta-Analysis. Konrad et al. Sports Med. (2024)
Effects of Self-Myofascial Release on Athletes' Physical Performance: A Systematic Review. Martínez-Aranda et al. J Funct Morphol Kinesiol. (2024)
Acute effects of foam roller or stick massage on indirect markers from exercise-induced muscle damage in healthy individuals: A systematic review and meta-analysis. Flávia Medeiros, Wagner Martins, David Behm, Deise Ribeiro, Emmanuela Marinho, Wanderson Santos, Ricardo Borges Viana. J Bodyw Mov Ther. 2023
Myofascial release strategies and technique recommendations for athletic performance: A systematic review. Maria Elisa Duarte França, Mayane Dos Santos Amorim, Larissa Sinhorim, Gilmar Moraes Santos, Iramar Baptistella do Nascimento. J Bodyw Mov Ther. 2023
The Effects of Foam Rolling Training on Performance Parameters: A Systematic Review and Meta-Analysis including Controlled and Randomized Controlled Trials. Andreas Konrad, Masatoshi Nakamura, David George Behm. Int J Environ Res Public Health. 2022
Foam Rolling Acute Effects on Myofascial Tissue Stiffness and Muscle Strength: A Systematic Review and Meta-Analysis. Marcelo H Glänzel, Deivid R Rodrigues, Gustavo N Petter, Daniel Pozzobon, Marco A Vaz, Jeam M Geremia. J Strength Cond Res. 2022
The effects of foam rolling on ankle dorsiflexion range of motion in healthy adults: A systematic literature review. Rob Grieve, Brendan Byrne, Charlie Clements, Laura-Jayne Davies, Edward Durrant, Oliver Kitchen. J Bodyw Mov Ther. 2022
Foam Rolling Training Effects on Range of Motion: A Systematic Review and Meta-Analysis. Andreas Konrad, Masatoshi Nakamura, Markus Tilp, Olyvia Donti & David G. Behm. Sports Med. 2022
Chronic Effects of Foam Rolling on Flexibility and Performance: A Systematic Review of Randomized Controlled Trials. Jeffrey Cayaban Pagaduan,Sheng-Yuan Chang & Nai-Jen Chang. Int J Environ Res Public Health. 2022
The Accumulated Effects of Foam Rolling Combined with Stretching on Range of Motion and Physical Performance: A Systematic Review and Meta-Analysis. Andreas Konrad, Masatoshi Nakamura, Daniel Bernsteiner, Markus Tilp. J Sports Sci Med. 2021
A Comparison of the Effects of Foam Rolling and Stretching on Physical Performance. A Systematic Review and Meta-Analysis. Andreas Konrad, Markus Tilp, Masatoshi Nakamura. Front Physiol. 2021
A systematic review and meta-analysis of the effects of foam rolling on range of motion, recovery and markers of athletic performance. Skinner B, Moss R, Hammond L. J Bodyw Mov Ther. 2020
Acute Effects of Foam Rolling on Range of Motion in Healthy Adults: A Systematic Review with Multilevel Meta-analysis. Wilke J, Müller AL, Giesche F, Power G, Ahmedi H, Behm DG. Sports Med. 2020
Effects of foam rolling on performance and recovery: A systematic review of the literature to guide practitioners on the use of foam rolling. Hendricks S, Hill H, Hollander SD, Lombard W, Parker R. J Bodyw Mov Ther. 2020
Duration of myofascial rolling for optimal recovery, range of motion, and performance: A systematic review of the literature. Hughes GA, Ramer LM. Int J Sports Phys Ther. 2019
A Meta-Analysis of the Effects of Foam Rolling on Performance and Recovery. Wiewelhove T, Döweling A, Schneider C, Hottenrott L, Meyer T, Kellmann M, Pfeiffer M, Ferrauti A. Front Physiol. 2019
An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis. Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. Front Physiol. 2018
The effects of self-myofascial release using a foam roll or roller massager on joint range of motion, muscle recovery, and performance: A systematic review. Cheatham SW, Kolber MJ, Cain M, Lee M. Int J Sports Phys Ther. 2015
What is electrical muscle stimulation?
Humans have been firing electricity into their bodies for centuries — it is thought that ancient Egyptians, Greeks, and Romans used electric fish to treat all manner of issues. Emperor Palpatine also used electricity to “encourage” his foes to join his side of the force. Today, electrical muscle stimulation (EMS, also known as neuromuscular electrical stimulation [NMES], electromyostimulation, or transcutaneous electrical nerve stimulation [TENS]), which targets afferent nerves to blunt pain), is typically used by physiotherapists for rehabilitation. A common application is during rehab for patients recovering from a stroke, and systematic reviews conclude that electrical stimulation can enhance the recovery of motor control and increase gait speed in stroke patients (see here, here, & here). But, the therapeutic benefit of using electrical stimulation in patients who have poor muscle function should not be assumed to mean that it will benefit recovery or performance in a healthy person. After all, the ability to send nerve impulses to muscle cells is a highly-developed attribute in athletes.
In physiology labs, EMS is widely used as a “test-tube” biology method to study muscle contraction in cell culture experiments — PhD students and postdocs in my lab group extensively used this “exercise in a dish” approach to measure the effect of contraction on glucose metabolism. Electrodes are lowered into a cell culture dish containing muscle cells and the pulses of electrical current can be fired at different voltages, frequencies, and durations to mimic nerve impulses. This type of electrical stimulus can also be achieved in the muscles of live human subjects by attaching electrodes to the skin and applying an electrical current through the skin to a muscle to cause involuntary muscle contractions. Consumer EMS devices like Compex and Powerdot are commonplace in glossy and enticing videos on social media, claiming to “increase blood flow” and “flush out metabolic waste products”. These devices are also inexpensive and widely available. Although not a new technology — vigorous EMS was used in East German athlete training camps in the 1970s — with the increasing popularity and availability of consumer devices (including whole-body EMS systems), it is important to understand what EMS does and whether or not it is useful for the recovery of athletic performance.
We have known for decades that muscle contraction increases blood flow to the muscle and therefore increases both the delivery and removal of nutrients, metabolites, hormones, etc, to and from the muscle. Therefore, the sentiments that muscle contraction can “increase blood flow” and “flush out metabolic waste products” are certainly not inaccurate. Furthermore, using electrical stimulation to innervate afferent nerves to dull pain might, theoretically, reduce feelings of muscle soreness following exercise. But, is EMS useful for recovery, particularly when considering that low-intensity exercise would achieve the same effects — delivery and removal of nutrients, metabolites, and hormones — on multiple muscles, with the added bonus of a light cardiovascular load? A 2011 narrative review, “Does electrical stimulation enhance post-exercise performance recovery?”, concluded that while EMS may enhance post-exercise lactate removal and reduce creatine kinase activity, evidence for restoring (or enhancing) performance is lacking. However, more research has emerged since 2011. And, we can do better than a narrative review. So…
What is the scientific evidence on the effect of electrical muscle stimulation on recovery and athletic performance?
Please note that this summary is specific to recovery and performance, and is not relevant to managing pain or the rehab from injury or illness.
It is also important to note that electrical muscle stimulation is probably safe for most people, but if you have heart problems, epilepsy, wear a pacemaker, or are pregnant, EMS is not advised.
An EMS treatment following a session might help lower blood lactate levels more so than complete rest but not compared to active recovery (moving), which is superior to EMS (see Malone et al. 2014).
An EMS treatment following a session does not promote the recovery of delayed onset muscle soreness (DOMS) (see Menezes et al. 2024 and Menezes et al. 2022), nor does it improve the recovery of performance after exercise (see Malone et al. 2014).
Since muscle wasting is an unwanted symptom of ageing that reduces functional capacity and quality of life, it is encouraging that EMS shows potential for maintaining muscle mass and strength in untrained older-aged folks, including those with sarcopenia (muscle atrophy) or obesity (see de Oliveira et al. 2022 & Kemmler et al. 2022). But, further high-quality randomised controlled trials are needed to confirm these potential effects.
While the effects of strength training and EMS might be comparable in untrained older adults (see Šarabon et al. 2022), there is a lack of high-quality comparative studies. Given the known benefit of resistance training on preserving muscle mass and increasing strength in older adults (see Liu et al. 2009, Borde et al. 2015, and Chen et al. 2021), older adults should always choose strength training over EMS, unless physical limitations or injuries prevent such.
Some evidence also shows promise for EMS in maintaining muscle mass in chronic conditions that force muscular inactivity (e.g. hospitalisation or prolonged leg immobilisation. Therefore, EMS might be useful for preventing muscle atrophy during rehabilitation from an injury that forces you to be inactive. That said, high-quality studies are needed to test that hypothesis.
These effects of EMS on maintaining muscle mass or increasing strength are enticing, but they are in folks who are losing/have lost muscle mass and/or strength and, therefore, do not automatically infer a benefit of EMS in healthy people or trained athletes.
So, what’s the story…?
A 2011 systematic review by Filipovic et al. concluded that a stimulation intensity of at least 50% of a maximal voluntary contraction (MVC) is sufficient to activate strength adaptations, even in trained athletes, and estimated that this could be achieved with 3 EMS sessions per week for 4–6 weeks, using an electrical impulse width in a range of 200–400 microseconds (ms), an impulse frequency of 50–100 hertz (Hz), and an electrical impulse intensity of at least 50 milliamps (mA). A follow-up systematic review by Filipovic et al. additionally concluded that EMS treatment can increase maximal isometric strength, rate of force development, and power, as well as functional movements like jump height and sprint time, even in trained athletes. However, neither review by Filipovic et al. performed a meta-analysis or calculated summary effect sizes or confidence intervals.
More recent systematic reviews with meta-analyses (see Wirtz et al. 2019 and Pano-Rodriguez et al. 2019) conclude that EMS has trivial and non-meaningful effects on strength performance in trained athletes. But, due to the lack of studies, firm conclusions about the effect of EMS training on performance cannot yet be made. A 2022 network meta-analysis (see Micke et al. 2022) concluded that EMS might benefit strength, jump, and sprint performance in trained athletes but that the outcomes are heavily dependent on the protocol, where the most optimal outcomes are found in combined sport-specific training and low-volume high-intensity EMS approaches. However, again, this review noted the lack of high-quality studies.
A 2022 systematic review examined the effects of EMS on endurance performance outcomes finding potential effects on V̇O2maxYour maximal rate of oxygen consumption; a measure of cardiorespiratory fitness and maximal aerobic power, which contributes to endurance performance., running economyThe rate of energy expenditure (measured in kiloJoules [KJ], kilocalories [kcal] or oxygen consumption [V̇O2]) per kilogram body mass (kg) per unit of distance i.e. per 1 kilometer traveled. A runner with a lower energy cost per kilometer has a higher economy than a runner with a higher energy cost., and time trial performance (see da Mota Moreira et al. 2022). But, only 3 of the included studies examined athletes (2 studies of recreational runners and 1 of soccer players) while the remaining studies were of untrained middle to older-aged people. Furthermore, the comparative effect of EMS and endurance training on endurance performance outcomes remains to be tested using a high-quality randomised controlled trial.
In general, the current systematic reviews report a medium to high risk of bias in the experimental studies they have examined. Therefore, high-quality randomised controlled trials are urgently needed in the EMS field.
Furthermore, there is also no consensus on the optimal EMS protocol. For example, a large range of electrical impulse intensities (current), impulse frequencies, impulse durations, and total durations of treatments have been used. More standardisation is needed and dose-response studies are lacking.
Can electrical muscle stimulation enhance athletic recovery and performance?
In sedentary older-aged folks (including those with sarcopenia/muscle loss or obesity), regular electrical muscle stimulation (EMS, NMES, TENS) is likely to help preserve muscle mass and increase strength. The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is medium.
Some evidence also shows that regular electrical muscle stimulation can help preserve muscle mass during periods of forced muscular inactivity and, therefore, it might be a useful tool during the recovery from injury.
Electrical muscle stimulation after exercise is unlikely to improve feelings of muscle soreness or the recovery of performance.
Some evidence shows that transcutaneous electrical nerve stimulation (TENS) applied during endurance exercise can reduce perceived pain and produce small improvements in lab endurance tests. Due to insufficient research, it is unclear whether this findings extends to improved performance in trained athletes.
Due to the lack of research, it is also unclear how these effects compare between trained athletes and untrained folks, and between males and females.
Keep in mind: there is a small number of studies and the studies have a low number of participants. Furthermore, there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported. (especially because blindingBlinding is when people in a study don’t know which treatment they’re getting. It stops expectations or beliefs (from patients or researchers) from skewing the results. “Single-blind” means participants don’t know; “double-blind” means participants and researchers don’t know; “triple-blind” means that the participants, researchers, and data analysts are kept in the dark. The goal is simple: fair tests and trustworthy findings. is impossible), and publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research. is likely. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: electrical muscle stimulation doesn't appear to hurt recovery or performance (unless of course you are causing more muscle damage and consequently delaying recovery). So, if you like it and believe it works for you, give it a whirl. But, remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
How to use this: INSERT.
Full list of meta-analyses examining electrical muscle stimulation for recovery.
Here are the meta-analyses I've summarised above:
Effects of photobiomodulation, intermittent pneumatic compression and neuromuscular electrical stimulation on muscle recovery: Systematic review with meta-analysis. Canez et al. J Bodyw Mov Ther. 2025
Electrical Stimulation and Muscle Strength Gains in Healthy Adults: A Systematic Review. Swarup Mukherjee, Jeryn Ruiwen Fok, Willem van Mechelen. J Strength Cond Res. 2023
Is Electrical Stimulation Effective in Preventing or Treating Delayed-onset Muscle Soreness (DOMS) in Athletes and Untrained Adults? A Systematic Review With Meta-Analysis. Mayara Alves Menezes, Danielle Alves Menezes, Lucas Lima Vasconcelos, Josimari Melo DeSantana. J Pain. 2022
Effects of electromyostimulation on physiological determinants of endurance-performance in healthy subjects: a systematic review. Ivo da Mota Moreira, Anne Krause, Daniel Memmert. J Sports Med Phys Fitness. 2022
Effects of electromyostimulation on performance parameters in sportive and trained athletes: A systematic review and network meta-analysis. Florian Micke, Steffen Held, Jessica Lindenthal, Lars Donath. Eur J Sport Sci. 2022
Efficacy of Whole-Body Electromyostimulation (WB-EMS) on Body Composition and Muscle Strength in Non-athletic Adults. A Systematic Review and Meta-Analysis. Wolfgang Kemmler, Mahdieh Shojaa, James Steele, Joshua Berger, Michael Fröhlich, Daniel Schoene, Simon von Stengel, Heinz Kleinöder, Matthias Kohl. Front Physiol. 2021
Effects of whole-body electromyostimulation on health indicators of older people: Systematic review and meta-analysis of randomized trials. Túlio M D de Oliveira, Diogo C Felício, José E Filho, Diogo S Fonseca, João Luiz Q Durigan, Carla Malaguti. J Bodyw Mov Ther. 2022
Resistance Exercise, Electrical Muscle Stimulation, and Whole-Body Vibration in Older Adults: Systematic Review and Meta-Analysis of Randomized Controlled Trials. Šarabon N, Kozinc Ž, Löfler S, Hofer C. J Clin Med. 2020
Effects of Whole-Body Electromyostimulation on Strength-, Sprint-, and Jump Performance in Moderately Trained Young Adults: A Mini-Meta-Analysis of Five Homogenous RCTs of Our Work Group. Nicolas Wirtz, Ulrike Dörmann, Florian Micke, André Filipovic, Heinz Kleinöder, Lars Donath. Front Physiol. 2019
Effects of whole-body electromyostimulation on health and performance: a systematic review. Alvaro Pano-Rodriguez, Jose Vicente Beltran-Garrido, Vicenç Hernández-González, Joaquim Reverter-Masia. BMC Complement Altern Med. 2019
Neuromuscular electrical stimulation during recovery from exercise: a systematic review. Malone JK, Blake C, Caulfield BM. J Strength Cond Res. 2014
Electromyostimulation--a systematic review of the effects of different electromyostimulation methods on selected strength parameters in trained and elite athletes. Filipovic A, Kleinöder H, Dörmann U, Mester J. J Strength Cond Res. 2012
Electromyostimulation--a systematic review of the influence of training regimens and stimulation parameters on effectiveness in electromyostimulation training of selected strength parameters. Filipovic A, Kleinöder H, Dörmann U, Mester J. J Strength Cond Res. 2011
Share this content with your people:
What is acupuncture?
Acupuncture dates back about 2500 years and is inherent to traditional Chinese medicine. It involves inserting needles into specific “acupoints” of the body and is said to “restore the balance of energy flow in the body”. Acupuncture is claimed to treat many chronic conditions. A stack of Cochrane systematic reviews have dug into this field and, while the evidence doesn’t support many of the claimed benefits, there is reasonable evidence that acupuncture may help relieve migranes (Linde et al. 2016), headaches (Linde et al. 2016), and chronic (long-term) lower-back pain (Mu et al. 2020). But, just pick any chronic condition and you will be able to cherry pick a study showing how acupuncture prevents or treats it. Sounds wonderful. But there are some caveats…
The high prevalenceThe proportion of individuals in a population who have a specific condition or characteristic at a given point in time or over a specified period. of positive effects of acupuncture across the literature raises some concern about the quality of evidence. A systematic review of pain treatment found that 100% of trials (all 36 trials) conducted in China found a statistically significant benefit of acupuncture, while only 68% of trials (171 of 252) conducted worldwide favoured acupuncture over control. Alarm bells are ringing, and this discovery suggests that a publication/reporting bias is present in the literature (i.e. that only positive findings are being published). However, looking through more optimistic lenses, this unusual finding could simply arise because acupuncture is more effective in the country where it is traditionally practised, or that Chinese doctors are simply more skilled acupuncturists. That said, a 2008 review found that the majority of studies examining the effect of acupuncture on pain used study designs with an increased likelihood of a false positive outcome. So, in relation to pain management, study quality in the field of acupuncture must be questioned. For a great read on the general importance of study design, statistics, and “P-hacking”, I can thoroughly recommend the book, “The Art of Statistics”, by David Spiegelhalter.
Besides the generally low quality of evidence, adverse events are also not uncommon. The incidence of an adverse event caused by acupuncture ranges from 6% to 15% in studies conducted outside of China, and less in studies conducted within China. Some adverse events are severe (e.g. death) but rare, while the most common adverse events — local pain, bleeding, and bruising — are mild and attributed to the patient’s mental tension, poor clinical practice, and a lack of sterilization. Naturally, from an athlete’s perspective, additional pain, bleeding, and bruising outside of training and racing might not sound appealing.
With relevance to the recovery from exercise, some studies find that acupuncture reduces exercise-induced delayed-onset muscle soreness (DOMS) while other studies do not (see here and here). Similar dichotomies exist for other aspects of recovery, but now I am just cherry-picking studies — what does the entirety of the literature say?
What is the scientific evidence on the effect of acupuncture on recovery and athletic performance?
Please note that this is a summary of the evidence examining the effects of acupuncture on the recovery from exercise or performance, not the management of pain or rehabilitation from injury or illness.
A 2013 systematic review by Urroz et al. identified 1 study showing that acupuncture performed after exercise decreased heart rate, blood lactate, and oxygen consumption during the 30 to 60-min post-exercise recovery period. However, this review only identified 4 trials in total and concluded that “given the heterogeneity of the interventions and the paucity of robust RCTs, the pooling of effect sizes across studies for meta-analysis was not considered appropriate”. In other words, the authors of the review were unable to create a summary effect size of all studies because of a lack of high-quality randomised controlled trials.
A 2020 systematic review by Chang et al. identified 15 studies and found a trivial to small but meaningless effect of acupuncture on lowering exercise-induced delayed onset muscle soreness (DOMS) — meaningless because the 95% confidence intervalA margin of uncertainty — i.e. a plausible range of values within which the true value would be found 95% of the time if the data was repeatedly collected in different samples of people — if the range crosses zero, there is no effect. of the summary effect sizeAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful the effect is. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. crossed zero.
A 2020 systematic review by Huang et al. found a small to moderate beneficial effects of acupuncture on lowering muscle soreness and creatine kinase (CK, a muscle damage marker), and restoring muscle strength at 72 hours after prior exercise, but no effect at 24- or 48- hours after prior exercise. But there are very few studies examining these effects and the authors also reported a low quality of evidence in all studies included in their meta-analysis.
To summarise these two systematic reviews, acupuncture after exercise might alleviate muscle soreness and help restore muscle strength, but only when the next bout is 72 hours later. But, both reviews noted that there are very few studies and many methodological limitations, including inadequately powered randomised controlled trials (i.e. not enough subjects), a lack of thorough and/or standardized reporting of methods, a lack of appropriate placebo controls, and a lack of blinding of either the participants or the investigators. Furthermore, the exact acupuncture site on the body is often poorly described.
Also, since acupuncture is not without risk — rare adverse outcomes can include hospitalisation and death while more-common side effects include pain, bleeding, and bruising — as an athlete, you must consider whether such risks are worth it. Furthermore, pain, bleeding, and bruising are things that delay recovery not accelerate it. So, again, as an athlete, you must consider, is acupuncture worth it?
If you can handle the increased risk of adverse events during your recovery and you do not experience pain, bleeding, or bruising from your practitioner, acupuncture does not appear to harm the recovery of performance and might slightly lessen your feelings of muscle soreness. But, always remember that your recovery time is not supposed to cause more stimulus and further disruption to homeostasis than your training sessions already do — recovery is about restoring homeostasis, not smashing it.
Can acupuncture enhance athletic recovery and performance?
Acupuncture after exercise is unlikely to improve feelings of muscle soreness or the recovery of performance after exercise.
Due to insufficient research, it is unclear how effects compare between trained athletes and untrained folks, and between males and females.
Keep in mind: there are very few studies, high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported., and publication biasPublication bias in meta-analysis occurs when studies with significant results are more likely to be published than those with non-significant findings, leading to distorted conclusions. This bias can inflate effect sizes and misrepresent the true effectiveness of interventions, making it crucial to identify and correct for it in research. is likely. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment..
The nice part: acupuncture doesn't appear to hurt recovery or performance. So, if you like it and believe it works for you, give it a whirl. That said, if an acupuncture treatment causes pain, bleeding, or bruising, this will delay recovery not promote it. Always remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
How to use this: INSERT.
Full list of meta-analyses examining acupuncture for recovery.
Here are the meta-analyses I've summarised above:
Effects of Acupuncture on Delayed-Onset Muscle Soreness: A Systematic Review and Meta-Analysis. Chang WD, Chang NJ, Lin HY, Wu JH. Evid Based Complement Alternat Med. 2020
Does Acupuncture Benefit Delayed-Onset Muscle Soreness After Strenuous Exercise? A Systematic Review and Meta-Analysis. Huang C, Wang Z, Xu X, Hu S, Zhu R, Chen X. Front Physiol. 2020
Effect of acute acupuncture treatment on exercise performance and postexercise recovery: a systematic review. Urroz P, Colagiuri B, Smith CA, Cheema BS. J Altern Complement Med. 2013
What is cupping?
Cupping originates from alternative medicine and uses a glass or bamboo cup to create suction on the skin over the area of interest. There are two types of cupping: “dry cupping” pulls the skin into the cup without drawing blood, while “wet cupping” lacerates the skin so that blood is drawn into the cup. Either way, the result is a hematoma (a collection of blood under the skin), which causes a bruise.
Cupping is typically used by acupuncturists, and other therapists, to treat pain. However, while meta-analyses find that cupping might be effective for treating chronic (long-lasting) pain — lower back pain, neck pain, etc — there is no standardization in cupping protocols between studies and there is generally a low quality of evidenceA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomisation processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomised controlled trials. (see Shen et al. 2022, Wood et al. 2020, Cramer et al. 2020, & Kim et al. 2011). Therefore, further high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after exposure to the treatment and control. are needed to increase confidence in the pain-relieving effect of cupping.
You’ve probably seen famous athletes sporting their cupping “scars” — cupping was popularised at the 2016 Olympics when swimmers looked like they’d endured a brutal session with a massive octopus. Anecdotally, I’ve heard a couple of folks say they “feel” more recovered when they undergo cupping. But, was their “feeling” of recovery framed by seeing their sporting heroes donning their cupping scars? Perhaps. So...
What is the scientific evidence on the effect of cupping on recovery and athletic performance?
There is currently no systematic review or meta-analysis examining the effects of cupping on recovery from exercise, or performance. However, there are some experimental trials:
— Studies examining hamstring flexibility have found a benefit of cupping, but only when there is no control group (Warren et al. 2020 
) and not when cupping is compared to a control or sham condition (Williams et al. 2019 & Schafer et al. 2020). Meanwhile, another study (Hammons et al. 2022) found an improved range of motion in the gastrocnemius (calf) muscle following muscle-damaging exercise, but no benefit to delayed onset muscle soreness (DOMS).
— One study (Li et al. 2022) found that cupping might reduce ultrasound-measured muscle stiffness, but this was at rest and not in response to exercise.
— Another study (Ekrami et al. 2021) found that pre-exercise cupping might slightly blunt exercise-induced increases in circulating cytokines (markers of inflammation), but whether this has relevance to recovery or long-term training adaptations remains to be investigated.
Consequently, there is insufficient evidence to make a firm conclusion about the effect of cupping on recovery (this means we cannot know whether it is beneficial or harmful). That said, we do know that an adverse effect of each cup site is bruising (due to pooling of blood under the skin, aka a hematoma). This is damage that needs time and energy (i.e. ATP) to repair, thus delaying recovery rather than promoting it. In addition to bruising, other less common adverse effects of cupping include pain and burns (see Mohamed et al. 2023)
While cup marks might look cool, there are too many unknowns to risk using this approach at the expense of your recovery or performance — for now, perhaps choose a different route in your quest for coolness. There are also several other things you could better spend your time to optimise your recovery — sleeping, eating, and resting.
Despite the lack of a systematic review examining recovery or performance, there is a systematic review of the effects of cupping on musculoskeletal rehab (see Mohamed et al. 2023), which concluded that “the evidence of cupping on increasing soft tissue flexibility is moderate, decreasing low back pain or cervical pain is low to moderate, and treating other musculoskeletal conditions is very low to low” and that cupping “might be used as a useful intervention because it decreases the pain level and improves blood flow to the affected area”. Overall, Mohamed et al. found that the quality of evidence was quite poor and that there are very few studies, some with no control group and others with sham control groups or a range of other treatments (including heat, physiotherapy, electrical stimulation, stretching, or acetaminophen).
Can cupping enhance athletic recovery and performance?
Cupping might help lower chronic (long-lasting) pain, but this effect is based on a low quality of evidenceA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomisation processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomised controlled trials. due to a high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies and a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported..
Due to insufficient research, the effect of cupping on recovery and performance is unclear.
Similarly, it is unclear how effects compare between trained athletes and untrained folks, and between males and females.
High-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are urgently needed to have any certainty (confidence) in the claimed effects of cupping.
Due to the lack of research, it is also unknown whether cupping has a detrimental effect on recovery. That said, cupping can cause pain and burns, and each cup site causes a bruise, which is damage that needs time to repair — theoretically, this is delaying recovery rather than helping it. Always remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
How to use this: INSERT.
Full list of meta-analyses examining cupping for recovery.
Here are the meta-analyses I've summarised above:
Evidence-based and adverse-effects analyses of cupping therapy in musculoskeletal and sports rehabilitation: A systematic and evidence-based review. Ayman A Mohamed, Xueyan Zhang, Yih-Kuen Jan. J Back Musculoskelet Rehabil. 2023
The information I provide is not medical advice. NSAIDs and painkillers are drugs. They can have side effects and interact with other drugs, nutrients, and medical conditions (see info on aspirin, ibuprofen, naproxen, and paracetamol/acetaminophen). Always consult your doctor to ensure that a specific NSAID or painkiller is safe for you to use.
What are NSAIDs and painkillers?
The symptoms of inflammation include pain/soreness, swelling, redness, heat, and the loss of function in an organ or region of the body. Beneath the surface, inflammation is a complex series of events involving your immune system and the release of proinflammatory and anti-inflammatory molecules. “Acute” inflammation is triggered infections and injuries; it’s short-lived and transient, and is typically overcome within hours or days. On the other hand, “chronic” inflammation is long-lasting and persistent and is a characteristic of several chronic conditions, like arthritis, obesity, etc, and can be a causal pathogenic factor in other conditions, like diabetes and heart disease.
Inflammation sounds pretty terrible and, naturally, many folks surmise that if inflammatory molecules are flying around the body then that cannot be a good thing. But, consider this fact: exercise triggers an acute inflammatory response. Since every work-out you do causes an acute inflammatory response, which is detectable as a feeling of muscle soreness and measurable in the blood and muscles as an increase in proinflammatory cytokines, it might sound intuitive that taking an anti-inflammatory drug is an “essential” part of being an athlete — to help alleviate post-exercise feelings of muscle soreness and help “speed up” recovery.
Nonsteroidal anti-inflammatory drugs (NSAIDs) are a very popular type of drug, most commonly taken for their pain-killing benefits. Let’s be honest, who hasn’t thought, “I need a painkiller” when they have a headache? NSAIDs include common generic/brand-name drugs like aspirin, ibuprofen (e.g. Advil/Motrin/Nurofen), naproxen (e.g. Aleve), etc etc. Analgesic (painkilling) and antipyretic (fever-lowering) drugs like paracetamol (aka acetaminophen, e.g. Panadol/Tylenol) also sometimes get stuffed under the NSAID umbrella because they have similar effects. All of these drugs can be self-prescribed, are cheap to purchase, and are easily available over-the-counter with no questions asked.
NSAIDs work by rapidly-inhibiting cyclooxygenase (COX) enzymes, whose main function is to synthesise prostaglandins, which are a group of hormone-like molecules involved in inflammation that trigger fever and swelling as the associated feelings of pain/soreness. Aspirin, ibuprofen, and naproxen are “nonselective”, which means they inhibit all forms of COX enzyme, COX1 and COX2. Inhibiting COX1, which is expressed in all tissues, reduces the levels of prostaglandins. COX2 is predominantly found in gastrointestinal and kidney cells, inhibition of which increases the risk of stomach and kidney issues that are associated with NSAIDs. Paracetamol (acetaminophen) works via a different mechanism, which is less clearly understood — although it might inhibit COX enzyme activity, paracetamol does not reduce tissue inflammation, while its painkilling effects likely manifest via effects on serotonergic- and endocannabinoid-signalling. Paracetamol (acetaminophen) tends not to have the side effects associated withaspirin or ibuprofen and, while it is safe at recommended doses, it can be toxic to the liver, especially when mixed with alcohol.
So, using NSAIDs to blunt or prevent exercise-induced muscle soreness certainly has some rationale. This is why many folks use them. A 2018 survey of recreational endurance athletes found that 68% (2 in 3) had used NSAIDs (predominantly ibuprofen) in the past 12-months, 45% (1 in 2) immediately before or after a session. The majority of people surveyed expected anti-inflammatory and/or painkilling benefits. This was despite more than half of the respondents expecting an adverse effect on their liver or kidney. Interestingly, only 26% (1 in 4) used an NSAID under the guidance of a doctor or pharmacist. This high prevalenceThe proportion of individuals in a population who have a specific condition or characteristic at a given point in time or over a specified period. of NSAID use has also been reported in Ironman triathletes, Ultrarunners, and world-class footballers. In 2020, a survey of 806 recreational “Parkrunners” (a weekly 5km event in the UK), found that 88% (9 in 10) had used an NSAID (mostly ibuprofen) in the past 12-months and that, alarmingly, 1 in 4 had experienced an adverse drug reaction (usually gastrointestinal), and 1 in 5 of the NSAID users had a pre‐existing contraindication yet still used NSAIDs. The study also found that the longer the race, the more likely a person was to take an NSAID during the event. This finding was also true in Ironman triathletes and Ultrarunners (who reported taking NSAIDs during a race to prevent pain and fatigue). To summarise these athlete surveys, the prevalenceThe proportion of individuals in a population who have a specific condition or characteristic at a given point in time or over a specified period. of NSAID use is high, ibuprofen is the most common, knowledge of effects (and side-effects) is poor, clinical advice is rarely sought, and their use is justified to treat an injury or reduce during-exercise and/or post-exercise feelings of soreness and pain.
Now, here comes the crunch… People who reason their journey to using an NSAID as a core component of their training, fail to acknowledge that the acute inflammatory response to each work-out is what drives muscle adaptations to training (you can read all about that in my post at veohtu.com/muscleadaptations). So, it’s reasonable to hypothesise that regular use of an NSAID alongside your training might actually blunt recovery and training adaptations. There’s even some experimental evidence to support this idea. For example:
— In cell culture experiments, prostaglandins — the group of molecules that trigger fever, swelling, and associated pain — can stimulate muscle cell growth (see Markworth et al. 2011).
— In humans COX enzymes are certainly involved in muscle protein metabolism (see Trappe et al. 2011), as is the production of prostaglandins in muscle following exercise (see Trappe et al. 2001).
— Further work has found that a single over-the-counter dose of either ibuprofen or paracetamol (acetaminophen) may prevent resistance exercise-induced increases in muscle protein synthesis in young recreationally-active men, without suppressing muscle soreness or creatine kinase (a marker of muscle damage; see Trappe et al. 2002) and without blunting exercise-induced increases in muscle inflammation (see Peterson et al. 2003).
— Longer-term studies have shown that chronic COX-inhibitor consumption attenuates muscle growth (see Soltow et al. 2006 & Novak et al. 2009) and regrowth from inactivity-induced atrophy, in rodents (see Bondesen et al. 2006). While in humans, high doses of ibuprofen (but not aspirin) have been found to blunt training-induced increases in muscle mass (aka hypertrophy) and strength in younger previously-inactive adults (see Lilja et al. 2017), but lower doses of ibuprofen or paracetamol (acetaminophen) have not been shown to inhibit training-induced hypertrophy or strength gains in older previously-inactive men (see Trappe et al. 2011).
To summarise these data, it appears that using NSAIDs or paracetamol might not mitigate exercise-induced inflammation or feelings of soreness but might impair training-induced increases in muscle protein synthesis, hypertrophy, and strength, at least in younger adults.
But, what about the effect of NSAIDs on severe muscle damage? One study has shown that muscle damage was reduced with daily diclofenac sodium (e.g. Voltaren) administration for 2 weeks before and during 2-weeks of daily eccentric stair-stepping training, in young untrained adults (see O’Grady et al. 2000). A similar outcome was found where daily ibuprofen administration for 14-days before electrical stimulation-induced muscle damage of the quadriceps improved muscle repair despite a lack of effect on feelings of soreness or the restoration of strength or power (see Mackey et al. 2016). So, NSAIDs show potential for supporting recovery following severe muscle damage, but these are just two small studies.
That was a somewhat lengthy introductory narrative but it’s vastly important to understand the risks and the benefits of drugs that are nonchalantly quaffed down by some athletes. As is probably becoming clear, the topic of NSAIDs and exercise is quite bewildering! Several experimental studies and narrative reviews have tried to unravel the complexity of the topic, and I have included them in this post since they provide useful insight. But, to avoid a potentially epic cherry-picking expedition, as always, let’s dig into the systematic reviews. So…
What is the scientific evidence on the effect of NSAIDs and painkillers on recovery and athletic performance?
First off, always remember that NSAIDs and painkillers are drugs. As with all drugs, there are side effects, drug-drug interactions, and food/alcohol-drug interactions, the risks of which are increased in people with certain conditions. Widespread availability can easily lead to misuse and even abuse. Never self-prescribe a drug without medical advice and if you are unsure, always consult your doctor.
Secondly, using certain drugs violates the rules of sport. NSAIDs, like aspirin, ibuprofen, and paracetamol (acetaminophen), are not currently prohibited for in-competition or out-of-competition use by WADA, but such rules can change so always consult WADA’s prohibited list for up-to-date info, and always cross-check your meds against the Global DRO drug reference list
Thirdly, always remember that every work-out triggers an acute (short-lived, transient) inflammatory response — this is normal, not pathological. Yes, combining the accumulation of acute inflammation with impaired anti-inflammatory defences will develop into chronic inflammation and tissue dysfunction and, possibly, disease. But, the acute increase in inflammatory molecules following a workout is a powerful mediator that drives your training adaptations. Therefore, in the context of exercise, acute inflammation is your friend. And, your body’s natural anti-inflammatory response prevents acute inflammation from having a rager and developing into chronic inflammation.
So, armed with that knowledge, does it sound like you’re supposed to be popping NSAIDs after every session? Let’s find out…
But, what about recovery? Let’s see…
And, what about performance…
How to use this: INSERT. Here are the meta-analyses I've summarised above:
Several narrative reviews highlight that the numerous potential adverse effects of NSAIDs should not be ignored (see Pedersen et al. 2022, Ziltener et al. 2010, Lundberg et al. 2018, & Warden et al. 2010). Furthermore, an analysis of 18,820 patients found that NSAIDs accounted for 30% of hospitalisations caused by adverse drug reactions (see Pirmohamed et al. 2004). Additionally, some evidence shows that chronic NSAID use may affect bone growth, potentially delaying the healing of stress fractures, which is not surprising since prostaglandins — the molecules produced by the COX enzymes that are inhibited by NSAIDs — are required for bone metabolism (see Ziltener et al. 2010, Vuolteenahoet al. 2008). Plus, NSAID use during exercise has been associated with hyponatremia, reduced renal blood flow, and gastrointestinal bleeding (see Pedersen et al. 2022, Lundberg et al. 2018, & Warden et al. 2010), and taking NSAIDs before exercise can increase core temperature in some cases (see Emerson et al. 2020). Incidentally, this is among the reasons UTMB banned NSAIDs from in-competition use, which is sensible given the current lack of knowledge about the health effects of repeated NSAID use in athletes during exercise, especially prolonged exercise like ultras and multi-day stage races.
So, yes there are risks and downsides, but this is not to say that NSAIDs and painkillers are useless — they have their place for dealing with acute pain. Sports medicine physicians often use NSAIDs to treat pain in athletes (Ziltener et al. 2010) but due to their negative consequences on the long-term healing process, the duration of use is typically short (a few days), and the specific type of injury and levels of dysfunction and pain are always be considered. If you have injury-related persistent pain or training-related persistent muscle soreness, it’s unwise to use NSAIDs and painkillers long-term every day because doing so may impair healing and repair (following injury/damage) and/or adaptation and growth (during training). Therefore, self-prescribed treatment of pain with over-the-counter drugs to accelerate the return to training is unwise — if you have persistent pain, always consult a medical doctor and a qualified sports physiotherapist.
A meta-analysis by Morelli et al. 2017 found a small beneficial effect of NSAIDs on blunting muscle soreness, creatine kinase (a marker of muscle damage), and the loss of muscle strength following muscle-damaging exercise. But their analysis included both human and animal studies, all types of NSAID, and both pre- and post-exercise NSAID treatment, and found large variability between studies. Furthermore, the findings were restricted to ibuprofen (e.g. Advil/Motrin/Nurofen), naproxen (e.g. Aleve), and diclofenac sodium (e.g. Voltaren); there was no data included on aspirin or paracetamol (aka acetaminophen, e.g. Panadol/Tylenol). Therefore, further high-quality randomised controlled trials in humans are needed to bolster the current evidence.
In healthy people, taking an NSAID before or during exercise has neither a beneficial nor detrimental effect on performance, including muscle strength, power, time-trial, and time-to-exhaustion (see Cornu et al. 2020). However, because of the diversity of studies, the analysis by Cornu et al. pooled all types of outcomes into “performance” and more studies are needed to understand the specific effects of specific NSAIDs on specific types of exercise. That said, one meta-analysis by Grgic et al. 2021 found a trivial performance-enhancing effect of paracetamol on time-to-exhaustion endurance tests when ingested 45 to 60 min before exercise. However, only a few studies on less than 20 participants have been conducted to date and this meta-analysis, and the one by Cornu et al., included a diverse bunch of participants: untrained, trained, athletes — there is currently very little data in athletes. Furthermore, Grgic et al. 2021 found no beneficial effect of paracetamol on time trial performance. This lack of benefit on running time trial (and race) performance was recently confirmed in two randomised controlled trials (see de Souza et al. 2022 & Huffman et al. 2022).
In young healthy adults, the regular use of non-selective COX-inhibitors (e.g. NSAIDs like aspirin, ibuprofen, & naproxen) does not enhance performance (see Cornu et al. 2020 & Alturki et al. 2018), but the quality of evidence is low and further high-quality randomised controlled trials are urgently needed to make firm conclusions.
Where the story differs is in older-aged hospitalised people with acute inflammatory conditions, in whom a regular prescription of anti-inflammatory drugs (both non-selective COX inhibitors and selective COX-2 inhibitors) can enhance performance and muscle weakness (see Alturki et al. 2018). Furthermore, in older-aged people, regular use of non-selective Cox inhibitors (e.g. NSAIDs like aspirin, ibuprofen, & naproxen) may also enhance training-induced increases in muscle strength (see Alturki et al. 2018). This benefit in older folks is likely to be due to the lowering of ageing-related chronic inflammation, but the precise mechanisms are currently not known and while the quality of evidence is good, there are few studies, and additional high-quality randomised controlled trials would bolster the evidence.
Can NSAIDs and painkillers enhance athletic recovery and performance? Taking a nonsteroidal anti-inflammatory drug (NSAID) after muscle-damaging exercise is likely to lower feelings of muscle soreness. The effect sizeA standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. is small. NSAIDs and painkillers are unlikely to improve performance in healthy people. People with chronic pain or chronic inflammatory conditions should consult their doctor for information on exercise. Due to the lack of direct comparisons, it is unclear how effects compare between trained athletes and untrained folks, and between males and females. Further research is needed in females because they are underrepresented among studies in this field. Keep in mind: there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in study designs and effects between studies, and a moderate to high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the published results accurately reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, precision, and publication bias. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomization processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and future studies could easily change overall conclusions. The best way to improve the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomized controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after the exposure to treatment/control. are needed to increase the certainty (confidence) in the overall effect sizes reported in meta-analysesA meta-analysis quantifies the overall effect size of a treatment by compiling effect sizes from all known studies of that treatment.. Also keep in mind that a regular regime of taking NSAIDs or painkillers to mask pain and blunt muscle soreness may have a detrimental effect on training adaptations, recovery from injury, and muscle growth. Therefore, popping such drugs every day so you “feel” more recovered after every session is unwise — acute exercise-induced inflammation is an athlete’s friend. Consequently, the widespread regular use of NSAIDs and painkillers by athletes is not justified by the current evidence. That said, taking the occasional anti-inflammatory or painkilling drug is unlikely to be a problem. Always remember that time and money spent doing your recovery with something that has no obvious benefit might be better spent sitting down, resting, eating something nutritious, and doing something calm.
The information I provide is not medical advice. NSAIDs and painkillers are drugs and they can have side effects and interact with other drugs, nutrients, and medical conditions (see info on aspirin, ibuprofen, naproxen, and paracetamol/acetaminophen). Always consult your doctor to ensure that a specific NSAID or painkiller is safe for you to use.Full list of meta-analyses examining NSAIDs and painkillers for recovery.
Prevalence, frequency, adverse events, and reasons for analgesic use in youth athletes: A systematic review and meta-analysis of 44,381 athletes. Julie Rønne Pedersen, Alessandro Andreucci, Jonas Bloch Thorlund, Merete Møller, Louise Kamuk Storm, Alessio Bricca. J Sci Med Sport. 2022
Effects of Paracetamol (Acetaminophen) Ingestion on Endurance Performance: A Systematic Review and Meta-Analysis. Jozo Grgic, Pavle Mikulic. Sports. 2021
Non-steroidal anti-inflammatory drugs on core body temperature during exercise: A systematic review. Emerson DM, Chen SC, Kelly MR, Parnell B, Torres-McGehee TM. J Exerc Sci Fit. 2021
Effect of Non-Steroidal Anti-Inflammatory Drugs on Sport Performance Indices in Healthy People: a Meta-Analysis of Randomized Controlled Trials. Cornu C, Grange C, Regalin A, Munier J, Ounissi S, Reynaud N, Kassai-Koupai B, Sallet P, Nony P. Sports Med Open. 2020
Impact of drugs with anti-inflammatory effects on skeletal muscle and inflammation: A systematic literature review. Alturki M, Beyer I, Mets T, Bautmans I. Exp Gerontol. 2018
Effect of NSAIDs on Recovery From Acute Skeletal Muscle Injury. A Systematic Review and Meta-analysis. Kimberly M. Morelli, Laura B. Brown, Gordon L. Warren. Am J Sports Med. 2017
And, here is the list of narrative reviews I summarised above:
List ordered newest to oldest.
What is the Effect of Paracetamol (Acetaminophen) Ingestion on Exercise Performance? Current Findings and Future Research Directions. Jozo Grgic. Sports Med. 2022 → see a letter to the editor from Holgado et al. 2022 and the reply from Grgic et al. 2022.
Analgesic and anti-inflammatory drugs in sports: Implications for exercise performance and training adaptations. Lundberg TR, Howatson G. Scand J Med Sci Sports. 2018
Effects of prostaglandins and COX-inhibiting drugs on skeletal muscle adaptations to exercise. Trappe TA, Liu SZ. J Appl Physiol. 2013
The use of nonsteroidal anti-inflammatory drugs for exercise-induced muscle damage: implications for skeletal muscle development. Schoenfeld BJ. Sports Medicine. 2012
Non-steroidal anti-inflammatory drugs for athletes: An update. Ziltenera JL, Leala S, Fournier PE. Ann Phys Rehabil Med. 2010
Prophylactic Use of NSAIDs by Athletes: A Risk/Benefit Assessment. Warden SJ. Phys Sportsmed. 2010 → “Scientific evidence for such benefits is sparse, and athlete rationale for using prophylactic NSAIDs for their preemptive analgesic and anti-inflammatory effects appears at odds with current understanding of the underlying pathology of many sports related injuries”.
Non-steroidal anti-inflammatory drugs, cyclooxygenase-2 and the bone healing process. Vuolteenaho K, Moilanen T, Moilanen E. Basic Clin Pharmacol Toxicol. 2008
Nonsteroidal Antiinflammatory Drugs in Tendinopathy: Friend or Foe. Magra M, Maffulli N. Clin J Sports Med. 2006
Consult WADA’s prohibited list.
Cross-check your meds against the Global DRO drug reference list.
Only choose supplements that have been independently tested by Informed Sport or LabDoor.
You are the only person responsible for what goes in your body.
Ignorance is not an excuse!
Stay educated. Be informed.
What is the placebo effect?
In the 90s, I moshed my socks off to Brian Molko’s melodic voice at many a Placebo gig. He taught me one thing: placebos rock! Dad jokes aside, a placebo is sometimes used in clinical trials and usually takes the form of a sugar pill, a saline (salt water) injection (like saline), or sham (fake) surgery (where the surgeon opens you up, has some lunch, stitches you back up, and pretends they did a procedure on you). Theoretically, a placebo is supposed to have no effect on whatever variable of interest you are studying.
In a randomised controlled trial, there is a treatment group and a control group. Sometimes, the control group is given something that resembles the treatment to prevent participants from knowing which group they are in — this something is called a “placebo” and we’d call this type of trial a placebo-controlled clinical trial. If the variable of interest improves in the placebo-controlled group we might think that the placebo has had an effect. But, to make that conclusion, we would need an additional non-placebo control group to know whether the placebo has improved the variable of interest more than doing nothing at all — it could be that the disease naturally improves with time. When studies have a placebo-controlled group and a non-placebo control group, if the change in the variable of interest in the placebo group is larger than the change in the non-placebo group, we can conclude that there has been a placebo effect (and Brian Molko can sing about it). On the flip side, if the variable of interest worsens in the placebo-controlled group (and more so than in the non-placebo group), we conclude that there has been a nocebo effect (i.e. a negative or detrimental placebo effect).
Imagine this scenario:
You have an important race today and your coach gives you a supplement to take that they know is not proven to enhance performance. You don’t know what it is or what its effect is supposed to be but, because you want to win and you trust your coach, you take it. During the race, you feel like a monster — strong, fast, and highly motivated. You win! In your post-mortem of the race, you think, without a doubt, that the supplement made you win. Your mother says it was a coincidence. Your coach says you had trained for months, had been setting PBs, and were well-prepared and ready for a great race — you were always going to win.
Who is right… You? Your mother (who is usually right)? Or your coach?
If you are right, then this supplement could be called a placebo and your race win could be called a placebo effect. But, because you had a positive expectation of a performance-enhancing effect, the effect of the placebo could also be called a belief effect. But, a positive expectation of an effect is not always required for a placebo effect and having a positive expectation of an effect can even enhance the placebo effect. The important thing to learn is that while your experiences shape your beliefs, your experiences (i.e. you won after taking a supplement) must be combined with data (your training had been great, you’d set PBs all season, and were in the shape of your life) to accurately inform your belief. The other important thing to learn is that “you are the only person responsible for what goes in your body” — never take candy from a stranger your coach without knowing what it is, what it does, and whether it has been independently tested.
Now, that was just hypothetical — you’re probably wondering, does the placebo effect exist?
Several systematic reviews have found evidence for the placebo effect in a range of diseases in studies where participants are blinded to whether they are receiving a placebo or the real treatment. A couple of recent systematic reviews (see here & here) also find evidence for the placebo effect on a range of diseases when an “open-label” placebo (participants know they’re receiving a placebo) is compared to a non-placebo (no treatment) intervention, with summary effect sizesAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful that effect is. Common effect size measures include Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. of 0.88 (a large effect; 95% confidence interval (CI)A measure of uncertainty used in Frequentist statistics. The 95% confidence interval is a plausible range of values within which the true value (e.g., the true treatment effect) would be found 95% of the time if the data was repeatedly collected in different samples of people. If this range of values (the confidence interval) crosses zero, there is little confidence in the effect. 0.62 to 1.14) and 0.72 (a moderate effect; 95%CI 0.39 to 1.05).
Woah!
But… These findings should be interpreted with caution because, in longitudinal studies (when the variable of interest is measured before and after several days/weeks/months/years), as I said earlier, the disease might naturally improve with time without real treatment. (Putting that in the context of exercise, your performance might improve while training with a supplement but it might also improve with training alone.) Furthermore, most studies do not have non-placebo control group. Plus, there are only a few small “open label” trials and participants weren’t blinded to the placebo vs. no treatment groups and, in many cases, participants received “positive messages” in addition to the open-label placebos before being asked about their feelings of pain or discomfort. E.g. “This will make you feel better.” ... … “How are you feeling?”.
Sometimes doctors prescribe placebos (without the patient’s knowledge) instead of therapeutic drugs to help “treat” a condition. As is probably obvious, from an ethical stance, placebos are a touchy subject — patients have a right to know what they’re receiving rather than being “deceived”. Furthermore, since the “belief effect” (the extent to which a person believes the treatment will work) can also influence the effectiveness of interventions, it could be argued that instead of deceptively-prescribing placebos (in place of therapeutic drugs), doctors could use the placebo effect and/or the belief effect to enhance the therapeutic effect of a drug — if a patient believes the treatment will help them, perhaps they’ll be more motivated to adhere to it. But, I’m going a little off-piste and away from exercise performance.
In research, it is very difficult for scientists to get “deception” studies approved by ethical review boards. But there are some fun examples. In a 2015 study from Ramzy et al., trained 10 km runners received no treatment or daily injections of “OxyRBX”, which they were told had the same effects as EPO (a red blood cell-boosting hormone that improves endurance performance). Before and after 7-days of treatment/no treatment, runners completed a 3 km race. On average, athletes in the OxyRBX group improved their 3 km time by 9.73 seconds (95% CIA measure of uncertainty used in Frequentist statistics. The 95% confidence interval is a plausible range of values within which the true value (e.g., the true treatment effect) would be found 95% of the time if the data was repeatedly collected in different samples of people. If this range of values (the confidence interval) crosses zero, there is little confidence in the effect. 5.14 to 14.33 seconds) whereas untreated athletes ran 1.82 s faster (95%CI 2.77 s slower to 6.41 s faster; group-change comparison P=0.02A p-value is a statistical measure that indicates the probability that the result is at least as extreme than that observed if the null-hypothesis was true. If P is small, the observed difference is big enough to disprove (reject) the null hypothesis. In very basic terms, P = probability that the effect could be explained by random chance and a P-value of less than 0.05 means the results look so promising that there’s only a 1-in-20 (or 5%) chance that they would have occurred if the treatment had no effect at all. Common thresholds for significance are 0.05, 0.01, and 0.001.). OxyRBX-treated runners also reported less physical effort, increased motivation, and improved recovery. This is an example of a placebo effect, probably caused by a belief effect.
That’s kinda fun but now consider this thought experiment:
A study tests the hypothesis that a caffeine-containing beverage improves 3 km running performance in well-trained runners. The runners complete the 3 km time trial four times in a randomised cross-over design, at the same time of day following a 5-day period of identical nutrition, sleep, and training. In each trial, runners are given 250 mL of cold fluid to ingest 30-mins prior to the time trial — the fluid is either water (no treatment), coffee (treatment), decaf coffee (placebo 1), and the same type of decaf coffee (placebo 2). The runners do not see the drinks being prepared and cannot see the drinks until they sip from them, but they obviously taste that water is not coffee. However, they cannot taste the difference between the coffee and the decaf. But, in one of the decaf coffee trials, runners are told that “This coffee has a lot of caffeine and will massively improve your performance today.” — this is the “belief” trial. When the study is finished and analysed, the data show that 3 km performance was equally faster in the coffee and the “belief” decaf trials than in the water and decaf trials, which were not different from one another. The researchers conclude that there was no placebo effect of decaf placebo 1 but there was a placebo effect (or belief effect) of decaf placebo 2 — decaf coffee gave the same performance-enhancing boost as coffee when runners were told it would.
As you can see, the placebo effect is complicated yet has potential in the athletic arena. But, does it actually work? The obvious question to ask is…
What is the scientific evidence on the placebo effect’s impact on athletic performance?
The placebo (or nocebo) effect is a difficult thing to assess. The placebo must be indistinguishable from the actual treatment, a non-placebo control group must also be included, studies must control for participants’ knowledge of the effect of the treatment and/or placebo, and studies must control for the participant correctly guessing whether they have received treatment or not. These things are not always possible and are sometimes impossible (e.g. you cannot blind a person from doing a type of exercise; similarly, you can’t blind someone from being in a sauna).
In studies with a placebo control and a non-placebo control, the placebo effect explains ~50% of the beneficial effect of exercise training on cognitive function and psychological outcomes inc. anxiety, depression, mood, etc (Lindheimer et al. 2015: placebo effect sizeAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful that effect is. Common effect size measures include Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. = 0.20, 95% CIA measure of uncertainty used in Frequentist statistics. The 95% confidence interval is a plausible range of values within which the true value (e.g., the true treatment effect) would be found 95% of the time if the data were repeatedly collected in different samples of people. If this range of values (the confidence interval) crosses zero, there is little confidence in the effect. -0.02 to 0.41; exercise effect size = 0.37, 95%CI 0.11 to 0.63). But, more high-quality randomised controlled trials are needed to bolster these findings and none of these studies was on athletes.
When pooling several nutritional (inc. caffeine, beta-alanine, sodium bicarbonate, and anabolic steroids) and mechanical (inc. electrical muscle stimulation, kinesiology tape, blood flow restriction, magnetic bands, and cold water immersion) interventions in “deception” studies where participants are led to believe they’re receiving a beneficial treatment, there are small to moderate placebo effects (effect sizeAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful that effect is. Common effect size measures include Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. = 0.36) and nocebo effects (effect size = 0.37) on sports performance. But, the methodological quality of the studies is generally poor so more high-quality randomised controlled trials are needed.
When specifically assessing caffeine and buffering supplements (e.g. sodium bicarbonate) in blinded studies where participants don’t know if they’re on placebo or treatment, the placebo has a trivial (tiny) but meaningful (the 95% CIA measure of uncertainty used in Frequentist statistics. The 95% confidence interval is a plausible range of values within which the true value (e.g., the true treatment effect) would be found 95% of the time if the data were repeatedly collected in different samples of people. If this range of values (the confidence interval) crosses zero, there is little confidence in the effect. does not cross zero) effect on (effect sizeAn effect size is a quantitative measure of the magnitude of a relationship or difference between groups in a study. Unlike p-values, effect sizes show how large or meaningful that effect is. Common effect size measures include Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. = 0.09, 95%CI 0.01 to 0.17) vs. the non-placebo-control while the actual treatment has a small effect (effect size = 0.37, 95%CI 0.20 to 0.56) vs. no treatment on running and cycling time-to-exhaustion and time-trial performance. This means that ~25% of the performance-enhancing effect of caffeine or buffers is explained by a placebo effect.
Therefore…
Since placebo effects are real and have a small effect, it would be prudent for an athlete (and an athlete’s support team) to try to maximise the placebo effect of a proven treatment by encouraging belief in its effectiveness.
If you choose to use the placebo effect, a reasonable approach is to:
Instil belief in the approaches you use and in the people you work with. Note: this is based on effective doses used in research.
But be aware that psychological belief in an approach can be a replacement for science EXCEPT when science shows that the approach has detrimental physiological effects. For example, many athletes claim that they “recover better” when using ice baths (cold water immersion) after exercise. But, while cold water immersion may reduce feelings of soreness and improve feelings of recovery in the short-term, in the long-term, daily post-exercise cold water immersion has been shown to blunt training adaptations.
Can the placebo effect enhance athletic recovery and performance?
Using the placebo effect is likely to improve performance, but unlikely to improve recovery. On the flipside, the nocebo effect (negative expectations) is likely to impair performance.
The effect sizeAn effect size is a standardized measure of the magnitude of an effect of an intervention. Unlike p-values, effect sizes show how large the effect is and indicate how meaningful it might be. Common effect size measures include standardised mean difference (SMD), Cohen’s d, Hedges’ g, eta-squared, and correlation coefficients. for improving performance is small.
Due to insufficient research, it is unclear whether the effect is similar between trained athletes and untrained folks, or between males and females.
Keep in mind: the studies are small and few in number, and there is high heterogeneityHeterogeneity shows how much the results in different studies in a meta-analysis vary from each other. It is measured as the percentage of variation (the I2 value). A rule of thumb: if I2 is roughly 25%, that indicates low heterogeneity (good), 50% is moderate, and 75% indicates high heterogeneity (bad). High heterogeneity means there’s more variability in effects between studies and, therefore, a less precise overall effect estimate. (variability) in effects between studies with a high risk of biasRisk of bias in a meta-analysis refers to the potential for systematic errors in the studies included in the analysis. Such errors can lead to misleading/invalid results, and unreliable conclusions. This can arise because of issues with the way participants are selected (randomisation), how data is collected and analysed, and how the results are reported.. So, the overall certainty of evidenceCertainty of evidence tells us how confident we are that the results reflect the true effect. It’s based on factors like study design, risk of bias, consistency, directness, and precision. Low certainty means more doubt and less confidence, and that future studies could easily change the conclusions. High certainty means that the current evidence is so strong and consistent that future studies are unlikely to change conclusions. is lowA low quality of evidence means that, in general, studies in this field have several limitations. This could be due to inconsistency in effects between studies, a large range of effect sizes between studies, and/or a high risk of bias (caused by inappropriate controls, a small number of studies, small numbers of participants, poor/absent randomisation processes, missing data, inappropriate methods/statistics). When the quality of evidence is low, there is more doubt and less confidence in the overall effect of an intervention, and new studies could easily change overall conclusions. The most effective way to enhance the quality of evidence is for scientists to conduct large, well-controlled, high-quality randomised controlled trials.. Therefore, additional high-quality randomised controlled trialsThe “gold standard” approach for determining whether a treatment has a causal effect on an outcome of interest. In such a study, a sample of people representing the population of interest is randomised to receive the treatment or a no-treatment placebo (control), and the outcome of interest is measured before and after exposure to the treatment and control. are needed to increase the certainty (confidence) in the overall effect size of the placebo effect.
The nice part: the placebo effect doesn't appear to hurt recovery or performance. So, if you like using it and believe it works for you, give it a whirl.
But, remember that the placebo effect does not make an athlete — the placebo effect does not replace training and placebo effects are not additive. I.e., using and believing in 5 supplements that don’t have proven ergogenic effects won't give you 5-times the benefit. Instead, it will give you 5 more factors that cost money and time and could return a positive doping test. This creates stress, and stress is the antithesis of the recovery and adaptation required for high performance. Yes, the placebo effect is real but, before playing with magic, first invest your valuable time and money in the things that will guarantee high performance: Optimise your training load, and nail your sleep, nutrition, and rest. No tricks. Just learn to understand yourself, watch for patterns, and intervene.
How to use this: INSERT.
Full list of meta-analyses examining the placebo effect for recovery and performance.
Here are the meta-analyses I've summarised above:
Placebo and Nocebo Effects in Motor Performance: An Overview of Reviews. Brietzke et al. Brain Behav (2025)
Negative expectations and measurable movement mechanics: a scoping review of the nocebo effect on motor performance. Burgos-Tirado et al. Front Hum Neurosci (2025)
Placebo and Nocebo Effects on Sports and Exercise Performance: A Systematic Literature Review Update. Chhabra et al. Nutrients. (2024)
Nonplacebo Controls to Determine the Magnitude of Ergogenic Interventions: A Systematic Review and Meta-analysis. Felipe Miguel Marticorena, Arthur Carvalho, Luana Farias DE Oliveira, Eimear Dolan, Bruno Gualano, Paul Swinton, Bryan Saunders. Med Sci Sports Exerc (2021)
The Placebo and Nocebo effect on sports performance: A systematic review. Philip Hurst, Lieke Schipof-Godart, Attila Szabo, John Raglin, Florentina Hettinga, Bart Roelands, Andrew Lane, Abby Foad, Damian Coleman, Chris Beedie. Eur J Sports Sci (2020)
Consensus statement on placebo effects in sports and exercise: The need for conceptual clarity, methodological rigour, and the elucidation of neurobiological mechanisms. Beedie et al. (2018) Eur J Sport Sci
Quantifying the placebo effect in psychological outcomes of exercise training: a meta-analysis of randomized trials. Jacob B Lindheimer, Patrick J O'Connor, Rod K Dishman. Sports Med (2015)
Consult WADA’s prohibited list.
Cross-check your meds against the Global DRO drug reference list.
Only choose supplements that have been independently tested by Informed Sport or LabDoor. You are the only person responsible for what goes in your body. Ignorance is not an excuse! Stay educated. Be informed.
Nail your daily nutrition habits first, layer specific sports nutrition on top of that, and then start to consider supplements.
Disclaimer I occasionally mention brands and products, but it is important to know that I don't sell recovery products, supplements, or ad space, and I'm not affiliated with / sponsored by / an ambassador for / receiving advertisement royalties from any brands. I have conducted biomedical research for which I’ve received research money from publicly-funded national research councils and medical charities, and also from private companies, including Novo Nordisk Foundation, AstraZeneca, Amylin, the A.P. Møller Foundation, and the Augustinus Foundation. I’ve also consulted for Boost Treadmills and Gu Energy on R&D grant applications, and I provide research and scientific writing services for Examine.com. Some of my articles contain links to information provided by Examine.com but I do not receive any royalties or bonuses from those links. Importantly, none of the companies described above have had any control over the research design, data analysis, or publication outcomes of my work. I research and write my content using state-of-the-art, consensus, peer reviewed, and published scientific evidence combined with my empirical evidence observed in practice and feedback from athletes. My advice is, and always will be, based on my own views and opinions shaped by the scientific evidence available. The information I provide is not medical advice. Before making any changes to your habits of daily living based on any information I provide, always ensure it is safe for you to do so and consult your doctor if you are unsure.