“Massage” is not one thing. It is a whole family of genuinely different techniques, developed in different countries and centuries, aimed at different problems, and backed by evidence of wildly different quality, from genuinely solid to essentially none at all. This is a properly referenced guide to the major styles, what each one actually is, what the trial evidence for each one actually shows, good and bad, and where to go for the full, deeper guide on the ones we cover in depth ourselves.
As a starting point, general massage therapy as a category has a solid evidence base. A large series of systematic reviews and meta-analyses of randomised controlled trials, covering pain populations both general and surgical, found massage therapy effective for treating pain compared with no treatment or sham, and beneficial for anxiety and health-related quality of life compared with active comparators. That is a genuinely strong foundation for the category as a whole. It does not, however, tell you which specific style suits which specific problem, or which styles have far weaker evidence hiding behind that general, reassuring finding. That is what the rest of this guide is for.
Swedish massage
The default Western “relaxation massage,” and the technique most people picture when they hear the word at all. Developed in the early nineteenth century and built around long gliding strokes (effleurage), kneading (petrissage), and lighter tapping and vibration movements, generally performed with oil directly on skin. It is the foundation most other Western massage styles, including deep tissue and sports massage, are built on top of.
The evidence is genuinely encouraging for relaxation and anxiety specifically. A randomised, single-masked clinical trial at Emory University’s Mood and Anxiety Disorders Program found that twice-weekly Swedish massage monotherapy, with no other treatment, produced a significant reduction in symptoms of generalised anxiety disorder compared with a light-touch control condition over six weeks. It is worth being honest, though, that Swedish massage is not automatically the right choice for every condition it gets recommended for: a meta-analysis comparing complementary therapies for fibromyalgia found that shiatsu, a different, pressure-based style covered further down this page, reduced pain, fatigue and sleep disturbance, while Swedish massage specifically did not improve those outcomes.
Deep tissue massage
Uses slower strokes and considerably firmer, sustained pressure than Swedish massage, targeting deeper layers of muscle and the connective tissue surrounding it, generally to address chronic tension or a specific problem area rather than for general relaxation.
This is one of the better-evidenced styles for a specific, common problem: chronic low back pain. A prospective randomised trial found that two weeks of deep tissue massage produced pain reduction comparable to a course of non-steroidal anti-inflammatory drugs, a genuinely notable comparison given how routinely NSAIDs are reached for first. A separate randomised pilot study in patients with ankylosing spondylitis, an inflammatory form of spinal arthritis, found deep tissue massage produced meaningful improvements in lower back pain and functional capacity. The honest caveat: both of these are individual trials with modest sample sizes rather than large-scale meta-analyses, so “comparable to NSAIDs” should be read as a genuinely promising early finding rather than a settled clinical equivalence.
Sports massage
Massage specifically applied before, during or after athletic activity, generally combining Swedish and deep tissue techniques and adapted to the specific demands of a sport or a particular muscle group under strain.
The largest and most rigorous review of this specific question, a 2020 systematic review and meta-analysis of 29 studies and 1,012 participants published in BMJ Open Sport & Exercise Medicine, found no evidence that massage improves strength, jump height, sprint speed or endurance, but did find small, statistically significant improvements in flexibility and in delayed-onset muscle soreness (DOMS), the ache that typically peaks a day or two after unfamiliar or intense exercise. A separate meta-analysis focused specifically on DOMS found a significant reduction in soreness ratings at 24, 48 and 72 hours after intense exercise in massaged participants compared with those receiving no massage. Put plainly: the honest, current evidence supports sports massage for feeling less sore afterwards, not for making you measurably stronger, faster or more enduring.
Hot stone massage
Combines standard massage technique with smooth, heated stones, generally basalt, placed on or used to massage the body, adding thermal therapy to manual pressure.
The current evidence is genuinely encouraging for a narrower set of outcomes than the treatment is often marketed for. Randomised controlled trials in postoperative orthopaedic patients and in patients undergoing maintenance haemodialysis have both found hot stone massage significantly improved sleep quality and reduced anxiety compared with standard care. Evidence for hot stone massage specifically treating chronic musculoskeletal pain is considerably thinner: as recently as 2023, researchers describing the first proper randomised trial in this specific area explicitly noted that it remains under-studied and requires considerably more rigorous research before firm conclusions can be drawn.
Thai massage
Traditional Thai massage, sometimes called Thai yoga massage, combines rhythmic pressure along energy lines with assisted stretching, the practitioner moving the recipient’s own body through yoga-like positions, generally performed clothed on a floor mat rather than on a table with oil.
A randomised controlled trial comparing court-type traditional Thai massage directly against amitriptyline, a genuine prescription medication used for chronic tension-type headache, found significant reductions in headache pain intensity in the massage group at multiple assessment points. A broader systematic review concluded that traditional Thai massage possibly helps with chronic pain generally, with individual studies reporting pre- to post-treatment pain reductions ranging from 25 to 80 percent, alongside improvements in disability, muscle tension, flexibility and anxiety. This is a genuinely more encouraging evidence picture than several other styles on this list, while still falling short of large-scale, definitive proof.
Shiatsu and acupressure
Shiatsu, Japanese for “finger pressure,” uses sustained pressure from fingers, thumbs and palms along the body, generally performed clothed on a mat, and shares its theoretical basis with acupressure, a related but distinct technique using similar pressure points.
This is a genuine case where two closely related techniques have very different evidence bases, and it is worth being precise about that difference rather than treating “shiatsu and acupressure” as one thing. A major systematic review screening over 1,700 publications found the evidence for acupressure specifically reasonably strong, particularly for pain (including period pain, lower back pain and labour pain), post-operative nausea and vomiting, and sleep in institutionalised elderly populations. Shiatsu itself, the review found, had been studied far less rigorously, with only a single randomised controlled trial identified among the included studies, and was judged to have evidence of insufficient quantity and quality to draw firm conclusions, whatever its individual practitioners and clients report anecdotally.
Reflexology
Based on the theory that specific points on the feet, hands and ears correspond to other organs and systems throughout the body, with pressure applied to those points intended to produce effects elsewhere.
This is the style on this list with the least supportive evidence, and the honest, house style here is to say so plainly rather than soften it. A systematic review of randomised controlled trials, later updated with 23 qualifying studies, found that the nine highest-quality trials generated negative findings, against five lower-quality trials reporting positive results, and concluded that the best available clinical evidence does not convincingly demonstrate reflexology to be an effective treatment for any medical condition. That does not mean reflexology feels unpleasant, foot massage of almost any kind tends to feel good, but it does mean the specific therapeutic claims tied to reflex points and internal organs are not well supported by controlled research.
Manual lymphatic drainage
A very light, specific, rhythmic technique intended to encourage lymph fluid movement, most commonly used clinically alongside compression therapy for lymphoedema following breast cancer surgery, distinct from the much firmer pressure used in most other styles on this list.
The evidence here is genuinely mixed rather than clearly positive or negative, which is itself the honest finding worth reporting. A 2020 meta-analysis of 17 randomised controlled trials and almost 1,900 patients found that, overall, manual lymphatic drainage did not significantly reduce lymphoedema volume compared with control treatment, though a meaningful benefit did emerge in patients under 60 and when the intervention was delivered within the first month after surgery. Other reviews have reached broadly similar conclusions: some genuine benefit in specific circumstances and populations, no consistent benefit as a general add-on to standard compression therapy. It remains a legitimate clinical technique used by trained lymphoedema therapists, but the blanket marketing claim that it “detoxifies” or benefits everyone equally is not well supported.
Indian head massage
A traditional technique, often taught under the name Champissage, working across the scalp, temples, jaw, neck and, in its full professional form, the upper back and shoulders. Manual pressure applied to the scalp and neck is consistent with the well-established finding that massage therapy can meaningfully ease tension-type headache, and a small study of standardised scalp massage found that daily sessions over 24 weeks increased hair thickness, with gene expression analysis pointing to mechanical stretching forces reaching the dermal papilla cells that regulate hair growth. That hair-thickness study is genuinely interesting but should be read with real caveats: it involved only nine men, used a mechanical device rather than hand massage, and ran for a full 24 weeks, considerably longer than most people commit to before expecting results.
Journal · Living Well
How to Give Yourself an Indian Head Massage
The full technique, step by step, and what the evidence actually supports.
Myofascial release
Self-administered pressure with a foam roller or ball, marketed as “releasing” tight fascia. The mechanical claim in the name does not hold up well under scrutiny: a frequently cited mathematical modelling study calculated that the pressures needed to actually deform dense fascial tissue are far beyond what a hand, elbow or foam roller can generate, meaning the palpable “release” people feel cannot be genuine mechanical tissue deformation. Chronic, repeated foam rolling has produced conflicting results for flexibility and no consistent benefit for athletic performance in systematic review, and the more plausible mechanism is neurological, sustained pressure calming the nervous system’s protective muscle guarding, rather than mechanical. Applied clinically for chronic low back pain, however, the evidence is genuinely stronger, with a systematic review and meta-analysis of randomised trials finding meaningful improvements in pain and disability.
Journal · Living Well
Myofascial Release: What the Evidence Actually Shows
Why the name oversells the mechanism, and what is genuinely happening instead.
Facial self-massage
Gentle, directional massage worked into an ordinary moisturising routine, rather than a separate treatment in its own right. Two mechanisms here are genuinely plausible and reasonably well supported: a real but short-lived reduction in morning puffiness from mechanically encouraging overnight-pooled lymphatic fluid to move, and a slower, cumulative muscle-toning case that suits an unhurried evening routine better than a rushed morning one. Neither mechanism supports the more dramatic marketing claims sometimes attached to facial massage, permanently lifting the jawline or eliminating wrinkles, and the honest, more modest case for it is good enough without the exaggeration.
Journal · Living Well
Self-Massage While You Moisturise
Different techniques for morning and night, and an honest look at what facial massage can and cannot achieve.
Aromatherapy massage rituals
Massage built specifically around a chosen oil blend, where touch and the oil’s own aromatic and topical properties are both doing genuine work, rather than a generic carrier oil standing in as a lubricant. Our own twice-weekly hair oil ritual uses rosemary essential oil, which has a genuinely rare thing for this category, a randomised comparative trial finding it performed comparably to 2% topical minoxidil for androgenetic alopecia over six months. Our post-exercise warm-down oil is built around black pepper and ginger, chosen for a real, if largely sensory, warming rubefacient effect rather than a claim that the oil itself repairs muscle damage. Both are worked examples of a wider principle: an aromatherapy ritual is only as good as the evidence behind the specific oils chosen for it, not the massage technique alone.
Botanical Rituals
The Twice-Weekly Hair Oil
A scalp-massage ritual built around a specific hair oil blend.
Botanical Rituals
The Warm-Down
A warming post-exercise self-massage oil for tired muscles.
Setting up your space
Whichever style you are practising, the setting makes a genuine difference to how worthwhile it feels. We have written up the practical, low-effort details separately.
Journal · Living Well
How to Create an Inviting Massage Space at Home
Temperature, lighting, towels and surface: the details that turn self-massage into a proper ritual.
Frequently asked questions
Which type of massage has the strongest evidence behind it?
For a specific, common problem, deep tissue massage for chronic low back pain and Thai massage for tension-type headache both have genuinely encouraging randomised trial evidence, including direct comparisons against standard medical treatments. Acupressure, distinct from shiatsu, also has a comparatively strong evidence base for pain, nausea and sleep. General massage as a category has solid evidence for pain and anxiety across a wide range of styles.
Which type of massage has the weakest evidence behind it?
Reflexology, honestly. Systematic reviews of randomised controlled trials have concluded that the best available clinical evidence does not convincingly demonstrate it to be an effective treatment for any medical condition, with the highest-quality trials specifically producing negative results more often than positive ones. Shiatsu, as distinct from the better-studied acupressure, and manual lymphatic drainage as a general-purpose treatment also have thinner or more mixed evidence than their popularity suggests.
How often should I do self-massage?
Most of the self-administered styles above are designed as short, frequent habits rather than occasional treatments: a five-minute daily or twice-weekly scalp or facial routine, or thirty to sixty seconds per muscle group for myofascial release before or after exercise. Daily use is generally safe for all of these techniques, provided you are not working directly over an injury, inflamed area or condition your doctor has advised against.
Can self-massage replace a professional massage?
Not fully. A trained practitioner can access muscles you cannot fully relax yourself, since your own arms and shoulders are always doing some work while your hands are active, and can apply techniques, pressure and modalities such as heated stones or assisted stretching that a self-administered version cannot replicate. Self-massage is genuinely worthwhile as a frequent, low-cost habit that complements rather than replaces occasional professional treatment.
Do I need oil for self-massage, or does it work dry?
It depends on the style. Scalp and myofascial-release techniques work perfectly well dry and need nothing added. Facial massage and aromatherapy rituals such as our hair oil and warm-down blends are built specifically around an oil, both for slip and for the oil’s own topical properties, so those are genuinely better done with one.
Does massage actually break up “knots” in muscle?
Probably not in the literal sense the phrase suggests. What is commonly felt as a “knot” is more likely an area of localised muscle tension and heightened nerve sensitivity than a physical tangle of tissue that pressure mechanically undoes. Sustained pressure genuinely does reduce that tension and sensitivity through the nervous system, which is why it feels effective, even though the physical “untangling” description is more metaphor than mechanism.
Is it normal for self-massage to hurt a little?
A strong, tolerable discomfort over a genuinely tight area is normal and expected, particularly with a foam roller or firm scalp pressure. Sharp, shooting or worsening pain is not, and is a signal to stop and, if it persists, seek proper assessment rather than continuing to push through it.
References
- Rapaport, M.H. et al. (2016). Acute Swedish massage monotherapy successfully remediates symptoms of generalized anxiety disorder: a proof-of-concept, randomized controlled study. Journal of Clinical Psychiatry, 77(7), e883-891. PubMed
- Majchrzycki, M., Kocur, P. & Kotwicki, T. (2014). Deep tissue massage and nonsteroidal anti-inflammatory drugs for low back pain: a prospective randomized trial. The Scientific World Journal, 2014, 287597. PMC3953439
- Davis, H.L., Alabed, S. & Chico, T.J.A. (2020). Effect of sports massage on performance and recovery: a systematic review and meta-analysis. BMJ Open Sport & Exercise Medicine, 6(1), e000614. PubMed
- Damapong, P. et al. (2015). A randomized controlled trial on the effectiveness of court-type traditional Thai massage versus amitriptyline in patients with chronic tension-type headache. Evidence-Based Complementary and Alternative Medicine, 2015, 930175. PMC4587431
- Robinson, N., Lorenc, A. & Liao, X. (2011). The evidence for Shiatsu: a systematic review of Shiatsu and acupressure. BMC Complementary and Alternative Medicine, 11, 88. PMC3200172
- Ernst, E., Posadzki, P. & Lee, M.S. (2011). Reflexology: an update of a systematic review of randomised clinical trials. Maturitas, 68(2), 116-120. PubMed
- Liang, M. et al. (2020). Manual lymphatic drainage for lymphedema in patients after breast cancer surgery: a systematic review and meta-analysis of randomized controlled trials. Medicine, 99(49), e23192.
- Koyama, T. et al. (2016). Standardized scalp massage results in increased hair thickness by inducing stretching forces to dermal papilla cells in the subcutaneous tissue. Eplasty, 16, e8. PMC4740347
- Chaudhry, H., Schleip, R., Ji, Z., Bukiet, B., Maney, M. & Findley, T. (2008). Three-dimensional mathematical model for deformation of human fasciae in manual therapy. Journal of the American Osteopathic Association, 108(8), 379-390. PubMed
- Pagaduan, J.C., Chang, S.Y. & Chang, N.J. (2022). Chronic effects of foam rolling on flexibility and performance: a systematic review of randomized controlled trials. International Journal of Environmental Research and Public Health, 19(7), 4315. PMC8998857
- Myofascial release for chronic low back pain: a systematic review and meta-analysis. Frontiers in Medicine, 8, 697986 (2021). Frontiers
- Panahi, Y., Taghizadeh, M., Marzony, E.T. & Sahebkar, A. (2015). Rosemary oil vs minoxidil 2% for the treatment of androgenetic alopecia: a randomized comparative trial. Skinmed, 13(1), 15-21. PubMed 25842469