Massage and Athletic Performance

Massage and Performance Recovery[1]

A Meta-Analysis from Sports Medicine, 2016

     Post-exercise massage is frequently touted as a recovery and recuperation aid by many athletes and coaches. However, the evidence for this claim is scarce. I REALLY want this claim to be true for I frequently engage in a weekly deep-tissue massage. It feels amazing, and I believe it helps – but does the evidence support my habit?

Sports Massage

     Bottom line up-front (BLUF): Massage can be effective in short duration (5-12 minutes) and if the recovery interval is also brief (up to 10 min). The average effects were larger for intensive mixed exercise (CrossFit style), followed by strength, then endurance exercise. Massage appeared to be more effective for untrained subjects than for athletes.

Massage is formally defined as “mechanical manipulation of body tissues with rhythmical pressure and stroking for the purpose of promoting health and well-being.” Various mechanisms have been proposed as to why massage is beneficial for recovery, including biomechanical, physiological, neurological, and psychological effects. There is little evidence for any of these except for psychological aspects(increased relaxation)[2] and some evidence that it reduces muscle inflammation.[3]

The authors of this paper conducted a systematic review and found that there were actually 22 randomized, controlled trials (RCT’s; these are the most powerful type of studies and the ONLY ones that can give us a clue to a cause-effect relationship). Five of these were automated methods(vibratory, pneumatic, water jet) and 17 were manual massage. Thus, the majority of the findings apply primarily to manual massage, possibly even self myofascial release as well (your friendly “roller”). More to come on self-myofacial release soon…..

Twenty-two studies met their criteria (270 total subjects in all – not a lot). The weighted average performance change due to massage was 3.5% and the weighted average effect size was g = 0.19.  Without getting all statistical, just know that a small effect size is around 0.1, moderate about 0.3, and large 0.5 or greater. An effect size of .19 indicates that small effects on performance recovery can be expected from massage.

Hopkins et al.[4] defined the smallest worthwhile enhancement (i.e. the minimum improvement making a certain strategy worthwhile) as the value increasing the chance of victory for an athlete by 10 %. For half-marathon and marathon races, they calculated values of 1%; for shorter distances including sprints, they found values of 0.5 %. The average performance change, although low at 3.5% falls into Hopkins’ range as being a worthwhile measure to include in your training regimen.

It was interesting that a tendency towards larger effects for shorter massage durations were found. The largest effects were found for short massage interventions of 5–12 min (on the target muscle group), while massage protocols lasting 15 min or more showed hardly any effect at all. However, this was for massage after an exercise session. It is possible that this tendency was due to a potential detrimental effect of prolonged massage on subsequent performance.

When looking at manual massage interventions directly after exercise, and performance testing within 10 minutes after the massage was completed, the strongest effect was for intense mixed exercise such as HIIT or CrossFit workouts.   This was followed by strength training and lastly endurance training.

The effects of massage by physical activity performed:

Modality Effect Size Performance Improvement
HIIT/CrossFit 0.45 7.9%
Strength Training 0.20 4.5%
Endurance Training 0.12 1.3%


     The length of massage on the target muscle groups favored shorter sessions:

Length of Massage (min) Effect Size Performance Improvement
5-6 0.32 7.8%
8-12 0.35 6.1%
15-20 0.08 0.9%
30 0.06 1.1%


Timing of the massage after training favored immediately after

Time of massage after training Effect Size Performance Improvement
5-10 minutes 0.45 7.9%
20-35 minutes 0.19 2.1%
1-6 hours -0.01 -0.4%
24 hours 0.07 1.7%
48 hours 0.08 2.8%
72 hours 0.09 3.9%
96 hours -0.02 -1.1%


What were the authors recommendations for Athletes and Coaches?

Massage duration of 5–12 min appears to be sufficient to maximize effects, and is most effective if the recovery period is short and preceded by intense, maximal strength or mixed-type exercise.   Massage appears to be ineffective in benefitting performance in endurance athletes.

The most profound effects on performance was immediately before the post-test, suggesting a short-term pre-massage effect that may be akin to using a foam roller for self-myofacial release. This is likely less relevant during competition than it might be during intense training cycles where enhancing medium to long-term recovery is important.

The fact that massage makes one “feel better” is important, as psychological effects are also very important in athletic performance.  This, along with the above data on strength and HIIT training are enough for me to continue using a foam roller on target muscle groups before and after intense training, as well as getting a whole body deep-tissue session weekly. Yay for me!


Lanny Littlejohn, MD

[1] Poppendieck W, Wegmann M, Ferrauti A, et al. Massage and performance recovery: A Meta-analytical review. Sports Med. 2016;46:183-204.

[2] Weerapong P, Hume PA, Kolt GS. The mechanisms of massage 
and effects on performance, muscle recovery and injury preven- 
tion. Sports Med. 2005;35(3):235–56.

[3] Crane JD, Ogborn DI, Cupido C, et al. Massage therapy attenu- ates inflammatory signaling after exercise-induced muscle dam- age. Sci Transl Med. 2012;4(119):119ra13.

[4] Hopkins WG, Hawley JA, Burke LM. Design and analysis of research on sport performance enhancement. Med Sci Sports Exerc. 1999;31(3):472–85.


Sharing is caring!

One response to “Massage and Athletic Performance

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.