Compression garments worn for hours after hard exercise give small benefits for strength recovery and soreness, but wearing them during endurance exercise does not improve performance
Interpretation
Several meta-analyses agree on a small recovery benefit: 23 studies, overall ES 0.38 and strength ES 0.62 (Brown 2017); 12 studies after damaging exercise, soreness g = 0.40 and strength g = 0.46 (Hill 2014); a newer 27-study analysis finds smaller strength and power effects (g = 0.21 and 0.23; Li 2025). In endurance athletes specifically, only 9 trials (158 people) exist and certainty is very low (Hagner-Derengowska 2026). For performance while wearing them, 51 randomised trials (899 runners) show no effect on race time or time to exhaustion (Wang 2025), consistent with a half-ironman study of 36 triathletes (Del Coso 2014) and a 20-runner marathon trial (Zaleski 2019). Graded moderate because the direction is consistent across reviews, but most studies are unblinded, use non-triathlete populations and eccentric damage protocols, rarely report sex splits, and newer estimates are smaller.
Practical implication
Wear compression, if you like it, for hours after damaging sessions and races and on long journeys home; don't expect it to make you faster, and treat it as an add-on to sleep, food and easy movement. Evidence in triathletes is thin and partly vulnerable to placebo. Sports garments are not medical compression: people with clotting history, circulatory problems or prescribed stockings should follow clinical advice.
Sources
8 sources, independently verified.
Compression garments produced a small overall recovery benefit (ES 0.38). Strength recovery benefited most (ES 0.62), with large effects at 2-8 h (1.14) and beyond 24 h (1.03); after resistance exercise the benefit beyond 24 h was ES 1.33. Recovery from endurance (metabolic) exercise was smaller overall, though next-day cycling performance showed ES 1.05 at 24 h. Pressure and training status did not clearly modify effects, and few studies reported applied pressure; participants were mostly not triathletes and could not be blinded.
Compression garments had moderate effects on soreness (Hedges' g = 0.40), strength (0.46), power (0.49) and creatine kinase (0.44) measured 24-72 h after muscle-damaging exercise. Damage protocols were mostly eccentric exercise, not endurance racing.
Three-level meta-analysis of 27 studies: compression garments reduced the loss of strength (g = 0.21) and power (g = 0.23) after fatiguing exercise, effects smaller than earlier meta-analyses; benefits in trained and untrained people, more pronounced in trained.
Only 9 trials (158 participants) tested post-exercise compression in endurance athletes; 10 of 11 risk-of-bias assessments were high risk. Soreness at about 24 h favoured compression (g = 0.62) but the 95% CI ran from -0.57 to 1.82; effects on later endurance performance could not be pooled. Certainty very low; limited blinding leaves results vulnerable to expectation effects.
Compression clothing had no significant effect on running performance from 400 m to the marathon, oxygen uptake, lactate, heart rate or perceived temperature; small positive effects on time to exhaustion, running economy and perceived exertion, and larger effects on post-run soreness.
Wearing compression while running did not improve race time (SMD -0.07, 95% CI -0.22 to 0.09) or time to exhaustion (SMD 0.04); garment type, race type and surface made no difference. Compression reduced soft-tissue vibration (SMD -0.43). Certainty of evidence low to very low.
In a half-ironman at 29 °C and 73% humidity, race time was 310 min with stockings vs 315 min without (no significant difference). Post-race falls in jump height (-9.2% vs -8.5%) and leg power, myoglobin, creatine kinase, soreness and RPE were similar between groups. Not blinded; matched rather than randomised allocation.
Creatine kinase rose 273% from baseline by the finish and 1,094% at 24 h; compression socks worn during the marathon did not change either rise. Small, unblinded trial.