Thirty-two randomized controlled trials containing 1098 patients with delayed onset muscle soreness were included in this meta-analysis comparing heat therapy, cold therapy, and control (passive recovery) for the reduction of pain. Comparisons between heat therapy vs. control, cold therapy vs. control, and heat therapy vs cold therapy were performed, and outcomes were assessed at ≤24 hours and >24 hours follow up. Additionally, sub-group analyses were performed by intervention type, for cold water immersion therapy, other cold therapies, hot packing, and other heat therapies. Results revealed statistically significantly favourable pooled pain scores in the cold therapy group and heat therapy group vs control at ≤24 hours (p<0.05 for both). At >24 hours, pooled pain scores in the heat therapy group were statistically significantly favourable vs the control group (p=0.004), but not in the cold therapy group vs control (p=0.05). No statistically significant differences in pooled pain scores were observed between the heat therapy and cold therapy groups at both ≤24 hours and >24 hours follow up (p>0.05 for both). For the subgroup analyses by intervention type, cold water immersion therapy resulted in statistically significantly favourable pooled pain scores at both ≤24 hours and >24 hours vs. control (p<0.05 for both). However, for other cold therapies, pooled pain scores were statistically significantly favourable vs. control at ≤24 hours only (p=0.03). For hot packing vs. control, statistically significant differences in pooled pain scores were observed in favour of the hot packing group at both ≤24 hours and >24 hours (p<0.05 for both). In the sub-group analyses of other heat therapies vs. control, no statistical differences in pooled pain scores were observed at both time-points (p>0.05 for both).