Objectives Wearing a soft knee brace has been shown to reduce self-reported knee instability in persons with knee OA. There is a need to assess whether a soft knee brace has a beneficial effect on objectively assessed dynamic knee instability as well. The aims of the study were to evaluate the effect of a soft knee brace on objectively assessed dynamic knee instability and to assess the difference in effect between a non-tight and a tight soft knee brace in persons with knee OA. Methods Thirty-eight persons with knee OA and self-reported knee instability participated in a laboratory study. A within-subject design was used comparing no brace vs brace and comparing a non-tight vs a tight brace. The primary outcome measure was dynamic knee instability, expressed by the perturbation response (PR). The PR reflects deviation in the mean knee varus-valgus angle during level walking after a controlled mechanical perturbation. Linear mixed-effect model analysis was used to evaluate the effect of a brace on dynamic knee instability. Results Wearing a brace significantly reduced the PR compared with not wearing a brace (B = -0.16, P = 0.01). There was no difference between a non-tight and a tight brace (B = -0.03, P = 0.60). Conclusion This study is the first to report that wearing a soft knee brace reduces objectively assessed dynamic knee instability in persons with knee OA. Wearing a soft brace results in an objective improvement of knee instability beyond subjectively reported improvement. Trial registration Nederlands Trial register (trialregister.nl) NTR6363@en