Aim: To compare foot joint kinetics and energetics in male paediatric boys with and without blood-induced ankle joint destruction to these of matched control groups. Methods: A cross-sectional study was conducted in which 3D gait analysis data were collected from thirty-five male children (6-21 years) with severe or moderate haemophilia and twenty-six typically developing boys. Structural integrity of the tarsal foot joints of all haemophilic patients was assessed using the IPSG-MRI scale. All participants walked barefoot while adopting a physiological gait pattern. Three subgroups were created based on the IPSG-MRI scores: a group with no joint involvement (HealthyHaemo), with uni-or bilaterally involvement (PathoHaemo) and with only unilaterally involvement (Haemo_Unilateral_Patho). Results: The PathoHaemo group presented a significant lower Lisfranc peak dorsiflexion angular velocity (34.7°/s vs 71.4°/s, P = .000, Cohen d = 1.31) and a significantly higher Lisfranc peak plantarflexion angular velocity (−130.5°/s vs −51.8°/s, P = .000, Cohen d = 0.98) compared to the control group. The Haemo_Unilateral_Patho side had a significant higher Chopart peak dorsiflexion angular velocity compared to the Haemo_Unilateral_Healthy side (41.7°/s vs 31.9°/s, P = .002, Cohen d = 1.16). Conclusion: No evidence for mild and severe gait deviations could be demonstrated. Internal moments, used as a surrogate measure of joint loading, quantified by the multi-segment foot model were found to be similar within the three subanalyses. We suggest that the ongoing musculoskeletal development in children compensates for structural damage to the ankle joint.