The aim of our study was to compare the healing response of biomechanically and biochemically different scaffolds in osteochondral defects created in rabbit medial femoral condyles. A block copolymer comprised of poly(ethylene oxide terephthalate) and poly(butylene terephthalate) was used to prepare porous scaffolds. The 70/30 scaffold (70 wt % poly(ethylene oxide terephthalate)) was compared to the stiffer 55/45 (55 wt % poly(ethylene oxide terephthalate)) scaffold. Nine 6-month-old rabbits were used. Osteochondral defects were filled with 55/45 scaffolds (n = 6); 70/30 scaffolds (n = 6); or left empty (n = 6). Defect sites were allowed to heal for 12 weeks. Condyles were macroscopically evaluated and analysed histologically using the O'Driscoll score for evaluating repair of osteochondral defects. Repair tissue in 70/30 scaffolds consisted of cartilage-like tissue on top of trabecular bone, whereas the tissue within the 55/45 scaffolds consisted predominantly of trabecular bone. O'Driscoll scores for 70/30 scaffolds were significantly better (p = 0.024) in comparison to untreated osteochondral defects and 55/45 scaffolds. This study reveals that the biomechanical and biochemical properties of the scaffold play an important role by themselves, and can affect the healing response of osteochondral defects. Scaffolds with low mechanical properties were superior in cartilage repair tissue formation.