One-hundred-fourteen patients with isolated ACL rupture were randomized to receive a bone–patellar tendon–bone (BPTB) graft (n = 58) or a hamstring graft (n = 56). The primary outcome was side-to-side anterior laxity measured with the KT-1000 arthrometer. Secondary outcomes included isokinetic knee-extension/flexion strength, Lachman and pivot-shift tests, KOOS, Tegner and Lysholm scores, and Kellgren–Lawrence (KL) osteoarthritis grade. Outcomes were assessed up to a mean of 19 years (range 18–20). Overall, the study revealed no significant between-group difference in KT-1000 laxity. However, hamstring-graft knees showed a 10.7 % lower peak flexion torque and 17.2 % lower total flexion work at 60 deg/s. Patient-reported scores, clinical laxity tests, and KL OA grades were similar. These findings suggest comparable long-term stability but a persistent hamstring-strength deficit with hamstring autografts.