The purpose of the present investigation was to evaluate physical outcome after anterior cruciate ligament (ACL) reconstruction with early versus late initiation of open kinetic chain (OKC) exercises for the quadriceps in patients operated on either patellar tendon or hamstring grafts. Sixty-eight patients, 36 males and 32 females, with either patellar tendon graft (34 patients) or hamstring graft (34 patients) were enrolled in this study. All patients were randomly allocated to either early (the 4th postoperative week) or late (the 12th postoperative week) start of OKC exercises for the quadriceps, resulting in four subgroups: patellar tendon reconstruction, early start (P4) or late start (P12) of OKC quadriceps exercises, hamstring tendon reconstruction, early start (H4) or late start (H12) of quadriceps OKC exercises. Prior to surgery and 3, 5 and 7 months later, assessments of range of motion (goniometer), anterior knee laxity (KT-1000), postural sway (KAT 2000), thigh muscle torques (Kin-Com dynamometer) and anterior knee pain (anterior knee pain score) were evaluated. No significant group differences were found in terms of range of motion 3, 5 and 7 months postoperatively. The H4 group showed a significantly higher mean difference of laxity over time of 1.0 mm (CI: 0.18-1.86) than the P4 group (P=0.04). Within the same type of surgery, the H4 against the H12, the mean difference over time was 1.2 mm (0.37-2.1) higher in the H4 group than in the H12 group (P=0.01). There were no significant group differences in terms of postural sway or anterior knee pain at the different test occasions. Significant differences in trends (changes over time) were found when comparing the four groups, for both quadriceps muscle torques (P<0.001) and hamstring muscle torques (P<0.001). All groups, except the P4 group, reached preoperative values of quadriceps muscle torques at the 7 months follow-up. In the H4 and the H12 groups, significantly lower hamstring muscle torques at the 7 months follow-up compared with preoperative values were found. In conclusion, early start of OKC quadriceps exercises after hamstring ACL reconstruction resulted in significantly increased anterior knee laxity in comparison with both late start and with early and late start after bone-patellar tendon-bone ACL reconstruction. Furthermore, the early introduction of OKC exercises for quadriceps did not influence quadriceps muscle torques neither in patients operated on patellar tendon nor hamstring tendon grafts. On the contrary, it appears as if the choice of graft affected the strength of the specific muscle more than the type of exercises performed. Our results could not determine the appropriate time for starting OKC quadriceps exercises for patients who have undergone ACL reconstruction with hamstring tendon graft. Future studies of long-term results of anterior knee laxity and functional outcome are needed.