The aim of this prospective study was to compare two arthroscopic techniques for reconstructing the anterior cruciate ligament, the "outside-in" (two incisions) and the "all-inside" (one incision) techniques. The results obtained for 30 patients operated on using the "outside-in" technique (group I) were compared with those for 29 patients operated on using the "all-inside" technique (group II). Before surgery, there were no significant differences between the groups in terms of Lysholm score, Tegner activity level, patellofemoral pain score, or knee laxity. Both groups displayed significant improvements in Lysholm score after 24 months, from 69 (16) to 91 (9) in group I and from 70 (17) to 90 (15) in group II (means (SD)). There were also significant improvements in patellofemoral pain scores in both groups, from 13 (6) to 18 (5) in group I and from 14 (6) to 18 (4) in group II after 24 months. No difference was found between the groups in knee stability at the 24 month follow up. The IKDC score was identical in both groups at follow up. The operation took significantly longer for patients in group I (mean 94 (15)) than for those in group II (mean 86 (20)) (p = 0.03). The mean sick leave was 7.7 (6.2) weeks in group I and 12.3 (9.7) weeks in group II (p = 0.026), indicating that there may be a higher morbidity associated with the "all-inside" technique. It can be concluded that there were no significant differences between the two different techniques in terms of functional results, knee laxity, or postoperative complications. The results were satisfactory and the outcome was similar in both treatment groups.