Introduction There are several postoperative evaluation methods for anterior cruciate ligament reconstruction (ACLR), and patient-reported outcome measurement (PROM) is one of the most important for evaluating postoperative clinical results. However, conventional PROMs have a high risk of a ceiling effect at one and two years postoperatively and do not accurately reflect functional improvement over time. Therefore, a longitudinal evaluation using PROM with a low risk of ceiling effect is necessary. The forgotten joint score-12 (FJS) was developed for clinical evaluation after arthroplasty. It is considered an evaluation method after arthroplasty with a low risk of demonstrating a ceiling effect. However, few studies have used the FJS as an evaluation method after ACLR, particularly those tracking changes over time. This study aimed to longitudinally evaluate FJS at one and two years after ACLR. Methods This is a retrospective observational study of postoperative patients using existing data and a questionnaire-based survey. This study included patients who underwent primary double-bundle ACLR between August 2017 and August 2021. We compared the FJS, knee injury and osteoarthritis outcome score (KOOS), and Lysholm knee scale (LKS) scores at one and two years post-surgery using the Wilcoxon signed-rank test. The ceiling effect for each PROM was calculated at one and two years post-surgery. A ceiling was defined as obtaining a perfect score in PROMs, and the risk of demonstrating a ceiling effect was the percentage of perfect scores among all cases. The risk of demonstrating a ceiling effect for each PROM was compared using the McNemar test. To identify factors influencing FJS one-year post-ACLR across all cases, multivariate linear regression analysis was conducted for the FJS.  Results Finally, 87 patients were included in this study. Fifty-six participants were women, and 31 were men, with an average age of 28.5 ± 11.8 years at the time of surgery and a BMI of 23.2 ± 3.7 kg/m. Fifty-eight patients with meniscus injuries requiring treatment were observed. A total of 54 patients were used to compare the results at one and two years, while 87 patients were included in the multivariate analysis for FJS at one year. The median of FJS at one and two years post-surgery were 90.6 and 95.8, respectively. The FJS showed a statistically significant improvement from one to two years (p = 0.033). Question #12 in FJS showed a significant improvement from the first to the second year postoperatively (1.74 ± 1.46 vs 1.15 ± 1.25 at one year vs two years, respectively; p = 0.0016). FJS had a lower risk of demonstrating a ceiling effect than KOOS ADL at one and two years (FJS vs KOOS ADL; at first year: 20.4% and 48.1%, p < 0.001; at second year: 33.3% and 63.0%, p = 0.0013). There was no difference compared to the other PROMs. According to the multivariate linear regression analysis, predictive factors for higher FJS scores at one year post-surgery were younger age and limb symmetry index of single-leg hop test(SLH-LSI) on the affected side that was close to that of the healthy side (SLH-LSI > 0.9). Conclusions The FJS continued to improve over two years after ACLR. The FJS post-ACLR was higher in younger individuals and those with SLH-LSI of 0.9 or higher.