Background: Anterior cruciate ligament (ACL) has been reported to alter knee proprioception following injury/reconstruction. Joint position sense (JPS) is a submodality of proprioception which is objectively assessed by measuring the acuity of reproducing a predetermined target angle. Recent evidence warrants further investigation of the psychometric properties of knee JPS tests following ACL reconstruction (ACLR). The purpose of this study was to evaluate the test-retest reliability of a passive knee JPS test in individuals who underwent ACLR. Methods: This study utilized a cross-sectional repeated-measures design. Nineteen male participants, aged 26.3 ± 4.4 years, had undergone a unilateral ACLR, before 6.3 ± 2.5 months, completed two sessions of knee JPS evaluation, five days apart. The absolute error (AE), constant error (CE) and variable error (VE) of the passive knee JPS were measured for both limbs in the extension and flexion directions at two target angles (30° and 60° knee flexion) per each direction using the ipsilateral angle reproduction method in the sitting position. Intraclass correlation coefficients (ICCs), 95% confidence intervals (CIs), standard error of measurement (SEM) and smallest real difference (SRD) were calculated to evaluate the test-retest reliability. Results: ICCs were higher for JPS CE (operated knee: 0.43–0.86 non-operated knee: 0.32–0.91) compared to AE (0.18–0.59 0.09–0.86) and VE (0.07–0.63 0.09–0.73). The CE of the 90°–60° extension test showed moderate to excellent reliability for the operated knee (ICC 0.86 [95% CI 0.64–0.94], SEM 1.63°, SRD 4.53°) and good to excellent reliability for the non-operated knee (ICC 0.91 [95% CI 0.76–0.96], SEM 1.53°, SRD 4.24°). Conclusion: Test-retest reliability varied from poor to excellent depending on the target angle, direction of movement and outcome measure. The CE appeared to be a more reliable outcome measure than the