Background: To improve the understanding of the psychological impacts of anterior cruciate ligament (ACL) injury, a systematic review synthesizing the evidence on knee self-efficacy, fear avoidance beliefs and kinesiophobia following ACL injury is needed. Objective: The aim of this systematic review was to investigate knee self-efficacy, fear avoidance beliefs and kinesiophobia following ACL injury, and compare these outcomes following management with rehabilitation alone, early and delayed ACL reconstruction (ACLR). Methods: Seven databases were searched from inception to April 14, 2022. Articles were included if they assessed Tampa Scale of Kinesiophobia (TSK), Knee Self-Efficacy Scale (KSES), or Fear Avoidance Beliefs Questionnaire (FABQ). Risk of bias (RoB) was assessed using domain-based RoB tools (ROBINS-1, RoB 2, RoBANS), and GRADE-assessed certainty of evidence. Random-effects meta-analyses pooled outcomes, stratified by time post-injury (pre-operative, 3–6 months, 7–12 months, > 1–2 years, > 2–5 years, > 5 years). Results: Seventy-three studies (70% high RoB) were included (study outcomes: TSK: 55; KSES: 22; FABQ: 5). Meta-analysis demonstrated worse kinesiophobia and self-efficacy pre-operatively (pooled mean [95% CI], TSK-11: 23.8 [22.2–25.3]; KSES: 5.0 [4.4–5.5]) compared with 3–6 months following ACLR (TSK-11: 19.6 [18.7–20.6]; KSES: 19.6 [18.6–20.6]). Meta-analysis suggests similar kinesiophobia > 3–6 months following early ACLR (19.8 [4.9]) versus delayed ACLR (17.2 [5.0]). Only one study assessed outcomes comparing ACLR with rehabilitation only. Conclusions: Knee self-efficacy and kinesiophobia improved from pre-ACLR to 3–6 months following ACLR, with similar outcomes after 6 months. Since the overall evidence was weak, there is a need for high-quality observational and intervention studies focusing on psychological outcomes following ACL injury.