This study investigated associations of ACTN3 rs1815739 and ACE rs1799752 with a pooled non contact knee ligament rupture (KLR) phenotype in elite handball players. 329 elite Polish handball players (18–35 years; ≥ 5 years of competitive experience) were included. Self-reported non-contact KLRs in 103 cases were analyzed as a pooled phenotype; 226 injury-free players were controls. TaqMan assays were used. Logistic regression and multifactor dimensionality reduction (MDR) were applied. ACE rs1799752 showed no association with the pooled KLR phenotype. ACTN3 rs1815739 genotype frequencies in cases versus controls were 46.6% versus 32.7% for RR, 39.8% versus 51.8% for RX, and 13.6% versus 15.5% for XX. The codominant genotype distribution differed before but not after sex adjustment (FDR-corrected global p = 0.046; sex-adjusted FDR corrected global p = 0.06). Under the dominant model, X-allele carriers were less frequent among cases than controls (RX + XX: 53.4% versus 67.3%) and had lower odds of the pooled KLR phenotype than RR homozygotes in both the unadjusted analysis (OR = 0.56, 95% CI: 0.35–0.90; FDR-corrected p = 0.02) and the sex-adjusted analysis (OR = 0.57, 95% CI: 0.35–0.92; FDR-corrected p = 0.02). MDR analysis identified a significant two locus model including ACE rs1799752 and ACTN3 rs1815739 (p = 0.01), within which the DD/RR, DD/XX, and ID/RR genotype combinations occurred more frequently among cases than controls. ACTN3 X-allele carriers had lower odds of pooled KLR under the dominant model. Specific ACE–ACTN3 genotype combinations were more frequent in the pooled KLR group. MDR findings require cautious interpretation and independent replication.