Part 1 of this genetic association series highlighted several genetic variants independently associated with elite status in rugby. However, it is highly likely that the genetic influence on elite status is polygenic due to the interaction of multiple genes. Therefore, the aim of the present study was to investigate whether polygenic profiles of elite rugby athletes differed from non-athletes utilising 13 genetic polymorphisms previously associated with tendon/ligament injury. Total genotype score (TGS) was calculated and multifactor dimensionality reduction (MDR) was used to calculate SNP-SNP epistasis interactions. Based on our elite rugby data from Part 1, mean TGS was significantly higher in elite rugby athletes (52.1 ± 10.7) than non-athletes (48.7 ± 10.8). There were more elite rugby athletes (54%) within the upper TGS quartile, and fewer (46%) within the lower quartile, compared to non-athletes (31% and 69%, respectively; P = 5·10-5), and the TGS was able to distinguish between elite rugby athletes and non-athletes (area under the curve = 0.59; 95% confidence interval 0.55-0.63; P = 9·10-7). Furthermore, MDR identified a three-SNP model of COL5A1 rs12722, COL5A1 rs3196378 and MIR608 rs4919510 that was best able to predict elite athlete status, with a greater frequency of the CC-CC-CC genotype combination in elite rugby athletes (9.8%) than non-athletes (5.3%). We propose that elite rugby athletes possess ‘preferable’ musculoskeletal soft-tissue injury-associated polygenic profiles that have helped them achieve success in the high injury risk environment of rugby. These data may, in future, have implications for the individual management of musculoskeletal soft-tissue injury.