The purpose of this study was to determine the reliability of 1 repetition maximum (1RM) split squat (SS) and establish the efficacy of collecting 1RM-SS and 1RM bilateral squat (BLS) data in the same session, for a non-resistance-trained recreationally active population. Fourteen males performed a submaximal familiarization session and 5 testing sessions. After familiarization, the 1RM-SS was tested in the following 3 sessions. In session 4, subjects were tested in both 1RM-SS and 1RM-BLS, with half performing SS then BLS and the remainder BLS then SS. In session 5, the testing order was reversed. Reliability statistics calculated included the following: changes in mean across sessions, coefficient of variation calculated from the typical error (TE) scores (%CV(TE)), and test-retest reliability (intraclass correlation coefficient [ICC]) of 1RM-SS. Statistically significant differences between the mean 1RM-SS in sessions 1 and 2 (2.14 kg, p = 0.001), and sessions 1 and 3 (2.86 kg, p < 0.003) were found, indicating the requirement for an additional familiarization session before 1RM-SS data collection. The %CV(TE) was 2.53% and the ICC was 0.97 for the 1RM-SS protocol. Performing SS before BLS tended to increase the mean 1RM-BLS (+2.1%), although the difference was not significant (p = 0.055). A reliable measure of 1RM-SS can be determined after 1 submaximal and 1 maximal familiarization session in non-resistance-trained recreationally active men. Analysis of the current data suggests that it is appropriate to perform both 1RM-SS and 1RM-BLS tests within the same testing session if 1RM-SS is performed before 1RM-BLS. However, further testing is warranted to firmly establish the effects of 1RM-SS on subsequent 1RM-BLS.