The purpose of this study was to assess 1) how treadmill slope variance affected external power output (PO) and propulsion technique reliability; and 2) how PO is associated with propulsion technique. Eighteen individuals with spinal cord injury performed two wheelchair treadmill exercise blocks (0% and 1% treadmill slope, standardized velocity) twice on two separate days. PO, velocity, and 14 propulsion technique variables were measured. In a follow-up study, N=29 performed wheelchair treadmill drag tests. Target and actual slope were documented and PO, intraclass correlation coefficients (ICC) and smallest detectable differences (SDD) were calculated. Within and between visits, the reliability study ICCs were perfect for velocity (1.0), weak for PO (0.33-0.46), and acceptable (>0.70) for five (0% slope) and 10 (1% slope) propulsion technique variables, resulting in SDDs of 35-196%. Measured PO explained 56-90% of the variance in key propulsion technique variables. In the follow-up, PO ICCs were weak (0.43) and SDDs high. Bias between target and actual slope appeared random. In conclusion, PO variability accounts for 50-90% of the variability in propulsion technique variables when speed and wheelchair set-up are held constant. Therefore, small Manuscript received xxx; revised xxx; accepted xxx. Date of publication xxx; date of current version xxx.This study was supported by the ICORD Visiting Scholar Grant.