Prior to implementing a normalisation method, the standardisation and reliability of the method needs to be examined. This investigation aimed to assess the reliability of EMG amplitudes and test outputs from proposed normalisation methods for the triceps surae. Sixteen participants completed isometric (maximum and sub-maximum); isokinetic (1.05 rad/s, 1.31 rad/s and 1.83 rad/s) squat jump and 20 m sprint conditions, on 3 separate occasions over 1 week. The EMG data was collected from the medial and lateral gastrocnemius (MG and LG) and soleus (SOL). Log transformed typical error measurements (TEMCV%) assessed EMG signal and test output reliability across the three sessions. Only the squat jump provided acceptable EMG reliability for all muscles both between days (SOL: 13%; MG: 14.5%; LG: 11.8%) and between weeks (SOL: 14.5%; MG: 12.9%; LG: 8.9%), with the sprint only showing poor reliability in the LG between days (16.3%). Acceptable reliability for the isometric and isokinetic conditions were muscle and re-test period dependant. Reliable output was found for the squat jump (4.1% and 3.6%), sprint (0.8% and 0.6%) and 1RM plantar flexion test (2.8% and 3.5%) between days and weeks, respectively. Isokinetic plantar flexion displayed poor reliability at all velocities between days and weeks. It was concluded that the squat jump provides a standardised and reproducible reference EMG value for the triceps surae for use as a normalisation method.