Transcranial magnetic (TMS) and motor point stimulation have been used to determine voluntary activation (VA). However, very few studies have directly compared the two stimulation techniques for assessing VA of the elbow flexors. The purpose of this study was to compare TMS and motor point stimulation for assessing VA in non-fatigued and fatigued elbow flexors. Participants performed a fatigue protocol that included twelve, 15 s isometric elbow flexor contractions. Participants completed a set of isometric elbow flexion contractions at 100, 75, 50, and 25% of maximum voluntary contraction (MVC) prior to and following fatigue contractions 3, 6, 9, and 12 and 5 and 10 min post-fatigue. Force and EMG of the bicep and triceps brachii were measured for each contraction. Force responses to TMS and motor point stimulation and EMG responses to TMS (motor evoked potentials, MEPs) and Erb's point stimulation (maximal M-waves, Mmax) were also recorded. VA was estimated using the equation: VA% = (1−SITforce/PTforce) × 100. The resting twitch was measured directly for motor point stimulation and estimated for both motor point stimulation and TMS by extrapolation of the linear regression between the superimposed twitch force and voluntary force. MVC force, potentiated twitch force and VA significantly (p < 0.05) decreased throughout the elbow flexor fatigue protocol and partially recovered 10 min post fatigue. VA was significantly (p < 0.05) underestimated when using TMS compared to motor point stimulation in non-fatigued and fatigued elbow flexors. Motor point stimulation compared to TMS superimposed twitch forces were significantly (p < 0.05) higher at 50% MVC but similar at 75 and 100% MVC. The linear relationship between TMS superimposed twitch force and voluntary force significantly (p < 0.05) decreased with fatigue. There was no change in triceps/biceps electromyography, biceps/triceps MEP amplitudes, or bicep MEP amplitudes throughout the fatigue protocol at 100% MVC. In conclusion, motor point stimulation as opposed to TMS led to a higher estimation of VA in non-fatigued and fatigued elbow flexors. The decreased linear relationship between TMS superimposed twitch force and voluntary force led to an underestimation of the estimated resting twitch force and thus, a reduced VA.