Transcranial magnetic stimulation (TMS) is a non-invasive brain stimulation technique used to study human neurophysiology. A single TMS pulse delivered to the primary motor cortex can elicit a motor evoked potential (MEP) in a target muscle. MEP amplitude is a measure of corticospinal excitability and MEP latency is a measure of the time taken for intracortical processing, corticofugal conduction, spinal processing, and neuromuscular transmission. Although MEP amplitude is known to vary across trials with constant stimulus intensity, little is known about MEP latency variation. To investigate MEP amplitude and latency variation at the individual level, we scored single-pulse MEP amplitude and latency in a resting hand muscle from two datasets. MEP latency varied from trial to trial in individual participants with a median range of 3.9 ms. Shorter MEP latencies were associated with larger MEP amplitudes for most individuals (median r = − 0.47), showing that latency and amplitude are jointly determined by the excitability of the corticospinal system when TMS is delivered. TMS delivered during heightened excitability could discharge a greater number of cortico-cortical and corticospinal cells, increasing the amplitude and, by recurrent activation of corticospinal cells, the number of descending indirect waves. An increase in the amplitude and number of indirect waves would progressively recruit larger spinal motor neurons with large-diameter fast-conducting fibers, which would shorten MEP onset latency and increase MEP amplitude. In addition to MEP amplitude variability, understanding MEP latency variability is important given that these parameters are used to help characterize pathophysiology of movement disorders.