Reliability, variability, and validity of stroke rate measured via a Digitrainer accelerometer and GoPro 60 Hz video camera were determined. Six elite New Zealand kayakers (three males, three females) performed three trials of 300-m sprints in a single kayak (K1) mounted with the Digitrainer and camera. Average individual within-trial reliability and between-trial reliability and variability were calculated using data from 40 strokes beginning at 200-m. Both Digitrainer and video showed good reliability (Mdiff% =5%; ES =0.6), and moderate variability (ICC <0.67; TE% <10%). There was good agreement between Digitrainer and video stroke rates (r=0.86, p=0.000), however the Digitrainer overestimated stroke rate by 4 ±5 spm. Both systems can assess relative change in stroke rate, however video should be used when valid stroke rates are required.