To assess the between-game variation in measures of physical performance during 11 v 11 soccer match-play, over a short period of time, in highly trained youth soccer players. A single cohort observational study design was employed. Physical match performance data were collected from 17 male, highly trained youth soccer players (age: 13.3 ± 0.4 y) over three, 2 x 20min, 11 v 11 matches. Using 10 Hz GPS, the variables selected for analyses were total distance (TD), high-speed running (HSR), very high-speed running (VHSR), number of high-speed running efforts (HSReff) and number of very high-speed running efforts (VHSReff). Match data was also separated into cumulative 5 min epochs, to identify the peak 5 min epoch and the mean of the cumulative 5 min epochs for each match. Variability was quantified using the coefficient of variation (CV), Standard error of measurement (SEM) and intra-class correlation coefficient (ICC). Between- and within-player smallest worthwhile changes (SWC) were also calculated for each variable to aid in the interpretation of the data. Analysis of the variance between games reported a low CV for TD (3.8%) but larger CVs for HSR (33.3%), HSReff (35.4%) and VHSR and VHSReff (59.6 and 57.4 %, respectively). Analysis of 5 min epochs (peak and average) found an increase in the CVs beyond that of the values reported for the whole match. Between-player SWC in high intensity physical performance data ranged from 24.7 – 42.4 %, whereas within-player SWC ranged from 1.2 – 79.9%. The between-game variability of high and very high intensity activities in youth soccer players, across three soccer matches over a short period of time (2 weeks), is relatively ‘large’ and specific to the individual, thus highlighting the need for caution when interpreting physical performance data between games and players.