The aim of this study was to analyze changes in gait variability and symmetry in distance runners. Fourteen competitive athletes ran on an instrumented treadmill for 10,000 m at speeds equivalent to 103% of their season's best time. Spatiotemporal and ground reaction force data were recorded at 1500, 3000, 5000, 7500 and 9500 m. Gait variability and inter-leg symmetry were measured using median absolute deviation (MAD) and the symmetry angle, respectively. There were no overall changes during the running bout for absolute values, symmetry angles or variability, and there were only moderate changes in variability between successive testing distances for three variables. Even with these few changes, variability was low (<4%) at all distances for all variables measured and, on average, the athletes were symmetrical for five of the seven gait variables measured. There were greater mean asymmetry values for flight time (1.1-1.4%) and for impact force (2.0-2.9%), which might have occurred because of muscle latency as the lower limb responded passively to impact during initial contact. Although most athletes were asymmetrical (>1.2%) for at least one variable, no one was asymmetrical for more than four of the seven variables measured. Being asymmetrical in a few variables is therefore not abnormal and not indicative of asymmetrical gait and given many practitioners analyze symmetry (and variability) on an individual, case-study basis, caution should be taken when assessing the need for corrective interventions.