Sixteen female games players completed 15 min of intermittent exercise on a treadmill. The speed and incline of the treadmill were varied to elicit a similar heart rate response to that found during women's hockey match‐play. The mean heart rate response was 169 ± 8 beats min^‐1 and the mean work:rest ratio was 1:1.7. Oxygen uptake was estimated by applying the oxygen cost determined from an individual's heart rate‐oxygen uptake regression equation to the average heart rate each 5 s during the intermittent exercise. A second esimation of oxygen uptake was made from a time‐motion analysis by assigning a specific oxygen cost to each discrete movement during the intermittent exercise protocol. The true oxygen cost was calculated from the analysis of the expired gas. The heart rate analysis overestimated the true cost by a mean of 4.3 ± 5.3% (0.09 ± 0.11 1 min^‐1), and the time‐motion analysis underestimated the true value by a mean of 15.7 ± 6.3% (0.34 ± 0.15 1 min^‐1). There was a significant difference (P < 0.01) between the true value and both the estimates. It was concluded that heart rate analysis gives a much better indication of oxygen uptake during intermittent activity than time‐motion analysis.