This study examined the suitability of the Wingate test as a means of assessing the maximal oxygen deficit (MOD), and the influence of the anaerobic capacity on the fraction of the MOD used during a 30- and a 45-s Wingate test in 19 male subjects. The MOD incurred in constant-intensity supramaximal exercise was higher (P < 0.01) than that for the 45-s and 30-s Wingate tests [68.6 (3.4) vs 60.9 (2.2) and 53.7 (1.6) ml x kg(-1), respectively], corresponding to a 10% higher value for the 45-s compared to that for the 30-s test (P < 0.001). A close correlation was found to occur between MOD and the oxygen deficit incurred during the 30- and 45-s Wingate tests, as well as between both all-out tests (r = 0.86-0.90; P < 0.001). The oxygen deficit accumulated during the first 30 s of the 45-s Wingate test was similar to that accumulated during the 30-s Wingate test. The intraclass correlation coefficient for the oxygen deficit after 30 s of all-out exercise (two treatments) was 0.96. The higher the MOD the lower was its fractional recruitment during the 30-s (r = -0.88, P < 0.001) and during the 45-s (r = -0.74, P < 0.01) Wingate tests. In conclusion, 80-90% as an assessment of the oxygen deficit incurred during a Wingate test is valid as an estimate of the anaerobic capacity. The fraction of the anaerobic capacity used in a 30- and 45-s all-out test is inversely related to the anaerobic capacity.