Human endurance performance can be predicted from maximal oxygen consumption (Vo(2max)), lactate threshold, and exercise efficiency. These physiological parameters, however, are not wholly exclusive from one another, and their interplay is complex. Accordingly, we sought to identify more specific measurements explaining the range of performance among athletes. Out of 150 separate variables we identified 10 principal factors responsible for hematological, cardiovascular, respiratory, musculoskeletal, and neurological variation in 16 highly trained cyclists. These principal factors were then correlated with a 26-km time trial and test of maximal incremental power output. Average power output during the 26-km time trial was attributed to, in order of importance, oxidative phosphorylation capacity of the vastus lateralis muscle (P = 0.0005), steady-state submaximal blood lactate concentrations (P = 0.0017), and maximal leg oxygenation (sO(2LEG)) (P = 0.0295), accounting for 78% of the variation in time trial performance. Variability in maximal power output, on the other hand, was attributed to total body hemoglobin mass (Hb(mass); P = 0.0038), Vo(2max) (P = 0.0213), and sO(2LEG) (P = 0.0463). In conclusion, 1) skeletal muscle oxidative capacity is the primary predictor of time trial performance in highly trained cyclists; 2) the strongest predictor for maximal incremental power output is Hb(mass); and 3) overall exercise performance (time trial performance + maximal incremental power output) correlates most strongly to measures regarding the capability for oxygen transport, high Vo(2max) and Hb(mass), in addition to measures of oxygen utilization, maximal oxidative phosphorylation, and electron transport system capacities in the skeletal muscle.