In 25 male cross-country (XCO) mountain bikers, intermittent field tests ridden straight after a 140-min fatiguing protocol (with repeated efforts at 105-110% of critical power) predicted a later XCO time trial better than the same tests ridden fresh (fatigued 10 s:20 s test in W/kg: 89% of the variance), and power after fatigue fell about 15-20% in the slower riders against about 6-10% in the faster ones; the predictions and the faster/slower split are correlational

Inoue and colleagues asked three things about Olympic cross-country mountain biking: whether intermittent field tests predict XCO performance better when ridden fatigued, whether a long fatiguing ride lowers power in those tests, and whether holding power under fatigue separates better from worse riders.

Twenty-five male XCO riders did three intermittent formats (30 s on / 15 s off, 10 s on / 20 s off, 3 min on / 2 min off) under fresh and fatigued conditions in random order. In the fatigued condition each test followed straight on from a 140-min ride containing repeated efforts slightly above critical power (105-110%).

An individual XCO time trial 72 h after the last session served as the performance criterion and split the riders into high and low performers. Every test correlated with time-trial performance (r = -0.53 to -0.95), more strongly when fatigued; the fatigued 10/20 test in W/kg was the best single predictor, explaining 89% of the variance.

Power fell in all three tests after the fatiguing protocol, by about 15-20% in low performers and about 6-10% in high performers. The authors conclude that sustaining power in repeated efforts under fatigue matters for XCO time-trial performance and competitive level.