OBJECTIVE To determine the validity of field-derived mean maximum power (MMP) values for monitoring maximal cycling endurance performance. METHODS Twenty-seven male professional cyclists performed 3 timed trials (TTs) of 1-, 5-, and 20-minute duration that were used as the gold standard reference. Field-based power output data (3336 files; 124 [25] per cyclist) were registered during the preparatory (60 d pre-TT, including training data only) and specific period of the season (60 d post-TT, including both training and competitions). Comparisons were made between TT performance (mean power output) and MMP values obtained for efforts of the same duration as TT (MMP of 1-, 5-, and 20-min duration). The authors also compared TT- and MMP-derived values of critical power (CP) and anaerobic work capacity. RESULTS A large correlation (P < .001, r > .65) was found between MMP and TT performance regardless of the effort duration or season period. However, considerable differences (P < .05, standard error of measurement [SEM] > 5%) were found between MMP and TT values for all effort durations in the preparatory period, as well as for the derived CP and anaerobic work capacity. Significant differences were also found between MMP and TT of 1 minute in the specific period, as well as for anaerobic work capacity, yet with no differences for MMP of 5- and 20-minute duration or the derived CP (P > .05, SEM < 5%). CONCLUSIONS MMP values (for efforts ≥5 min) and the associated CP obtained from both training sessions and competitions can be considered overall accurate indicators of the cyclist's maximal capabilities, but specific tests might be necessary for shorter efforts or when considering training sessions only.