High agreement between laboratory and field estimates of critical power in cycling

The purpose of this study was to investigate the level of agreement between laboratory-based estimates of critical power (CP) and results taken from a novel field test. Subjects were fourteen trained cyclists (age 40±7 yrs; body mass 70.2±6.5kg; VO_2max 3.8±0.5L·min ^-1). Laboratory-based CP was estimated from 3 constant work-rate tests at 80%, 100% and 105% of maximal aerobic power (MAP). Field-based CP was estimated from 3 all-out tests performed on an outdoor velodrome over fixed durations of 3, 7 and 12min. Using the linear work limit (W_lim) vs. time limit (T_lim) relation for the estimation of CP1 values and the inverse time (1/t) vs. power (P) models for the estimation of CP2 values, field-based CP1 and CP2 values did not significantly differ from laboratory-based values (234±24.4W vs. 234±25.5W (CP1); P<0.001; limits of agreement [LOA], -10.98-10.8W and 236±29.1W vs. 235±24.1W (CP2); P<0.001; [LOA], -13.88-17.3W. Mean prediction errors for laboratory and field estimates were 2.2% (CP) and 27% (W′). Data suggest that employing all-out field tests lasting 3, 7 and 12min has potential utility in the estimation of CP.