This report aims to generate an evidence-based debate of the Critical Power (CP), or its analogous Critical Speed (CS), concept. Race times of top Spanish runners were utilized to calculate CS based on three (1500-m to 5000-m; CS 1.5-5km ) and four (1500-m to 10000-m; CS 1.5-10km ) distance performances. Male running world records from 1000 to 5000-m (CS 1-5km ), 1000 to 10,000-m (CS 1-10km ), 1000-m to half marathon (CS 1km-half marathon ), and 1000-m to marathon (CS 1km-marathon ) distance races were also utilized for CS calculations. CS 1.5-5km (19.62 km h −1 ) and CS 1.5-10km (18.68 km h −1 ) were different (p < 0.01), but both approached the average race speed of the longest distance chosen in the model, and were remarkably homogeneous among subjects (97% ±1% and 98% ±1%, respectively). Similar results were obtained using the world records. CS values progressively declined, until reaching a CS 1km-marathon value of 20.77 km h −1 (10% lower than CS 1-5km ).Each CS value approached the average speed of the longest distance chosen in the model (96.4%-99.8%). A power function better fitted the speed-time relationship compared with the standardized hyperbolic function. However, the horizontal asymptote of a power function is zero. This better approaches the classical definition of CP: the power output that can be maintained almost indefinitely without exhaustion. Beyond any sophisticated mathematical calculation, CS corresponds to 95%-99% of the average speed of the longest distance chosen as an exercise trial. CP could be considered a mathematical artifact rather than an important endurance performance marker. In such a case, the consideration of CP as a physiological "gold-standard" should be reevaluated.

This study aimed to predict the velocity corresponding to the maximal lactate steady state (MLSS(V)) from non-invasive variables obtained during an incremental maximal running test (University of Montreal Track Test, UMTT) and to determine whether a single constant velocity test (CVT), performed several days after the UMTT, could estimate the MLSS(V). During a period of 3 weeks, 20 male junior soccer players performed: (1) a UMTT, and (2) several 20-min CVTs to determine MLSS(V) to a precision of 0.35 km·h(-1). Maximal aerobic velocity (MAV) and velocity at 80% of maximum heart rate (V80%HRmax) were strong predictors of MLSS(V). A regression equation was obtained: MLSS(V)=(1.106·MAV) - (0.309·V(80%HRmax)) - 3.024; R2=0.60. Running velocity during CVT (V(CVT)) and blood lactate at 10 (La10) and 20 (La20) minutes further improved the MLSS(V) prediction: MLSS(V)=V(CVT)+0.26 - (0.812·ΔLa(20-10)); R2=0.66. MLSS(V) can be estimated from MAV and V(80%HRmax) during a single incremental maximal running test among a homogeneous group of soccer players. This estimation can be improved by performing an additional CVT. In terms of accuracy, simplicity and cost-effectiveness, the reported regression equations can be used for the assessment and training prescription of endurance in team sport players.

The aim of this study was to investigate whether the speed associated with 90% of maximal heart rate (S90%HRmax) could predict speeds at fixed blood lactate concentrations of 3 mmol·L(-1) (S3mM) and 4 mmol·L(-1) (S4mM). Professional team-sport players of futsal (n = 10), handball (n = 16), and basketball (n = 10) performed a 4-stage discontinuous progressive running test followed, if exhaustion was not previously achieved, by an additional maximal continuous incremental running test to attain maximal heart rate (HRmax). The individual S3mM, S4mM, and S90%HRmax were determined by linear interpolation. S3mM (11.6 ± 1.5 km·h(-1)) and S4mM (12.5 ± 1.4 km·h(-1)) did not differ (p > 0.05) from S90%HRmax (12.0 ± 1.2 km·h(-1)). Very large significant (p < 0.001) relationships were found between S90%HRmax and S3mM (r = 0.82; standard error of the estimates [SEE] = 0.87 km·h(-1)), as well as between S90%HRmax and S4mM (r = 0.82; SEE = 0.87 km·h(-1)). S3mM and S4mM inversely correlated with %HRmax associated with running speeds of 10 and 12 km·h(-1) (r = 0.78-0.81; p < 0.001; SEE = 0.94-0.87 km·h(-1)). In conclusion, S3mM and S4mM can be accurately predicted by S90%HRmax in professional team-sport players.