The blood lactate profile (HLa-P) is an accepted method of evaluating athletes and providing a basis for the prescription of training intensity. For both logistic and public health reasons HLa-P is less than optimal. In this study we evaluate the relative velocity or the %HR-max, obtained during a training session, as alternatives to HLa-P. Competitive speed skaters (N = 20) performed HLa-P consisting of 5.2000 m/400 m at incremental velocities ranging from very slow to maximal (time = 3.0-5.0 min). Blood lactate measured during a 60-s interval following each repetition was used to construct HLa-P and to predict the velocity associated with steady state (HLa = 4.0-6.5 mmol.l-1). Relative velocity was calculated relative to the velocity of the maximal trial. A plot of relative velocity and %HRmax vs HLa demonstrated that HLa = 4.0-6.5 mmol.l-1 occurred at a relative velocity of 78-88% (R2 = 0.807) and at 84-92 %HRmax (R2 = 0.748). In a separate training session the relative velocity and %HRmax models were cross validated by having the subjects skate 9.2000 m/400 m at constant velocity. HLa changes during the training session defined the presence/absence of steady state (delta HLa < 1.0 mM from trial 3 to 9). Comparing the velocity during the training session vs the velocity predicted from HLa-P, relative velocity model and %HRmax model allowed a test of the accuracy of bloodless means of defining steady state. HLa-P correctly predicted 81% of training session HLa responses, the relative velocity model correctly predicted 78%, and the %HRmax model correctly predicted 68%.(ABSTRACT TRUNCATED AT 250 WORDS)