The critical power test is a well-established procedure that provides
estimates of two important parameters characterizing work performance;
anaerobic work capacity (AWC) and critical power (CP). The concept pr
oscribes a hyperbolic relationship between power output (P) and time t
o exhaustion (t), given by (P - CP)t = AWC. Since evidence now exists
that the procedure overestimates CP and underestimates AWC, this study
was undertaken to investigate the effect of relaxing the requirement
that the time asymptote necessarily be at zero. Using data from a prev
ious study, it is shown that in so doing, (1) a time asymptote signifi
cantly less than zero is obtained, (2) significantly smaller estimates
of CP and larger estimates of AWC are obtained, (3) a third parameter
is introduced that theoretically represents maximal instantaneous pow
er, (4) it implies that the maximal power that could be developed at a
ny instant is proportional to the amount of AWC remaining at that inst
ant, which in turn implies that (5) at exhaustion not necessary all of
AWC is consumed.