We have previously pointed out how the Anaerobic Threshold is important in order to adjust the intensities of training to the improvements in the performance of the athlete (see article on Measuring the AT). 

We have also observed that for each athlete, the AT value is related to the speed in a race, especially when climbing and time trialing: the power the athlete is developing in the race when going “all-out” is the so-called “CRITICAL POWER”. 

The latter is related to the duration of the effort that the athlete has to do: it will be higher for an effort of a few minutes, while for a performance of 30-40’ will be lower. 

I usually measure the critical power in training simulating an uphill time trial of 15'-30' or similar efforts on a flat course. 
It is also possible to check the critical power during a race, directly through a power meter device, or indirectly evaluating the VAM on the climb. 

The relationship between AT and critical power varies from athlete to athlete and depends, in my opinion, from muscle characteristics as well as training methods. 
On a climb of 30’ the majority of cyclist is able to express a critical power which is 10-30 watts higher than the AT value measured at 4mM/l with an incremental test. 

But a certain number of athletes (Armstrong was one of them) are only able to develop a critical power that is slightly inferior to the AT at 4 mM/l, independently from the level of fitness. 
Most likely these cyclists have a very high percentage of type I muscle fibers, aerobic ones, that are not able to tolerate elevated lactic acid concentration. 

In any case for every athlete there is a close correlation between AT and critical power: if the AT improves, also the critical power that the athlete can maintain increases. 
Such behavior is relatively constant for every cyclist, even though it is possible to have an influence on it with training and the habit of performing all-out efforts. 

Usually I start evaluating the critical power after at least 8-10 weeks of specific training, when the organism has properly adapted to certain training intensities. 
During the racing season it is useful, if not essential, to measure the critical power (as well as the AT) at the end of a long 4-5 hours training session, so to simulate typical racing conditions. 

The available glycogen sources in liver and muscles will determine the ability of expressing a good critical power in the crucial phases of the race, which usually are close to the finish. 

Glycogen availability will once again depend on a high value of AT so to face the first difficulties of the race event with a more efficient energetic consumption: if the cyclist will be close to his critical power already on the first climbs, he will burn much more glycogen than a cyclist climbing with an effort below his anaerobic threshold.