The 3min all-out protocol captures both CP/FTP/“threshold” and work capacity above threshold (ie. FRC or W’) in the same protocol. So the threshold estimate will not should not be affected by the athlete’s work capacity or phenotype.
The idea behind the all-out effort is that it should fully deplete FRC/W’/“work capacity”, leaving the athlete with only aerobically produced power by the final 30sec of the test, where the threshold-estimate is taken from.
An athlete with higher FRC/W’/“work capacity” will perform more work (kJ) above threshold, but they should still reach their threshold by the end. Threshold would be over-estimated if the athlete paced the effort and conserved their work capacity such that they were still able to draw on FRC/W’/“work capacity” during the final 30sec of the test.
Note in the chart above the ‘end-test power’ and ‘work capacity’ on the right represent these two components.
If you have a big glycolitic capacity, could this test underestimate aerobic power? Ie if you create a big oxygen debt in the initial minutes of the test, wouldn’t the oxygen you are consuming be split between creating power, and repaying the oxygen debt?
Great question. I’m actually struggling to answer.
I’m pretty sure the answer to your first question is ‘no’.
But the answer to your second question is ‘yes’
Let’s try to reason it out.
Athlete A with a larger anaerobic capacity (W’/FRC/“work capacity above threshold”), but the same CP/FTP/“threshold” than Athlete B:
We would see a larger work capacity above threshold, expressed as more work (kJ). ie. pink area in the chart above.
More anaerobic resources depleted and metabolic byproducts accumulated means a larger disturbance to metabolic homeostasis.
Therefore more aerobic energy production capacity (greater component of VO2) occupied by trying to restore homeostasis, and therefore not available for direct locomotion.
We’re defining CP as the same, so the blue area is the same between athletes.
This implies the same VO2 attributable to external workload (locomotion/power output).
But Athlete A with a bigger ‘anaerobic hole’ to fill will have greater VO2 attributed to that, on top of locomotion?
Therefore Athlete A will show a higher VO2max at the same end-test CP?
What if Athlete A with the larger work capacity, and Athlete B with smaller work capacity both have the same VO2max?
Same as above we would see larger work (kJ) produced by athlete A (pink area) and a greater metabolic disturbance to homeostasis
At the end of the test they both have the same VO2max to work with, while Athlete A would have relatively more of that VO2max occupied by trying to restore homeostasis, while Athlete B has more left over for locomotion
Therefore Athlete A will show a lower CP at the same end-test VO2max?
This tracks with what I would expect for a more ‘anaerobic phenotype’ athlete: relatively lower CP/FTP/“threshold” at lower %VO2max, and lower metabolic efficiency (GE, economy). Tracks with VLamax & VO2max concepts. But I can’t quite mechanistically make the link, so not sure I have this correct.
I also wonder about underestimates of “FTP”. I think some people can have fairly high anaerobic contributions to their “aerobic” power over prolonged durations. If the 3 min max completely depletes their FRC, it will leave them with their pure aerobic power - but this would be lower as what they can usually sustain.
The other problem, at least with the last study, is that “all out” wasn’t even close to all out. The study shows:
Peak power (PP) in the 3-minute all-out test was measured at 738 ± 170 W, and CP was determined at 305 ± 32 W
738 watts peak power for someone with a 300+ watt threshold is not “all out”. At my measly 220FTP, I can put out an “all out” max effort >1,000 watts. And exceed 738 watts in a sprint at the end of a 5+ hour ride.
Results: CP estimated from the 3-min all-out approach was significantly higher than from the traditional approach (402 ± 33, 351 ± 27 W, P < .001), while W’ was lower (15.5 ± 3.0, 24.3 ± 4.0 kJ, P = .02). Five different combinations of 2 or 3 self-paced efforts led to CP estimates within the threshold of agreement, with only 1 combination deemed accurate for W’.
Conclusions: In elite cyclists the 3-min all-out approach is not suitable to estimate CP when compared with the traditional method. However, reducing the number of tests used in the traditional method lessens testing burden while maintaining appropriate parameter accuracy.
2nd Study:
Conclusions: EP and WEP obtained from a 3-min all-out test overestimate severe-intensity performance in competitive cyclists.
3rd Study from the abstract:
Significant differences were observed between EP and CP1 (+37 W, P<0.001), between WEP and W’1(-6.2 kJ, P=0.001), between EP and CP2 (+31 W, P<0.001) and between WEP and W’2 (-4.2 kJ, P=0.006). Average SEE values for EP-CP1 and EP-CP2 of 7.1% and 6.6% respectively were identified. Data suggest that using an isokinetic mode 3-min all-out test, while yielding a reliable measure of EP, does not provide a valid measure of CP.
Survey says: all 4 studies say this protocol isn’t a reliable method of estimating CP
right, but I don’t know if 3 minutes can actually do that. Rider X who can rider above FTP for >3minute…creates a problem? That is not an issue for a 20kj FRC rider
Wow, intervals.icu is sweet! Seems to include most of the cool stuff that Strava has put behind a paywall (and more).
I think it’s underestimating my FTP (eFTP is 30 watts less than most recent ramp test), but that’s probably because I haven’t done enough short all-out efforts?
