CP is evaluated as a static threshold — but it moves in an unpredictable way during a race or a training session.

With MMPs, on the other hand, we always know the margin between the current effort and the best recorded performance.

This makes MMPs a practical and robust reference for real-time and post-hoc analysis, regardless of fatigue state.

What if, instead of assessing all the margins in isolation, we drop all the moving averages and show only the one closest to its own MMP limit?

At any moment during a race or training, this single value tells you which duration is most critically loaded relative to the athlete's best performance.

It provides a compact, interpretable signal that captures the dominant physiological stress at each point in time.

We can never be sure the profile is truly maximal.

The main limitation of the power profile approach is that we do not know how "real" it is — i.e. whether the recorded efforts were truly maximal and representative of the athlete's capacity.

Margins are always computable — and informative.

When MMPs are used to prescribe training or post-process a session, we can always evaluate the margin between the current race and the best recorded performance. The MMP with the smallest margin may also tell us about the dominant exercise limitation at that moment.

The approach is model-free and practical.

No assumptions about the shape of the power-duration curve are needed. Any training software can compute rolling averages — making this method accessible and easy to apply in real-world settings.