Two obvious limitations explaining the discrepancy of between model predictions and data are 1) VO2, La, PCr kinetics are not accounted for (thus energy system contributions and rate of W' depletion change with time), and 2) W' is in reality not one system but has a component from each.
But regardless of averaging, yes, most people "fail" with quite a bit of sprint capacity left. Psychology and decision making is not part of the MPA-model... :) Assessing the model using points with partial W' depletion (e.g., 50%) might be more informative. Example of representative data.
Accurate, although it depends on the duration of the sprint (e.g., 5s is repressed much more than 1s under W' depletion). I'm currently collecting experimental data on this. Strictly speaking Pmax (and thus MPA) is the best power over an indefinitely short duration. But averaging matters in practice
The three-dimensional impulse-response model uses three training load metrics as inputs and three performance metrics as outputs. These metrics are represented by a three-parameter CP model and reflect the stress imposed on each of the three systems. journals.plos.org/plosone/arti...
How to quantify the training load of higher intensities and sprints, separately from aerobic training load, simply from power data? We suggest a 3D strain score (alternative to the "1D" TSS/TRIMP/work -> fitness paradigms) to capture the nuances of training CP, W', and Pmax.
Thanks for your interest and suggestion! It is possible we include a code in the updated version of the preprint and/or the final, published paper.
Also, looking forward to hearing any feedback if you read it - trying to make it ready for peer review. It's a rather theoretical model, cutting some corners, so I'm curious how it's received amongst exercise physiologists.
Yes, I think their system has a strong rationale and deserves to be put under scientific scrutiny! More (experimental) studies to come in the future, I hope 🤞
We suggest that quantifying load using 3 dimensions rather than one better predicts adaptations specifically in CP, W', and Pmax. Using a global amalgamation of load such as total work, TRIMP, or TSS inevitably leads to more uncertainty about what exact attributes training is targeting.
Accumulating work at different intensities relative to CP, Pmax, and remaining W' results in somewhat distinct adaptive responses, which should be taken into account when trying to model training and predict performance gains.
Our new preprint proposes a new way to quantify training loads based on power output. We describe how a 3-parameter CP model may be used to estimate contributions of each energy system, resulting in a "strain score" which overcomes some limitations of other TL metrics. doi.org/10.48550/arX...
Tom's Weekend Warrior study is out! Training either 4x (short) or 2x (long) weekly resulted in the same adaptations, when overall volume and content of training was fixed.
Performance, thresholds, efficiency, Hbmass, and ox cap ↑ equally in both groups.
onlinelibrary.wiley.com/doi/full/10....
