Source notes

A training-load score is useful only if it represents something relevant to the result you care about. The authors explain training load as exposure to exercise, separating work performed from the internal response that work produces. They argue that choosing a measure should begin with the intended outcome and the process thought to connect training with that outcome. Heart rate, running speed, duration, and perceived effort describe different aspects of a session. No label or formula makes one of them a universal measure of the stimulus responsible for every adaptation.

A central distinction is prediction versus cause. A score might predict later performance without representing something that would improve performance if deliberately changed. An association alone does not establish that connection. Likewise, combining intensity and duration into one total can conceal how the exposure was distributed. The value of a measure depends on the question, rather than on whether its calculation looks sophisticated.

How Does This Show Up in Real Life?

For a reader looking at a dashboard, the useful question is “What does this number actually tell me about my goal?” The authors use sprint development to illustrate the problem: heart rate may describe cardiovascular demand without adequately representing the muscle-and-nerve stimulus relevant to sprinting. A running measure might be a better candidate, but that link still needs evidence or a clearly stated hypothesis. This does not mean simple totals are useless. It means the same total can be adequate for one purpose and inadequate for another. The paper mainly changes how training data should be interpreted. It provides no universal target score, automatic rule for adjusting exercise, or verified threshold that guarantees performance improvement or freedom from injury.

The authors develop a framework rather than test a monitoring system against an alternative. Many relevant biological processes are difficult to measure during ordinary training, and identifying their causal role is challenging. The examples illustrate assumptions that need testing. The paper therefore cannot show that adopting its framework produces a particular improvement in performance or long-term health.

Why We Cite It

For Alpekin, the distinction between intermediate results and final outcomes creates a useful boundary for health claims. A change in strength or another measured capacity should not automatically be presented as proof of a broader health benefit. Alpekin could specify which outcome was actually assessed before connecting a result to a larger promise.

Treat this as a guide to asking better questions of training data, not as a new scoring method. Its practical value is intellectual caution: a measurement, a prediction, and a controllable cause are different things, even when they appear together on the same screen.