Lactate Threshold: Why Watches, Formulas and the Lab Give Different Zones

The key points in 30 seconds

• A new systematic review included 70 running studies and found great heterogeneity in definitions, protocols and reference methods.
• MLSS, IAT and OBLA describe a similar region of intensity but are not interchangeable measurements of LT2.
• A zone estimated by a watch is a useful estimate, not a direct measurement of a physiological threshold.
• For tracking progress, repeating the same protocol matters more than comparing numbers obtained with different methods.

Why a runner needs the lactate threshold

Almost every running programme divides intensity into zones: easy, moderate, threshold, interval. The boundaries between them are often tied to lactate thresholds. The idea is simple: below a certain speed the body copes with lactate, above it lactate starts to accumulate faster and sustained running becomes harder and harder. Knowing this boundary helps to choose a pace for long sessions and to judge whether easy days are not turning out too hard.

The problem is that the "lactate threshold" has no single definition. The name covers different criteria, different protocols and different ways of calculation — and this is exactly what the new review examined.

What LT1 and LT2 are

In a stepwise test, running speed is raised at equal intervals, and between steps a drop of blood is taken to measure lactate. At first its concentration stays relatively stable and then starts to rise noticeably. The first lactate threshold, LT1, is linked to the first sustained rise above baseline. The second, LT2, corresponds to a higher intensity beyond which lactate accumulation accelerates and a prolonged steady state becomes increasingly unlikely.

These are not two perfectly sharp "switches". The result depends on stage duration, the rate of load increase, where and when blood is sampled, the mathematical criterion and the athlete's condition: recent training, nutrition, fatigue.

Why the methods give different numbers

The review by Lu and colleagues classified methods as measured, calculated and predictive. LT1 was determined by the first rise, a breakpoint, a fixed concentration or a deviation from baseline. For LT2 several criteria were used:

• maximal lactate steady state (MLSS) — the highest load at which lactate still does not rise without limit;
• the individual anaerobic threshold (IAT), calculated from the shape of a particular person's curve;
• the fixed 4 mmol/L mark — OBLA;
• other mathematical criteria.

MLSS requires several long constant-load tests on different days and is physiologically well linked to metabolic stability. But even its protocols differ, so the review rated confidence as only moderate. OBLA is simpler, yet an identical 4 mmol/L ignores the individual shape of the curve: in one person a sharp rise in lactate may begin before this mark, in another after it. IAT is individualised but exists in several calculation variants.

Watches, formulas and predictive models

Sports watches do not take blood. They estimate the threshold from pace, heart rate, heart rate variability or results of past races. Such estimates belong to predictive models — and the quality of their validation turned out to be a weak point of the review.

Among the 18 studies that developed predictive models, the risk of methodological bias was high in every case; among 23 studies of models or devices, overall risk was high in 65.2%. This does not mean watches are useless. It means that the accuracy of a particular formula cannot be transferred to any population and protocol: a formula that worked well in one group of runners in one laboratory may be wrong in another.

How to read this information

When you see "threshold 4:30 per kilometre" in a watch report or a laboratory report, it is useful to ask three questions. By what method was this number obtained — MLSS, IAT, a fixed 4 mmol/L or a formula? What protocol was used — step length, speed increment, type of track or treadmill? How reproducible is the number — do two tests with the same protocol give a similar result? Without answers, comparing two thresholds obtained in different ways is like comparing temperatures on different scales.

Practical meaning

For everyday training, calculated zones can be a convenient starting point. They are best combined with pace, heart rate, perceived effort and the ability to talk. A laboratory test is useful when a decision truly needs greater precision — for example, preparing for a specific race — but its result should always be read together with the name of the method.

When reassessing, use comparable conditions: the same protocol, similar preparation, the same equipment and the same calculation method. A difference between two methods does not necessarily mean worse or better fitness.

What the review does not show

The review does not say which method is "best" for practice and does not give a universal correction between methods. It does not assess whether training by zones calculated in different ways improves race results. It is devoted specifically to the methodology of determining thresholds in running — its results cannot be transferred automatically to other sports.

How convincing is the evidence

The strength of the review is that it systematises 70 studies across several decades. Its main weakness is also its main conclusion: definitions, references and analytical procedures differ too much for one universal correction. The PROSPERO registration was retrospective — made after searching and screening had begun.

Sport in Luxembourg conclusion

A training zone is a model of reality, not reality itself. MLSS looks the most physiologically grounded reference among the LT2 criteria considered, but it is not an infallible "gold standard". Watches are not necessarily wrong; they give an estimate whose accuracy depends on the formula and the user.