ULTRA SCIENCE
Threshold assessment and critical speed
Define which threshold you mean before turning a result into a training zone.
Different tests, different estimates
Lactate and ventilatory tests identify different physiological markers; critical speed models the speed-duration relationship. They are not automatically the same boundary. Threshold terminology is inconsistent, so retain the exact definition with every result.
First decide whether your question concerns the transition out of easy exercise, the boundary near harder sustained work, or how to anchor a particular workout. A report labelled only “threshold” leaves this ambiguous. Ask for the name of the marker, the associated speed or workload and how it was identified. Avoid translating labels between laboratories or watches without checking their definitions.
Choose and document a protocol
For laboratory assessment, ask which lactate or ventilatory definition, stage length and interpretation will be used. For critical speed, work with a qualified coach on a published time-trial protocol, sufficient recovery and accurate distances. Keep raw times and model fit. Do not substitute one hard run or two arbitrary watch points for a validated protocol.
Lactate testing uses repeated blood samples during a specified exercise progression; it requires competent sampling and interpretation. Ventilatory testing uses respiratory measurements and requires the provider to identify and explain the selected breakpoint. Critical-speed assessment instead fits performance over different durations. Ask the coach which published protocol is appropriate, why its efforts are selected and how recovery will be managed. A field result is an estimate, not a substitute laboratory sample.
Validity, reliability and error
Field critical-speed testing can be repeatable under suitable conditions. Pacing, effort, rest and trial selection matter. Correlation with performance does not prove individual agreement with a physiological threshold; request protocol-specific uncertainty rather than assuming an exact boundary.
Record surface, gradient, conditions, device, footwear and how each effort was completed. A distance measured poorly or an effort stopped early can distort a performance model. A model that fits a small number of points neatly does not by itself establish physiological validity. When retesting, changing the test design at the same time as training makes the source of any difference harder to identify.
Apply cautiously
Use results to check training-intensity descriptions against perceived effort and repeated sessions. Do not treat critical speed as a sustainable ultra pace or a fresh threshold as unchanged late in a race. Avoid forcing a standard percentage correction on every runner.
Translate the report into a small number of usable effort descriptions with your coach, including what to do on climbs, in heat and when tired. If the recommended workload repeatedly feels inconsistent with the intended session, review the interpretation rather than forcing compliance. Critical speed may be worth measuring when a coach needs a performance anchor; it is not necessary just to obtain another pace number.
Who, costs and frequency
Worth considering when training zones remain confusing or a coach will use the result. Unnecessary if effort-based training already works and nothing would change. Labs cost more; field trials cost recovery and need safe maximal-effort eligibility. Repeat after an appropriate block, with the same protocol; not as an extra weekly hard session.
Choose the least burdensome assessment that answers your question. Beginners, runners returning from illness or injury, and anyone with symptoms should not add maximal trials without appropriate assessment. A qualified provider can decide whether testing should be modified or postponed. Keep the original report so future comparisons preserve the definition, protocol and uncertainty rather than only a rounded heart-rate number.
Use this with
Evidence and limits
Evidence: Moderate. Relevant to threshold interpretation; indirect to a specific ultra training prescription. Older foundational review; cannot validate current consumer algorithms.
Version and review
AI-assisted evidence appraisal is not credentialed clinical review. Publication authorization and application release dates are separate from scientific review.
Evidence markers identify statements and their limits; practical examples are not validated individual prescriptions.
Copy the site guide and this page. Paste into your AI and ask naturally.