Missing heart rate does not necessarily mean missing training

The run happened, and the watch saved the time and distance, but heart-rate data was absent or unusable. The session still belongs in your log. Some apps can also use manually entered perceived effort to provide training-load feedback. This can be a valuable addition if you understand what the input means. It does not restore a missing sensor signal; it records a different kind of personal observation.

The current Strava Relative Effort description identifies inputs from either heart-rate data or a Perceived Exertion self-rating. The feature is available for subscribers’ activities with the required data, so perceived effort for a session can play a role without a sensor. The resulting score belongs to the app’s own system; it is not a training-load unit shared by every sports platform. Source: Strava Relative Effort.

The useful question is not whether manual data is less “real”, but what it observes. A heart-rate series provides physiological data that changes over time, with its own measurement conditions. A retrospective overall impression describes how intense the whole task felt. The two sources may be related, but they are not identical. Keeping this distinction clear allows a manually supplemented session to remain interpretable in later weekly comparisons.

What does Strava’s self-rating scale ask?

The Strava Perceived Exertion description asks you to rate perceived intensity on a scale from one to ten, from easy to maximal effort. You can enter it when saving or editing an activity. The system also lets you choose it instead of heart rate as the input for associated feedback on a particular session. The 18 September guide mentions this option for missing or inaccurate heart-rate data too. Source: Strava Perceived Exertion.

Do not enter the number as a grade for the success of the workout. A run planned to be easy and experienced as easy is not worse for receiving a low rating. Nor is a session that felt hard automatically better. The scale asks about experienced intensity, rather than how proud you are of the final pace or how well the social post turned out. Consistent meaning matters more than an impressively high weekly score.

In your own routine, first think through the whole task, then consider which description fits it. Do not automatically promote the single hardest second into a rating for the entire session. Follow the specific app’s guidance, because numbers on different scales do not necessarily have identical meanings. Do not copy a familiar rating from another system without explanation. Manual data becomes useful when you repeatedly answer the same question with it.

A woman runner reviews her notebook beside a café window to look for connections in her running log.Enlarge image
AI-generated editorial illustration of reviewing a running log; not a documentary photograph.

An overall impression does not preserve the run’s internal timeline

In an illustrative situation of our own, two runs both last forty minutes and receive the same overall self-rating. One is steady; the other contains several short fast sections with easier running between them. The overall impression may be similar while the pattern over time differs. A single manual number does not identify the minute in which the harder section occurred. Preserve that through the session description or any detailed data available.

The FIT Activity file description treats individual records, laps and summaries as separate data. This is a reminder of what a single retrospective value lacks: the sequence of detail. If you saved a session without heart rate, a manual effort rating does not fill the missing heart-rate field with real measured points. The file and subjective note can be interpreted together, but they describe different things. Source: Garmin FIT Activity.

Briefly identify the structure in your log: for example, a steady run, a hilly loop, several short repetitions or a route with many stops. Every manual entry does not need a detailed study. A few precise words protect you from later assuming identical session histories from identical totals. A training-load score is one of the system’s summaries, while the actual content of a session is built from several pieces of information.

Do not invent the formula from the displayed score

Strava describes the purpose and inputs of Relative Effort, but that does not justify inventing a precise internal formula. A number on the manual scale is not automatically a simple multiplier from which the contribution of every minute can be reconstructed. Treat the displayed total as the result of the manufacturer’s own algorithm. If another app also shows a training-load number, the shared word “effort” or “load” does not prove that the calculation is identical.

In an index devised solely for illustration, someone could multiply duration by a manual intensity rating. Thirty minutes rated four would give 120 units, as would sixty minutes rated two. This is not the Strava Relative Effort formula or an exercise prescription. In the example, two different tasks produce the same result. It helps show how a summary number can necessarily conceal details even when the arithmetic is correct.

If you use an index like this in your own log, identify the formula and scale, and keep the app’s result separate. Do not relabel a simple invented index as a manufacturer score. Retaining the original values allows later interpretation too. When changing platforms, it is particularly important not to silently join scores produced by different formulas into one continuous progress graph.

Your recording habits also shape weekly comparisons

If you give every run a self-rating one week but only rate the hardest sessions the next, the data does not cover your actual training in the same way. Missing data does not prove rest or an easy day. Coverage therefore matters in your weekly review: how many sessions have an appropriate input, and which are missing? The system’s convenient summary cannot fully replace a personal observation that was never entered.

Strava’s weekly Relative Effort view shows the current week’s cumulative score and a suggested range based on a three-week average. This is a manufacturer reference, not a check of every tissue or your entire life situation. If your recording method or activity uploads change, the weekly picture may be built on a different basis too. Examine differing scores alongside their data origins first. Source: weekly Relative Effort graph.

A simple habit can support consistency: make a brief training note at a similar time and answer the same question. If you fill in many entries weeks later, mark them as recalled retrospectively. Strava’s current description says self-ratings can be added to earlier activities, but each must be edited individually. That technical possibility does not make a distant memory the same observation as a freshly recorded impression. Source: self-ratings for earlier sessions.

Heart-rate and manual inputs: keep their origins visible

One day you have a good heart-rate series; another day you use a manual value. This is not inherently poor logging, but it is worth marking the method change. Two different information sources may underlie the weekly score series. If one differs unusually from the other, do not immediately turn that difference into a performance change. First ask what changed in the input, duration, sport or interpretation of the overall impression.

In your own log, adding “heart rate”, “manual effort” or “uncertain sensor data” beside the entry may be enough. App interface options vary, so the note can be kept in a separate personal log too. You do not need to create false detail for missing data. The aim is to understand, even a few months later, why a run received that total and what data produced the assessment.

If heart-rate data is inaccurate, do not insist on retaining it as the primary input simply because “the sensor is objective”. A faulty measurement is a data-quality issue too. A manual rating does not automatically become more accurate because the other input is incomplete, however. Each needs to be treated in its own role: a documented sensor signal and a consistently recorded subjective observation. Choosing appropriately requires knowledge of the particular recording and the platform’s supported operation, rather than a universal ranking.

While reviewing training-log data, a runner writes a separate note about an uncertain measurement.Enlarge image
AI-generated editorial illustration of clearly marking suspicious sensor data; not a documentary photograph.

A research load metric is not a universal language for every demand either

Bonato and colleagues’ 2026 observational study followed twelve weeks of training by 48 recreational marathon and half-marathon runners. Internal load was analysed using eTRIMP, a metric linked to heart-rate zones. This does not validate Strava’s manual effort algorithm. The methodological example shows that a training-load number always has a specific definition and research purpose. Source: Bonato and colleagues.

Nor is an association observed in the study a general instruction to push your own score higher every week. The participants and method define the meaning of the result, and an association is not an individual causal prescription. If a number relates to performance in one sample, it still does not describe every need of every runner. Read the research paper in terms of its own method, rather than as automatic validation of an app graph.

Across different sports, an identical cardiovascular or overall-impression-based score does not mean an identical mechanical task for every local tissue. Cycling, running and strength training involve different movements. The log therefore needs the sport and a task description alongside the score. If you retain only a combined weekly total, much of the content is lost. The summary may help reveal a pattern, but it does not become a measure that replaces every form of load.

Do not let the score rewrite the session’s purpose

A high Relative Effort does not prove the value of an easy day. If the plan calls for relaxed activity, a lower score may be consistent with that under appropriate circumstances. Do not use the manual scale as a reward or a source of guilt. A log helps when it honestly describes the task and impression, rather than reshaping every outing to create a desired graph. Chasing points can change the question you originally collected data to answer.

In a weekly review, consider coverage, the origin of the input and actual session content together. If you see an unusual value, look for specific explanatory data: more time, another sport, a manual rating or a missing recording. It is also acceptable to conclude that no single cause can be proven. Do not turn your observations into an automatic, identical rule for increasing load every following week.

A manual effort rating is useful when it retains its subjective but consistently recorded nature. It does not need to masquerade as second-by-second heart rate to have value. It can add another perspective where sensor data is missing, and a system can summarise it in a defined way. Keeping the origin, scale and session structure together lets a training-load score remain an understandable signal without losing the actual run behind the log.

Sources

Helpful? Pass it on.