You’ve been exercising regularly. The scale hasn’t moved much, your watch has given you three different fitness scores, and yesterday’s session felt harder than last week’s. Is anything working?
The most useful answer comes from a few repeatable observations. Choose something you want to do better, record a starting point, and look for a pattern under reasonably similar conditions. You might do more of the same work, find it easier, or do it with greater control. Each tells you something. None tells you everything.
What are you trying to improve
Start by separating four related questions. Fitness concerns capacities such as strength and endurance. Performance is what you achieve in a particular task on a particular day. Body composition describes what your body is made of, including fat and fat-free tissues. Health is broader still. A faster walk cannot establish that your blood pressure is healthy, and a change in body weight cannot tell you how much stronger you are. [1, 9]
Even these categories overlap. Body composition is commonly included among health-related fitness attributes. The point is to name the outcome instead of allowing one convenient number to stand in for all of them. Steps and completed sessions record activity; they don’t directly measure capacity. [1]
Try completing this sentence: “I want to be better able to…” Perhaps it’s walking or wheeling around the neighbourhood, carrying shopping, or performing a familiar exercise. Choose one main measure that fits that goal, plus another perspective if useful. You don’t need a home laboratory or a score for every part of your body.
Make the comparison fair
A starting measurement is a baseline. It becomes more useful when you know how it was obtained. Record the exercise or route, equipment, resistance, repetitions or duration, and how hard it felt. Note an unusual circumstance such as illness, a hot day or a particularly tiring session beforehand.
Keep the important details similar when you compare results. That might mean the same walking route and surface, the same chair, or the same exercise technique and rest between sets. If you change the conditions, record the change. A heavier lift with a shorter range of movement answers a different question from the original lift.
Give yourself a chance to learn an unfamiliar task before treating its score as a firm starting point. In a 2011 Journal of Strength and Conditioning Research study, 30 men completed four one-repetition-maximum testing sessions. Those without previous resistance-training experience improved their bench-press and squat test scores across sessions; experienced men were more stable. A better score can partly reflect getting better at the test. [2]
That learning is useful. It simply means the result cannot tell you exactly how much muscle or general strength you gained. For everyday tracking, a few comparable observations are more informative than making the very first attempt your permanent reference.
Notice strength in familiar work
A training log can provide useful evidence without a maximum-effort test. Suppose you perform a familiar row with the same resistance, movement and number of repetitions. Over several sessions it feels more manageable. Or you can use a little more resistance while keeping the same repetitions and a similar level of effort. Those are encouraging signs that your capacity for that exercise is improving.
Record enough detail to make that interpretation credible. “Ten repetitions” leaves out the resistance, technique and effort. More repetitions with a fixed load may also reflect muscular endurance. Neither observation measures muscle size directly.
You don’t have to push to failure to collect useful information. Use work already appropriate for you, and avoid turning every session into an examination. If technique is unfamiliar, a qualified coach can help you choose and repeat a suitable movement. These logging examples are practical applications, not a validated diagnostic test or a universal progression formula.
Look at endurance from more than one angle
Choose a familiar, manageable walk, wheel, cycle or other activity that suits you. You could compare how hard the same task feels, or how much you comfortably complete in a similar period at a similar effort. Use the same approach each time. A route completed faster only because you pushed much harder doesn’t establish the same improvement as a faster route at comparable effort.
A laboratory measure such as VO2max, the maximal rate at which your body can use oxygen during intense exercise, answers a more specific question. It doesn’t capture every contributor to endurance performance. In a small JSCR experiment, competitive cyclists added eight weeks of heavy strength training to their usual endurance training. Cycling economy improved: they used less oxygen for a given workload. This improvement exceeded the comparison group’s, without a statistically detected change in VO2max. Thirteen cyclists completed the study. This is a useful example of different outcomes moving differently, not a beginner’s training prescription. [3]
Include something that matters outside exercise
Think about a daily task you want to preserve or improve. Is getting up from your usual chair more manageable? Can you complete a familiar outing with fewer pauses? Note what you actually did, including support or assistance. “Needed a pause halfway” is more informative than a general feeling that you’re doing badly.
Formal functional tests have their place. A 1999 study of 76 community-dwelling adults over 60 found that a standardised 30-second chair-stand test was reasonably repeatable and related to leg-press strength. That supports its use for a particular question in that population. It doesn’t turn any chair, any technique or an online age chart into a diagnosis. [4]
If balance, pain or mobility limitations make a task uncertain, get help choosing an appropriate assessment. Keep any support you need. Removing it to improve a score changes the task and may make it unsafe.
When should you check again
Keep a brief log as you go, then choose a review point after several weeks of reasonably consistent practice. That is an organisational suggestion, not a biological deadline. The useful interval depends on what you’re measuring, your starting point and how variable the result is.
Look across comparable attempts rather than comparing your best day with your worst. Results vary because of the measurement method, the person and the circumstances. A small change can reflect improvement, ordinary variation, or both. No single percentage proves improvement in every test. [5]
Add a short note about how you’re feeling between sessions. Persistent fatigue, disrupted sleep or difficulty doing your usual work deserves attention even when one score improves. These observations help you judge whether the routine is manageable; they cannot diagnose “overtraining” or establish a separate recovery-fitness score.
If progress seems absent, first check whether the measurements were comparable and the intended training actually happened. Then consider whether the activity matches your goal and whether the current demands are manageable. More exercise is one possible adjustment, rather than an automatic answer.
Use devices and body measurements for the questions they answer
A watch can make recording activity convenient. Its estimated VO2max is still an estimate. In a 2026 validation study, Apple Watch Series 10 estimates differed substantially from laboratory measurements in some participants. The comparison included 35 mostly fit young adults. It assessed agreement at one point, so it did not establish how accurately the watch tracks an individual’s training-related change. Results from this sample and testing period shouldn’t be treated as a verdict on every device or update. [6]
Body-composition estimates deserve similar care. In a study of 39 people, drinking two litres of water before repeat measurement altered estimates from bioelectrical-impedance devices, including estimated fat mass. That was a fluid experiment, not sudden fat gain. Follow the method’s preparation instructions and avoid interpreting a single small fluctuation as new muscle or lost fat. [7]
You can keep these measures if they serve a useful goal. They do not need to decide whether carrying the shopping became easier.
Choose a useful next step
Write down one capability you care about, a repeatable way to observe it and the circumstances worth recording. Then give the routine enough consistency for a pattern to emerge. Progress might mean improving, or maintaining a valued ability during a demanding period of life.
These examples are for activities you can already perform safely. Stop if a task causes chest pain, faintness or unusual breathing difficulty. Seek urgent medical help for possible heart-attack symptoms, severe breathing difficulty or fainting. Get individual advice for recurring symptoms, an unexplained decline, or a condition affecting which activities are appropriate. [8, 9]
The question to return to is simple: “What can I do now, how does it compare with before, and how confident am I that I’m comparing the same thing?”
Dig Deeper
Three optional explanations for readers who want another way into the question.
An analogy for a useful set of measures
Imagine checking whether you’re becoming a more capable cook. You could time one recipe, notice whether you need less help, or see whether the result is more consistent. Each observation is useful for a different purpose. The weight of the ingredients would tell you something else.
Fitness tracking works similarly. Choose observations that represent the capability you care about. Getting better at one familiar recipe also doesn’t demonstrate that you can cook every dish. Practice and transferable skill are both valuable, but they need different evidence.
The analogy cannot tell us which exercise test is valid, how much change matters, or how long adaptation takes. Those questions require measurement research and the person’s context.
What a change in numbers actually means
Here is an invented example. You usually complete a familiar route in about 20 minutes at a manageable effort. One day takes 19 minutes 50 seconds. The stopwatch shows a difference, but wind, crossings, pacing and ordinary variation could help explain it. Repeatedly completing the route in around 19 minutes at comparable effort would be more persuasive. Neither example supplies a scientific cutoff.
Professionals distinguish relative reliability, often described with an intraclass correlation coefficient, from absolute error in the measurement’s own units. People can maintain similar rankings while their scores shift. A high coefficient alone doesn’t establish that your small change exceeds error. [5]
Published “minimal detectable change” thresholds apply to particular protocols and populations. Detectable also differs from important: a small detectable change might not matter in daily life, while a noisy test might miss something useful. Ask both questions when choosing a measure. [5]
Follow the reasoning before accepting the claim
1. Name the outcome. Did the study measure a lift, oxygen uptake, a timed task, muscle size or a health outcome?
2. Check the people and protocol. Would the result plausibly apply to this person, activity and equipment?
3. Look for learning and variation. Did participants practise? Were conditions consistent? Was absolute error reported?
4. Separate agreement from change. Agreement with a laboratory measurement on one occasion does not establish that a device tracks changes accurately. [6]
5. Test the conclusion. A nonsignificant change is not proof of no effect, particularly in a small study. Different outcomes may change differently, as the cycling experiment illustrates. [3]
For coaches, record what would make you revise your interpretation: repeated comparable results, a second relevant measure, or an assessment better suited to the goal. Keep the conclusion as specific as the evidence. “This task became easier under these conditions” can be a useful finding without becoming a claim about every aspect of health.
Linked sources and evidence notes
Checked 13 October 2026. This is a focused editorial research pass, not a systematic review. Access notes identify what was inspected. The everyday tracking examples and review interval are practical interpretations, not a tested program. No independent clinical review is claimed.
Definitions of physical activity exercise and fitness
Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Reports. 1985;100(2):126–131.
Conceptual definitions, not a training trial. Our four-question framing is an editorial aid. Access: abstract and original scanned page 129 inspected, including the health-related fitness table and glossary.
Previous lifting experience and test session gains
Ritti-Dias RM, Avelar A, Salvador EP, Cyrino ES. Influence of previous experience on resistance training on reliability of one-repetition maximum test. Journal of Strength and Conditioning Research. 2011;25(5):1418–1422. doi:10.1519/JSC.0b013e3181d67c4b.
Thirty men with no previous resistance training or more than two years of experience performed bench-press and squat tests on four days. Significant session gains occurred in the inexperienced group. Supports familiarisation, but does not isolate every cause of improvement or prescribe a universal number of practice sessions. Access: abstract only; full-text appraisal remains incomplete.
Cycling economy can change without a detected VO2max change
Sunde A, Støren Ø, Bjerkaas M, Larsen MH, Hoff J, Helgerud J. Maximal strength training improves cycling economy in competitive cyclists. Journal of Strength and Conditioning Research. 2010;24(8):2157–2165. doi:10.1519/JSC.0b013e3181aeb16a.
Read the source | Full journal article
Sixteen assigned; thirteen completed, eight intervention and five control. The intervention added four sets of four-repetition-maximum half-squats three times weekly. This small trained sample and laboratory outcome limit generalisation. Access: full nine-page journal PDF inspected, including Table 2.
Chair stand testing in community dwelling older adults
Jones CJ, Rikli RE, Beam WC. A 30-s chair-stand test as a measure of lower body strength in community-residing older adults. Research Quarterly for Exercise and Sport. 1999;70(2):113–119. doi:10.1080/02701367.1999.10608028.
Seventy-six volunteers, mean age 70.5, completed repeated chair-stand and maximal leg-press tests on separate days. Chair-stand scores showed reasonable repeatability and correlations with weight-adjusted leg-press performance. This supports a standardised functional assessment in the studied population, not self-diagnosis or a universal norm. Access: abstract checked; attempted full-text retrieval did not return the paper.
Interpreting reliability and individual measurement error
Weir JP. Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM. Journal of Strength and Conditioning Research. 2005;19(1):231–240. doi:10.1519/15184.1.
Methodological review of relative reliability and absolute error; no intervention sample or universal cutoff. Our route timings are invented illustrations. Access: author-uploaded full text inspected.
Apple Watch Series 10 agreement with laboratory testing
Lambe R, Schumann M, Donnelly L, et al. Accuracy of VO2 max Estimates From Apple Watch Series 10. Mayo Clinic Proceedings: Digital Health. 2026;4(2):100357. doi:10.1016/j.mcpdig.2026.100357.
Outdoor estimates compared with maximal treadmill testing. Forty completed laboratory testing; thirty-five were analysed, ages 20–36. Data: October 2024–January 2025. Cross-sectional design limits claims about training-related change. Access: PubMed and indexed full-text sections inspected; direct retrieval blocked. Exact watchOS version unverified.
Fluid intake can change body composition estimates
Jeong S, Bonner R, Firari A, Kurti S, Saunders MJ, Womack CJ. The effect of acute hydration on body composition assessed by multi-frequency and single-frequency bioelectrical impedance. Journal of Sports Medicine and Physical Fitness. 2023;63(10):1069–1074. doi:10.23736/S0022-4707.23.14913-9.
Thirty-nine participants, twenty male and nineteen female, were assessed with two bioimpedance methods and DXA before and after drinking two litres of water. Estimates shifted, with increases in estimated fat percentage for both bioimpedance methods. This acute protocol does not quantify everyday error for every home scale. Access: abstract only; publisher full text unavailable.
Safety symptoms during physical activity
US Department of Health and Human Services, Office on Women’s Health. Stay safe during physical activity. Updated February 3, 2025.
Official patient guidance supporting stopping for concerning symptoms and seeking appropriate care. This is safety guidance, not a study validating any tracking example. Relevant page text checked.
A performance score does not establish healthy blood pressure
US Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Keep Your Heart Healthy. Updated August 19, 2026.
Official patient guidance states that high blood pressure can lack symptoms and must be measured. Supports the narrow health distinction and urgent-care context. Relevant page text checked; no disease-risk estimate is inferred from exercise performance.
A question for you
What everyday activity would you like to find easier, and how could you make a fair comparison with where you are now? Please avoid posting private medical details.
Reader discussion
Comments are reviewed before appearing publicly. Please keep discussion relevant and respectful.