Eight centimetres of vertical oscillation can become eight per cent, but it can also become less. The difference lies in step length and the measurement definition. Two similarly named running-watch metrics can therefore give different views of the same movement. Original numerical examples and research explore what to read together before judging technique from a percentage.
Centimetres and percentages answer different questions
Vertical oscillation describes the up-and-down displacement determined by a measurement procedure during running. Vertical ratio adds a reference length, comparing displacement with step length. One is an absolute length, the other a ratio. Both relating to the same run does not require them to change in the same direction.
The official description of Garmin Forerunner 965 data fields presents vertical oscillation as torso movement in centimetres; vertical ratio is oscillation divided by stride length. In the manufacturer’s terminology here, stride length runs from one footfall to the next. This is a specific function description, rather than a general definition for all research or other manufacturers’ measurements.
The English word stride makes checking the definition particularly important. A biomechanics paper may use it for the length of a complete same-foot cycle, while elsewhere it means a single step between opposite feet. Doubling the denominator in the formula would halve the ratio. Before comparing percentages from two devices or studies, check which length actually appears in the denominator.
The same eight centimetres produce two ratios
The following original examples use hypothetical numbers, rather than ideal movement values. In situation A, oscillation is eight centimetres and step length one metre. First make units consistent: one metre is one hundred centimetres. Eight divided by one hundred is 0.08, or eight per cent. Dividing centimetres directly by the number one representing one metre would incorrectly mix units.
In situation B, oscillation remains eight centimetres, but step length is 1.2 metres, or one hundred and twenty centimetres. The ratio is approximately 6.7 per cent. The percentage decreased while the centimetre value of vertical movement did not change at all. A larger denominator created the difference. Summarising this as the runner moving less up and down would be inaccurate.
In situation C, nine centimetres of oscillation accompanies a step length of 1.4 metres. The ratio is approximately 6.4 per cent. In this arithmetic example, oscillation is greater but the ratio is lower than in both A and B. Ranking therefore differs depending on the column you examine. Neither ranking is a complete assessment of running: each answers a different descriptive question.
| Situation | Oscillation | Step length | Ratio |
|---|---|---|---|
| A | 8 cm | 1.0 m | 8% |
| B | 8 cm | 1.2 m | 6.7% |
| C | 9 cm | 1.4 m | 6.4% |
The difference between A and B is approximately 1.3 percentage points. Relative to the initial eight per cent, this is a decrease of about 16.7 per cent, while oscillation remains eight centimetres. Percentage points and relative percentages are different descriptions; neither means an equivalent energy or time gain.
Enlarge imageFaster running can change the denominator
Speed relates to step frequency multiplied by step length defined consistently with it. Longer steps at the same frequency also increase travel speed. Imagining the earlier A and B examples at the same rhythm therefore produces tasks at different paces. A change in vertical ratio then compares different conditions rather than improving one movement at unchanged speed.
Do not automatically make a faster section’s lower percentage the target for an easy run. Faster movement may involve another step length and another overall movement pattern. To understand changes in your technique during similar tasks, first align speed and route characteristics. An identically named metric does not make different runs directly comparable.
Keep oscillation, step length and ratio alongside each other in your diary, together with their corresponding speed. Retaining only the last column loses one of the most important parts of understanding the change. Reading a ratio requires knowing both numerator and denominator. A smaller percentage then becomes explainable information rather than independent praise or criticism.
Torso movement is not automatically the whole body’s centre of mass
Research and manufacturer descriptions may link vertical displacement to different body points. A signal measured at the torso or pelvis differs conceptually from the estimated path of the whole body’s centre of mass. Sensor location, attachment and calculation method form part of measurement too. Saying the watch displayed this many centimetres does not describe the whole method.
The technique research involving 97 endurance runners used three-dimensional whole-body motion capture and respiratory-gas measurement. It analysed several movement characteristics, including variables related to vertical displacement of the pelvis. Its specific normalisation is not the same definition as a sports watch’s vertical ratio calculated against step length.
Do not combine the two data sources as though two screens displayed the same metric. Research may help explain relationships between movement and economy, while manufacturer documentation explains your own display. Both are useful when names and methods remain clear. Scientific background does not automatically validate every centimetre reported by every wearable.
A poorer result from exaggerated bouncing does not establish zero oscillation
The original experiment involving nine female distance runners compared several deliberately modified running conditions. Running with exaggerated vertical oscillation produced higher oxygen demand than habitual running. This supplies data about a specific exaggerated change, rather than showing that zero is everyone’s ideal vertical displacement.
The logical distinction matters. An unfavourable effect from excessively increasing a characteristic does not imply that reducing it without limit is favourable. The two directions are separate questions, and other parts of movement may change meanwhile. A runner does more than try to guide a vertical point along a straight line. Ground interaction and sustaining the whole movement are connected tasks.
The model analysing the mechanical power requirements of running connects several factors, from vertical movement through speed to air resistance. Minimising one indicator therefore does not necessarily optimise the whole task. Here, the model supports understanding of relationships rather than individual technique instructions. Pushing a watch column down is not itself an improvement in the energy demand measured in research.
A whole-session average can hide individual sections
Warm-up, faster sections, turns and cool-down may mix in one average. Comparing whole-run vertical ratios across two days can reveal differences simply because the sessions had different compositions. One contained more comfortable running, the other more fast sections with longer steps. Changing session structure differs from a change in one constant technique characteristic.
The method matters when averaging ratios too. An average of ratios calculated by section is not necessarily the same as the ratio of separately averaged oscillation and step length. Differences may be especially relevant when durations and weighting differ too. You need not recreate every app’s internal calculations, but do not automatically assume a quick division reproduces exactly the same summary.
The practical approach is to find comparable steady sections and name the view. Are you examining a whole-session average, lap average or a shorter section? Retain the original file and measurement notes. Later interpretation then has a chance to distinguish an actual pattern from differences produced by summary methods or varied training structure.
Enlarge imageConsistent measurement comes before judging technique
Record changing a watch, sensor or attachment location as a separate event. Do not immediately call a difference between two data series progress or deterioration. First clarify whether you are seeing the same quantity with the same definition and a similar method. This is simple measurement discipline and can prevent unnecessary self-criticism or excessive enthusiasm.
Using a securely attached device and following your own model’s instructions matter because the sensor does not measure independently of the body. Do not replace manufacturer instructions with online advice for another device. If something looks unusual, begin by describing the circumstances. A number appearing does not automatically supply a reliable explanation.
A movement change prompted by symptoms is a separate individual question. Vertical ratio is not a diagnosis, and a lower percentage does not prove recovery. If working with a professional, discuss the specific task, history and measurement conditions together. A general running-watch metric alone does not prescribe how each body part should move.
Four columns create a clearer picture
Include speed, oscillation in centimetres, consistently defined step length and the calculated ratio on your overview sheet. Add brief notes about surface, shoes and device. If the percentage changed, check which underlying value moved. This produces a specific question rather than making one display colour determine the assessment of your entire run.
Record the comparison’s purpose too. Do you want to see repetition of the same easy task, or understand movement at different speeds? These questions need different data series. Every run need not become a technique test. A metric becomes useful when connected with a clearly stated question and interpreted with the actual conditions retained.
Vertical oscillation and vertical ratio can therefore give different views of the same movement. Centimetres describe up-and-down displacement; a percentage describes its ratio to a chosen length. Keeping that distinction lets you look behind a smaller number: did oscillation really change, did step length differ or did comparison conditions change? This creates more accurate understanding than a rule that lower is always better.
Sources
- Folland and colleagues: running technique, economy and performance.
- Tseh and colleagues: deliberately modified movement and oxygen demand in female runners.
- Jenny and Jenny: a model of the mechanical power requirements of running.
- Garmin Forerunner 965: vertical oscillation, stride length and ratio data fields.
How Futószív articles are prepared: our editorial guidelines.


