METRIC DEFINITIONS
Heart-rate drift versus aerobic decoupling
Heart rate can rise while speed stays steady, or speed can fall while heart rate stays steady. Drift and decoupling describe different parts of that relationship.
Name the calculation before interpreting it
For this article, simple heart-rate drift means the percentage change between the average heart rates of two chosen segments: (late HR − early HR) ÷ early HR × 100. This is an explicit two-segment teaching calculation, not a claim that every app's field named “HR Drift” uses it.
For decoupling, compare a speed-to-heart-rate ratio across the same segments. We use (early ratio − late ratio) ÷ early ratio × 100, so a positive value indicates a falling ratio. RUNALYZE also describes decoupling as a comparison of early and late efficiency. Always check units, sign and segment definitions in the system you use.
The percentages are not interchangeable
Imagine equal, steady-speed segments at 3 m/s. Average heart rate is 150 bpm early and 165 bpm late. The values are illustrative.
| Calculation | Result |
|---|---|
| Heart-rate increase: (165 − 150) ÷ 150 × 100 | 10.00% |
| Early meters per beat: 60 × 3 ÷ 150 | 1.200 |
| Late meters per beat: 60 × 3 ÷ 165 | 1.091 |
| Decoupling: (1.200 − 1.091…) ÷ 1.200 × 100 | 9.09% |
The reciprocal relationship makes the percentages different. The example does not justify rounding both to the same metric or applying one threshold to the other.
Now change speed instead of heart rate
If heart rate stays at 150 bpm while speed falls from 3.0 to 2.7 m/s, the simple HR change is zero. Meters per beat falls from 1.20 to 1.08, making the defined decoupling 10%. A heart-rate-only comparison would miss the speed change.
Terrain, workout structure and recording gaps still matter. A planned slower section is not evidence of unexplained fading. A positive percentage alone cannot diagnose its cause or certify a runner's endurance.
Use the definition shown with your result
In Runome, use the metric explanation and available streams to understand the displayed field. Keep the app's result separate from a hand calculation if its window or method differs. Report the chosen segments alongside any value you calculate yourself.
Start with segment selection and then try the warm-up comparison. These checks are more useful than treating one percentage as a universal pass-or-fail test.