DRIFT WORKFLOW
Choosing comparable halves for a decoupling calculation
The calculation is short. Choosing the two intervals deserves more attention.

Choose the question before the midpoint
If you want to compare a steady effort, splitting an entire activity exactly in half may include warm-up in the first side and a hill or cooldown in the other. The result then describes those mixed conditions.
TrainingPeaks discusses decoupling as a comparison between output and heart rate over an aerobic effort. The public Runome calculator makes a simpler, explicit comparison of the two average-pace and average-heart-rate pairs you enter. It does not locate appropriate intervals automatically.
Mark an equal-duration comparison
The numbers in this example are illustrative, not a customer training record.
| Time interval | Purpose | Use in calculation |
|---|---|---|
| 00:00–15:00 | Gradual warm-up | Exclude |
| 15:00–45:00 | Steady running | First half |
| 45:00–75:00 | Steady running | Second half |
The comparison contains two 30-minute intervals. Use each interval’s own average pace and heart rate. Equal duration does not mean equal distance: if pace changes, the distances can differ. Do not force the distance totals to match by changing only one boundary.
Check interruptions and coverage
If there is a long stop at minute 52, the second interval is no longer directly comparable with the uninterrupted first one. A route reversal into a strong headwind or a sustained climb can also change the demand. Document the difference or choose a more comparable part of the record.
Look for missing heart-rate samples and abrupt plateaus. A perfectly steady average can conceal a sensor problem. Use the same treatment of pauses and unavailable data in both halves. If the comparison cannot be made fairly, leave it out of a trend rather than treating every calculated percentage as equally informative.
Calculate both percentages
Enter the two pairs in the decoupling calculator. At the same 5:00 min/km pace, a change from 140 to 147 bpm gives 5% heart-rate drift and about 4.76% decoupling. The latter compares efficiency ratios, so its denominator differs.
Keep the selected timestamps with the result. That small record lets you repeat the comparison on another outing and understand why an alternative selection produced a different number.