Heart rate climbing while power stays flat on a long ride: cardiovascular drift, decoupling, and what to do

What cardiovascular drift actually is

Cardiovascular drift is the steady upward creep of heart rate during prolonged steady work, even when power output never changes. The mechanism is mechanical, not a loss of fitness: stroke volume — the blood pumped per beat — declines after the first 10-20 minutes, and heart rate rises to hold cardiac output constant [Coyle & González-Alonso 2001].

The classic picture blamed rising skin blood flow as the body shed heat, but Coyle and González-Alonso reframed it: the progressive decline in stroke volume after 10-20 minutes of exercise is driven primarily by the rising heart rate itself, not by blood pooling in the skin as core temperature climbs [Coyle & González-Alonso 2001]. The practical upshot is the same either way. To keep delivering the same oxygen at the same 200 watts, a heart pumping less per beat has to beat more often. Over a two-to-three-hour ride that compensation routinely shows up as a 5-10% rise in heart rate at unchanged power.

Heat and dehydration amplify it sharply. Montain and Coyle had trained cyclists ride two hours at a fixed sub-threshold intensity in a 33C room while replacing different fractions of their sweat losses; the heart-rate rise and the stroke-volume decline both scaled almost perfectly with how dehydrated each rider got, with correlations of r=0.99 [Montain & Coyle 1992]. A rider who finishes a long summer ride 3% down on body weight will see far more drift than the same rider on a cool day with a full bottle every 45 minutes. The drift is real physiology responding to a real thermal and fluid load — it is not your power meter or your fitness lying to you.

How to see drift in a Strava ride (heart rate up, power flat)

On a Strava activity, open the analysis view and overlay the heart-rate and power streams. Drift looks like two lines diverging: power holds a flat band while heart rate ramps slowly upward across the back half. The cleaner the power line, the more obvious the divergence — and the more confidently you can call it drift rather than effort.

Strava records both streams honestly and plots them on the same time axis, but it does not interpret the gap between them — which is the recurring theme of the broader case for using Strava as a training tool rather than a journal: it syncs your data faithfully but does not read it. To see drift, scrub to a stretch where the terrain and your effort were steady, then compare the first hour to the last. If power is sitting at a flat 190 watts in both windows while average heart rate has climbed from 138 to 150, that 12-beat gap at identical power is drift, not a harder ride.

The trap is heart-rate-only riders reading the climbing line as effort. Strava's Relative Effort is built from weighted time-in-heart-rate-zone [Meyer 2018], so drift mechanically pushes late-ride minutes into higher zones and inflates the score on a ride that felt easy throughout — the exact pattern our companion piece on why Relative Effort runs high on easy rides walks through. If you only have heart rate, lean on perceived exertion as the tiebreaker; Foster's session-RPE is the cheapest valid internal-load check in cycling [Haddad et al. 2017], and a ride that drifted but felt like a 4 out of 10 was a 4.

If you have a power meter, the read is unambiguous. Normalized power flat across the ride plus heart rate trending up is the textbook signature [Allen et al. 2019]. The presence of power data is what lets you separate drift from a genuine intensity increase, because power answers the question heart rate cannot: did the actual work rate change, or only the cost of producing it?

Aerobic decoupling as a durability test, not a warning

Quantify the gap and it becomes a fitness signal. Aerobic decoupling — Pw:HR — compares the power-to-heart-rate ratio in the first half of a steady ride to the second half. A long ride that stays coupled (typically under about 5% drift) indicates good aerobic durability at that duration; a ride that decouples badly flags a ceiling you have not yet trained past.

The logic runs straight out of the drift mechanism. A well-trained aerobic engine maintains stroke volume longer, so heart rate rises less for the same watts and the power-to-heart-rate ratio stays stable across the ride. A 2025 analysis using machine learning to track training response in cyclists found exactly this — that attenuated cardiovascular drift, a smaller heart-rate rise for a given power output, is interpreted as a marker of robust aerobic endurance, and that drift shrinking over a block corresponds to genuine cardiovascular adaptation [Barsumyan et al. 2025]. Decoupling is drift, measured and turned into a number you can track.

Used as a test, it answers a question FTP cannot: not how much power you can hold for an hour, but how long you can hold a given aerobic power before the wheels start coming off. A rider whose two-hour endurance rides decouple at 3% has durability a rider who decouples at 12% at the same power does not, even if their FTPs match. For anyone training toward a long event — a gran fondo, a four-hour road race, a gravel day — that durability is the thing the event actually tests. TrainingPeaks surfaces Pw:HR and an Efficiency Factor for this directly; Intervals.icu computes decoupling on rides for free.

The honest caveat: a single decoupled ride is weak evidence on its own, because heat, a missed bottle, a poor night's sleep, or simply going out too hard all inflate it. Read decoupling the way you read any drift number — as a trend across several comparable rides in similar conditions, not a one-ride verdict. Drift trending down across a base block is the signal worth chasing; one ugly number on a hot, under-fueled afternoon is mostly noise.

What to do about it (usually nothing)

On a normal long ride, the correct response to drift is nothing. Do not log the ride as harder than it was, do not skip tomorrow's quality session, and do not add recovery you do not need. The two things drift should change: how you fuel and hydrate long rides, and whether you start treating decoupling as a durability metric to track.

The training decision is the whole point. A drifted heart rate is not evidence of fatigue or lost fitness, so it should not reclassify a zone 2 endurance ride as a hard day or talk you out of a planned interval session. Trust the inputs that do not depend on heart rate — power-based load if you have it, perceived exertion if you do not [Allen et al. 2019, Haddad et al. 2017]. A ride long enough to drift is doing the aerobic work a base block is built around, and the only mistake is reading the climbing heart-rate line as a problem to solve.

Where drift should change behavior is fuelling and heat strategy. Because dehydration scales the drift almost linearly [Montain & Coyle 1992], a rider who consistently sees large drift on long rides has a concrete, actionable fix: drink more and earlier, manage heat, and start long rides topped up. That is the rare case where a high drift number points at something you can act on — not your training plan, but your bottles.

Quick answers

Sources cited in this guide

1. 01
   Coyle & González-Alonso 2001. Cardiovascular drift during prolonged exercise: new perspectives. Exercise and Sport Sciences Reviews.
2. 02
   Montain & Coyle 1992. Influence of graded dehydration on hyperthermia and cardiovascular drift during exercise. Journal of Applied Physiology.
3. 03
   Barsumyan et al. 2025. Quantifying training response in cycling based on cardiovascular drift using machine learning. Frontiers in Artificial Intelligence.
4. 04
   Meyer 2018. Quantifying Effort through Heart Rate Data. Strava Engineering.
5. 05
   Allen et al. 2019. Training and Racing with a Power Meter (3rd ed.). VeloPress.
6. 06
   Haddad et al. 2017. Session-RPE Method for Training Load Monitoring: Validity, Ecological Usefulness, and Influencing Factors. Frontiers in Neuroscience.

More inside Training with Strava

Other articles in this series

1. 01

   What your Strava Fitness number means (and if yours is good)

   Strava Fitness is CTL — a 42-day weighted load average. What the number means, why it is personal, and the decisions to make from it.

   →
2. 02

   Apps that connect to Strava: read vs display

   How to tell which training apps actually read your Strava data and adapt versus the ones that only display your rides.

   →
3. 03

   Is Strava Premium worth it for a self-coached cyclist?

   What Strava Premium gives a training-focused rider, what it doesn't (no coaching), and when the free Intervals.icu chart beats paying.

   →
4. 04

   Relative Effort vs TSS: which to trust by workout

   A per-workout-type rule for when to trust Strava Relative Effort vs power-based TSS, and why they are not the same units.

   →
5. 05

   How to set up Strava for training: one-time configuration

   Configure Strava once for training: real FTP and max HR, honest zones, one sensor stack, and feed privacy that protects your plan.

   →
6. 06

   Strava segments as fitness tests: map efforts to tests

   Use Strava segments as scheduled benchmark tests: which profiles map to which test, how to schedule every 4-6 weeks, and controlling variables.

   →
7. 07

   What to look at on Strava after a ride: 4 metrics

   A 30-second post-ride routine: the four Strava metrics that matter after every ride, the ones to ignore, and why.

   →
8. 08

   Strava Fitness going down while training hard: the decay math

   Why your Strava Fitness (CTL) drops even when you train hard: the 42-day EWMA decay math, the real causes, and when a falling line is correct.

   →
9. 09

   Strava indoor power vs outdoor HR: Fitness chart jumps

   Mixing power-based TSS and HR-based Relative Effort splices incompatible units into your Strava Fitness chart. Why it jumps and how to fix it.

   →
10. 10

    Strava heart rate zones wrong: the whole-dashboard cascade

    How a wrong max HR in Strava cascades into bad zones, Relative Effort, and Fitness — and how to set zones from real data.

    →
11. 11

    Why your Strava Relative Effort is high on easy rides

    Relative Effort can spike on a genuinely easy ride — usually a mis-set max HR, not lost fitness. What inflates it on Strava and what to do.

    →