Cardiac Drift Calculator

What this page does

A cardiac drift calculator reads the two numbers your watch already keeps for each half of an effort: the pace you held and the average heart rate you ran at. From those it works out how fast your heart rate climbed while the pace stayed put, and how far your output and your heart rate came apart over the effort. You type the two halves; the page does the arithmetic and prints every step it used.

Cardiac drift is the climb in heart rate that happens while the pace does not move. The published definition is a gradual increase in heart rate despite a constant external workload (Barsumyan A., Soost C., Graw J. A., Burchard R. (2026), The relationship between cadence decline, cardiovascular drift and aerobic decoupling as a marker of fatigue in well trained cyclists, BMC Sports Science, Medicine & Rehabilitation 18(1):179, DOI 10.1186/s13102-026-01678-w). What sits behind it is not settled: one review describes it as a progressive fall in stroke volume that the heart rate partly compensates for (Coyle E. F., González-Alonso J. (2001), Cardiovascular drift during prolonged exercise: new perspectives, Exercise and Sport Sciences Reviews 29(2):88–92, DOI 10.1097/00003677-200104000-00009), and another calls it a live argument between skin blood flow displacement and sympathetic activation, with the mechanism still unknown (Souissi A., Haddad M., Dergaa I., Ben Saad H., Chamari K. (2021), A new perspective on cardiovascular drift during prolonged exercise, Life Sciences 287:120109, DOI 10.1016/j.lfs.2021.120109). This page therefore reports what the numbers are and does not claim what caused them.

Aerobic decoupling is the other half of the same picture: how much the gap between your heart rate and your output opened up between the two halves. The published protocol puts one 60-minute effort into two equal halves and compares them, and this page follows the same two-epoch structure, so your two halves are the two epochs.

No other page on this site reads a climb. The heart rate recovery calculator reads what happens after you stop, which is the opposite end of the same curve. The VO2 max calculator works from a test to a number; the lactate testing page works from a threshold to a pace. This page starts from two halves of an effort you have already run.

Everything runs in your browser. Nothing you type is sent to a server, there is no account and no upload, and the numbers never leave the page.

How the numbers are calculated

You supply six numbers: the pace and the average heart rate of each half, and how long each half took. Three different quantities come out of them, and they answer three different questions.

The drift rate, in bpm per hour. The climb in heart rate, divided by the time over which the climb happened:

rate = (HR₂ − HR₁) ÷ elapsed hours

The elapsed time is the gap between the middle of the first half and the middle of the second half, because those are the moments the two averages you typed describe. Two 30-minute halves put their midpoints 30 minutes apart. If your halves are 25 and 35 minutes, the gap is 12.5 plus 17.5, which is 30 minutes again, and the page shows you the number it used.

The drift, as a percentage. This is the published equation, unchanged (Barsumyan et al. 2026):

Cardiac Drift (%) = (HR(Second Half) − HR(First Half)) ÷ HR(First Half) × 100

The same two heart rates, stated as a share of where you started rather than as a speed. The published protocol reports this figure, and so does this page.

The aerobic decoupling. The published equation for it is written in terms of power per heartbeat (Barsumyan et al. 2026):

Pw:HR Decoupling (%) = (Pw:HR(First Half) − Pw:HR(Second Half)) ÷ Pw:HR(First Half) × 100

You do not usually know your power, so the page substitutes the running form of the same ratio: how much distance you covered for each heartbeat. Pace is the inverse of speed, and a ratio built on pace would flip the sign of the answer, so the page converts your pace to speed first and reports the ratio as metres covered per beat:

Pa:HR = (60,000 ÷ pace in s/km) ÷ HR     Decoupling (%) = (Pa:HR₁ − Pa:HR₂) ÷ Pa:HR₁ × 100

A positive decoupling number means the second half produced less for each heartbeat than the first: the gap between your effort and your heart rate opened up. A negative number means the opposite, that you got more out of each beat. Neither number is a verdict, and the page does not print one.

• HR₁, HR₂ — the average heart rate of the first half and of the second half, in beats per minute.
• pace — the average pace of that half, in minutes and seconds per kilometre. Multiply by 1.609344 for seconds per mile, then divide by 1609.344 to come back.
• elapsed — the time between the midpoints of the two halves, in minutes.
• Pa:HR — the pace-to-heart-rate ratio, stated as metres covered per heartbeat.
• 1 km = 1000 m and 1 mile = 1609.344 m exactly — definitions, not measurements.

A worked example: 4:30 per km at 155, then 4:35 per km at 160

First half: 270 seconds per kilometre is 222.22 metres per minute, so the heart rate of 155 gives a Pa:HR ratio of 222.22 ÷ 155 = 1.43 m per beat. Second half: 275 seconds per kilometre is 218.18 metres per minute, and 218.18 ÷ 160 = 1.36 m per beat. The decoupling is (1.4337 − 1.3636) ÷ 1.4337 = 4.89%.

The heart rate rose 5 beats, from 155 to 160. The two midpoints are 30 minutes apart, which is half an hour, so the drift rate is 5 ÷ 0.5 = 10 bpm per hour. As a percentage of the first-half average it is 5 ÷ 155 = 3.23%. The two halves covered 6.67 km and 6.55 km, so 13.21 km in an hour.

Read those three numbers together rather than apart. A climb of 5 beats over half an hour is small in itself; what makes it interesting is that the pace went the other way, so the climb happened while the work got harder rather than easier. If the second half had been 4:25 at 160, the Pa:HR ratio would have been 1.41 m per beat and the decoupling would have been negative.

What makes these numbers unreliable

Everything below changes the two numbers you type, or changes what they mean. The page lists them and does not correct for any of them: correcting needs a model of heat production, of the wind or of the gradient, and no source this page can stand behind supplies one. An unsourced correction would make the number look more precise than it is.

• Heat. The best documented distortion. In a study of women working at the same percentage of their maximum oxygen uptake, the group that produced more metabolic heat drifted more — 15 beats and 10% against 10 beats and 6% — while at a fixed rate of heat production the two groups did not differ at all (Stone T., Burnash S. G., Earley R. L., Mulholland A. M., Yoder H. A., Macdonald H. V., Richardson M. T., Wingo J. E. (2025), Metabolic Heat Production Modulates the Cardiovascular Drift-VO₂max Relationship Independent of Aerobic Fitness in Women, Medicine & Science in Sports & Exercise 57(1):181–191, DOI 10.1249/mss.0000000000003543). The same group reported larger drifts in the heat, where heart rate rose 19% while running from 15 to 45 minutes and stroke volume fell 20% (Wingo J. E., Stone T., Ng J. (2020), Cardiovascular Drift and Maximal Oxygen Uptake during Running and Cycling in the Heat, Medicine & Science in Sports & Exercise 52(9):1924–1932, DOI 10.1249/mss.0000000000002324).
• Wind. A headwind makes the same pace cost more, so the drift you measure is a drift against a harder task. The page cannot tell that apart from a drift against the same task.
• Gradient. A hill in the second half raises the work done at the same pace, which is the same problem as the wind: the halves are no longer comparable.
• Dehydration and fluid loss. The page cannot separate a climb caused by losing fluid from one caused by heat, and it does not adjust for either.
• An effort too short. Drift needs time to appear: the stroke-volume decline it is built on is described as starting after 10 to 20 minutes of exercise (Coyle & González-Alonso 2001). A ten-minute effort split in two cannot carry a drift.
• Uneven halves. Two halves of very different length are not two epochs. The page reads the gap between the midpoints so that the number stays honest, but it cannot make two unequal halves comparable.

The two lookup tables, and how to read them

The first table gives the heart-rate climb in bpm per hour. Each row is one elapsed time and each column is one climb, and every cell is the same single division: the climb divided by the row's minutes. The row nearest the elapsed time on your own numbers is highlighted.

The second table gives the same climbs as a percentage of a starting heart rate. Each row names a first-half average and each column is one climb, and every cell is the climb divided by the row's heart rate. The row nearest your own first-half average is highlighted.

Read the two tables together and the difference between them is the whole point of this page. The first table knows how long your effort was and says nothing about your fitness; the second table knows how fast your heart rate was beating and says nothing about how long you were out there. A climb of 10 beats is 5 bpm per hour over 30 minutes and 20 bpm per hour over 15 minutes, and the rate is the one that describes the climb, while the percentage is the one that describes your heart rate.

A rate and a percentage: two ways to state the same climb

Both numbers come out of the same two heart rates, and neither is the true one. They answer different questions, and mixing them up is how a small climb gets described as a large one.

The rate in bpm per hour is about time. It asks how fast your heart rate was climbing while you were out there. It gets larger as the effort gets shorter, because the same climb happens in less time. It is the number to quote when you want to say how the effort went.

The percentage is about heart rate. It asks how much of your own heart rate the climb was. A 5-beat climb is 3.2% at 155 beats per minute and 4.5% at 110. It gets smaller as your heart rate gets higher, because the same climb is a smaller share of a faster beat. It is the number to quote when you want to compare two efforts with different heart rates.

Neither is a grade. The percentage has no threshold behind it on this page, because a threshold needs an outcome and this page publishes no outcomes. What the percentage is good for is comparing two of your own efforts, run on the same protocol, where the protocol is the same in both cases.

Supported range and limits

• Pace: 1:00 to 30:00 per kilometre, written mm:ss. Anything outside that is refused rather than divided.
• Heart rate: 60 to 220 beats per minute, in either half.
• Half length: 1 to 360 minutes in each half.
• Elapsed: the gap between the two midpoints, which is half of the first half plus half of the second. Two 30-minute halves give 30 minutes.
• Rounding: rates and percentages to two decimals, distances to two decimals in kilometres, ratios to two decimals in metres per beat. One rounding, at the end.
• Cost: none. No account, no email, no paywall, no daily limit, no request to a server.
• Privacy: every number is computed in your browser. Nothing you type is sent anywhere or stored — reload the page and it is gone.
• Works offline once the page has loaded, and on a phone.

What this page deliberately does not do

• No verdict on the number. There is no band, no grade, no good and bad, and no threshold to compare against. A verdict needs a threshold with an outcome behind it, and this page publishes no outcomes. The published cohort figures are described where they appear as a description of that study's riders on that protocol, and not as a line you are measured against.
• No correction for heat, wind or gradient. These are listed as distortion sources, above, and nothing more. An unsourced adjustment would make the number look more precise than it is.
• No heart rate reserve conversion. This page does not convert between a percentage of your maximum heart rate and a percentage of your heart rate reserve; the fat burning heart rate page is the one that does that.
• No zone boundaries. The page does not print training zones and does not grade your heart rate against them; the heart rate zones page is the one that carries a named zone model with its source.
• No training plan and no fatigue model. No weekly volume, no session design, no progression, and no model of the heart rate slowing you down in the last kilometre. The page reads the halves you ran.
• No VO2 max estimate. A drift number is not converted into a fitness number; the VO2 max calculator works from a test instead.
• No medical reading. Nothing here is advice about your heart, and nothing here is a diagnosis. If a number from this page worries you, that is a conversation for a doctor, not for a calculator.
• No age or sex grading. The number is not adjusted for either, and no table on this page is split by either.
• No prediction. The page does not tell you what your next effort will show, or what a training block will produce.
• No accounts, no history, no watch sync. The page reads the six numbers you type. It does not read your watch, and it does not keep what you entered.
• No heart rate recovery. How far your heart rate falls after you stop is a different measurement, and the heart rate recovery calculator is the page for it.

Questions