A question that surfaces in triathlon chats roughly once a week: “Are heart rate zones as a percentage of threshold the same for running and cycling?” The answers split into two camps — “of course they are, they're percentages” and “on the bike your heart rate is always 5–10 beats lower.” Both camps are right in their own way, and both draw the wrong practical conclusion from it.

Let's take it in order, because there are three different questions here that usually get glued into one.

How much HRmax actually differs

The largest sample on the subject is 5311 endurance athletes (4043 runners and 1268 cyclists), incremental testing with gas analysis, Journal of Clinical Medicine, 2023.

Measured maximum:

running — 184.6 ± 9.8 bpm;
cycling — 182.7 ± 10.3 bpm.

The difference is statistically significant (p = 0.001) and amounts to less than two beats. Not five, and not ten.

The same paper also checked 13 formulas for predicting HRmax. The root mean square error was 9.1–10.5 beats in both sports, with a mean absolute error of about 4.7%. In other words, any formula of the “220 minus age” variety misses your real number by roughly ten beats — and that is an entire zone.

The average difference is small — and transfer still doesn't work

This is where the main trap hides. In 2009 the Journal of Strength and Conditioning Research published threshold testing on triathletes, done separately on a treadmill and on a cycle ergometer. The mean difference in heart rate at the anaerobic threshold between the two modes turned out to be just 3.1 beats.

But the correlation between running and cycling threshold within the same person was r = 0.32, and the total transfer error was 12.1 beats.

Read it like this: the group shows almost no difference because in some people the cycling threshold sits ten beats lower, in others eight beats higher, and on average it all cancels out to zero. That average is of precisely no use to you. The authors draw the only conclusion available: you have to test separately in each discipline.

Where the difference comes from

The mechanism is clear, and it explains why the spread is so wide.

Less muscle mass is working. Running engages the trunk, the arms and the stabilizers; cycling mostly uses the legs, while the torso is supported by the saddle and the bars. Less active mass means less peripheral demand, and a lower heart rate at the same oxygen uptake.

Body position. Seated, venous return is better and stroke volume is higher — the heart needs fewer contractions for the same cardiac output.

No impact loading. In running, shock absorption is added on top of the metabolic cost, and it too costs heartbeats.

And most of all, it is individual. How big the difference is for you specifically depends on how many years you have been turning the pedals and how long you have been running. A former cyclist who came to running recently will show the opposite picture to a former runner.

What practitioners do — and what they do in vain

There is a separate paper showing that practitioners themselves set the zones differently for the two sports. In 2025, Scientific Reports surveyed 778 endurance coaches and athletes. Lower zone boundaries as a percentage of maximum:

Zone	Cyclists	Runners
1	47.3 ± 18.0%	55.4 ± 10.2%
2	65.7 ± 8.2%	69.8 ± 6.7%
3	76.4 ± 6.7%	79.6 ± 5.7%

Cyclists systematically place the bottom three zones lower — the authors link this to longer sessions and lower mechanical load. Note the standard deviations: ±18% in zone 1 among cyclists means there is no shared understanding of “easy work” even inside a single sport.

A separate story is when zones are calculated as percentages of threshold and the upper boundary of zone 4 or zone 5 ends up above your measured maximum. That is not a paradox and not a watch error. It is a signal that the model doesn't fit you: either the percentage coefficients come from someone else's population, or the threshold itself was determined incorrectly. Checking the threshold is simple — you should be able to hold that heart rate for about an hour.

How to apply this

Test separately in each sport. Twenty to thirty minutes at the flattest effort you can manage — on the run and on the bike — will give you two different numbers. That is normal, not an error.
Don't carry over a “minus 5–10 beats” correction. The group difference is about two beats, while the individual spread is twelve. An “on average” correction will miss by more than a separate test would.
Use age-based formulas only as a rough frame. An error of about ten beats is the width of a whole zone.
Work from threshold rather than from maximum if you train in two sports: the threshold shifts along with your form, the maximum barely moves.
Revisit your zones when the season and the volume change. If your share of cycling doubled over the winter, the old bike zones are already wrong.

Key points

Measured maximum heart rate on the bike is on average 1.9 beats lower than in running (184.6 versus 182.7 in a sample of 5311 athletes).
The mean difference in threshold heart rate between running and cycling is 3.1 beats, but the within-person correlation is only r = 0.32, and the total transfer error is 12.1 beats.
The group average is useless here: in different people the difference runs in opposite directions and cancels itself out.
HRmax prediction formulas are off by 9–10 beats — an entire zone.
Practitioners set cycling zones 1–3 lower than running zones by 3–8 percentage points, with an enormous spread inside the group.
A zone's upper boundary sitting above your maximum is a sign that the threshold was determined incorrectly or the coefficients belong to someone else.
There is only one working solution: separate testing for each sport.

Sources: “Validity of the Maximal Heart Rate Prediction Models among Runners and Cyclists”, Journal of Clinical Medicine, 2023. https://doi.org/10.3390/jcm12082884 · “Transferability of Running and Cycling Training Zones in Triathletes: Implications for Steady-State Exercise”, Journal of Strength and Conditioning Research, 2009. https://doi.org/10.1519/JSC.0b013e31818767e7 · “Contextualizing the Norwegian standardized intensity zone framework in an international sample of endurance practitioners”, Scientific Reports, 2025. https://doi.org/10.1038/s41598-025-17023-z