At the top of a climb, one breath in ten goes to the muscles that breathe. Not to your legs.
Not a metaphor. A measurement. At maximal effort, about ten percent of all the oxygen your body takes in is burned by the diaphragm, the muscles between the ribs, the neck and the belly wall. In one study, women averaged closer to fourteen percent and men around nine at maximal ventilation. The cost rises steeply near the top, exactly where the race is decided.
Most riders hear that and think: fine, that's the price of going hard.
It isn't. That share is not fixed. And the reason nobody told you is that the test everybody uses can't see it.
At maximal effort, the measured average was about one breath in ten.
At maximal ventilation, one study reported about 14% in women and 9% in men.
The wrong test
A VO2max test measures the most oxygen your whole body can use in one minute of all-out effort. It's a ceiling. It's your income.
It tells you nothing about where that oxygen went. How much reached your legs, and how much was spent on the muscles that move your ribs. That second number is your rent. And you never see the rent by looking at the income.
Here's how blind the standard test can be. In 2012 a group in Zurich pulled together forty-six studies of breathing-muscle training. Across those studies, constant-load tests improved by about sixteen percent and intermittent incremental tests by about eighteen percent. Conventional incremental tests did not detect the same improvement. The size of the benefit also depended on fitness, with less-fit participants improving more.
Same training question. A different verdict, depending on which test was used.
The ramp test measures income. The constant-pace test measures what you can do with what's left after rent. The conventional ramp remains one of the standard ways this sport tests capacity.
How big the rent is, and how much it varies
Even at the same ventilation, the cost of breathing is not identical between healthy people. Sex, airway size, expiratory flow limitation and breathing mechanics can all change the work required to move air.
That variation is the opportunity. Nobody is going to give you a new ceiling. But the rent can move.
I know because I measured mine.
Same watts, three months apart
At the same power, 360 watts, three months apart, with nothing in my training changed:
Breathing rate went from 53 breaths a minute to 33.
Air moved went from 143 liters a minute to 117.
Heart rate went from 164 to 154.
Nose only, both times.
Same watts. Almost a fifth less air. Ten beats less.
Power held constant
360 W
BREATHS / MIN
53
AIR / MIN
143L
HEART RATE
164BPM
More air moved. More breaths taken. More heartbeats spent.
That's 26 liters less air moved every minute at the same power. The ceiling didn't move. The price of reaching it did.
And here's the part that matters for the rest of this series. I didn't train my breathing. No device, no box breathing, no protocol. I changed what my trunk allows, and the breath got slower and deeper on its own.
Why the share is not fixed
The physiology papers treat each person's breathing mechanics as fixed background. My work treats them as the variable. That's the whole difference.
A body that has lived for years under pressure carries muscles that never fully switch off. Jaw, neck, shoulders, the belly wall, the muscles between the ribs, the diaphragm itself. I wrote about this a few issues ago: the brain stops noticing a muscle that is always on. The signal is still there. You stopped receiving it.
A muscle that stays partly contracted all day is a motor idling at high revs. It burns fuel and makes carbon dioxide while you sit in the car. So your body has to breathe for work that nobody can see being done. That's rent you're paying before the ride starts.
Then the same tension decides how you're allowed to breathe. A held rib cage can't widen low and to the sides. A braced belly doesn't let the diaphragm drop. Depth is not available. So the only breath left is faster and higher, and sooner or later through the mouth.
That pattern can raise the price by three routes described in respiratory physiology. They were not measured separately in my two tests.
The first is waste. Your airways hold the same volume whether the breath is big or small, so a small fast breath spends a larger share of itself where no exchange happens. The tense chest breather has to move more air to get the same oxygen in.
The second is work. Air pulled in by the neck against a stiff rib cage costs more than air drawn by a diaphragm into a soft belly. The cost of breathing rises with breathing rate.
The third is blood. A diaphragm that is already working before the effort begins tires earlier, and a tired diaphragm sends a signal that tightens the vessels in your legs. That one gets its own episode. For now: the rent isn't only oxygen.
WASTE
Smaller, faster breaths leave a larger share where no gas exchange happens.
WORK
The neck pulls air against a rib cage that does not widen freely.
BLOOD
Fatigued breathing muscles can trigger a reflex that restricts blood flow to the legs.
What the breath is actually telling you
This is where I go further than the papers, because it's what I see in the room.
Fast, high, chest-led breathing is not a technique problem. It's a report. It's what a body does when it has energy with nowhere to go.
Every high performer I work with learned, somewhere, to brace. Most of them learned it long before sport, in the place they grew up. Pressure, stress, expectations, and a body that found the adjustments that let it live there. Holding the belly. Lifting the chest. Keeping the breath high and quiet. Those were not mistakes. They were the best adaptations a young body could find for that environment, and they worked.
Some learned it later, in the gym, where they were taught to brace and told it was core. And sport did the rest. Years of training on top of a held trunk set the pattern in concrete.
Either way, the brace became the resting state. And a body at rest that is organized around a threat pays rent on that threat every minute of the day, on the bike and off it.
So when I say the share is not fixed, I mean this. Your rent is not set by your lungs. It's set by how your body holds itself when nothing is happening. Change that, and the breath changes without being told to. That's not a claim. That's the 26 liters.
A rider, three weeks
A rider I work with came in with breathing as his limiter. Not fitness. Not legs. Breathing. Three weeks of conscious attention to the breath, together with releasing the chronic tension in his trunk and around his diaphragm. No breathing device.
Control of the breathing muscles improved. The legs felt much better. And breathing stopped being the thing that ended his efforts. That's how he described it, and I'm not going to improve on his words.
One honest thing about three weeks. It's enough to feel the difference. It's not enough for the new pattern to become automatic. It took years to distort that breathing and make it expensive, and it takes time to make it efficient again: for the brain to start noticing the muscles it stopped noticing, for the trunk to become soft and adaptable, able to expand and then compress. That's the work. Three weeks is where it starts to show.
What actually changes it
Not a breathing technique. Every technique you've tried works inside whatever room your trunk has. None of them can change the room.
The order is this. Structure first: the ribs, the belly wall and the diaphragm have to be allowed to expand and then compress. Then control of the breathing muscles. Then the breath slows and deepens on its own, and the nose stays the route for longer. You don't force the last step. It arrives.
And measure the price, not the ceiling. Pick a fixed, moderate power you can repeat. Note your breathing rate, how hard it feels, your heart rate, and the pace at which your mouth has to open. Repeat it in a few weeks. If air moved, heart rate and effort all fall at the same watts, your rent went down. No lab needed.
Your ceiling is a fact about your body. Your rent is a fact about how you live in it. Only one of those moved for me in three months, and it was the one that decides the last third of a race.
The cost of breathing, one of five.
JB
Sources and scope +
- One breath in ten. Aaron EA, Seow KC, Johnson BD, Dempsey JA. Oxygen cost of exercise hyperpnea: implications for performance. Journal of Applied Physiology. 1992;72(5):1818-1825. doi:10.1152/jappl.1992.72.5.1818.
- Women pay more. Dominelli PB, Render JN, Molgat-Seon Y, et al. Oxygen cost of exercise hyperpnoea is greater in women compared with men. Journal of Physiology. 2015;593(8):1965-1979. doi:10.1113/jphysiol.2014.285965.
- The wrong test. Illi SK, Held U, Frank I, Spengler CM. Effect of respiratory muscle training on exercise performance in healthy individuals: a systematic review and meta-analysis. Sports Medicine. 2012;42(8):707-724. doi:10.2165/11631670-000000000-00000.
- The rent is also blood. Harms CA, Babcock MA, McClaran SR, et al. Respiratory muscle work compromises leg blood flow during maximal exercise. Journal of Applied Physiology. 1997;82(5):1573-1583. doi:10.1152/jappl.1997.82.5.1573.
Scope note. The percentages above come from exercise physiology studies and describe study averages, not a fixed cost for every athlete. The sex comparison included nine women and nine men. The blood-flow study included seven male cyclists working at maximal effort. The 360-watt comparison is Jani's own repeated field measurement, not a controlled trial. The three-week rider account is an anonymized client report. This article is education, not medical advice.