Meet Kondy, the engine inside your muscles

The short version

Your muscles cannot burn food. Something has to convert it first, and that converter is a tiny engine called a mitochondrion. Training builds you more of it.

• Food is not energy. It is raw material, and it has to be converted.
• The converter is the mitochondrion. Kondy is one. You have far too many to count.
• Out on the bike, using oxygen gets you about 10 times more from the same fuel.
• Training builds you more engine, partly by growing the mitochondria you already have.
• Give a training block 8 to 12 weeks before you decide it is not working.

Food is not energy

You ate breakfast three hours ago. Right now your legs are turning the pedals. Somewhere in between, that breakfast became movement.

Here is the odd part. Your muscles cannot use food. Not toast, not pasta, not the gel in your back pocket. A muscle can only spend one thing, and food is not it.

Think of food as money sitting in your bank account. Your muscles only take cash. Something has to go to the machine and withdraw it — over and over, every second you are moving.

That cash has a name: ATP. You do not have to remember it. You only have to remember three things about it. It is the one thing a muscle can spend. You carry almost none of it. And you rebuild it, constantly, for as long as you are moving.

That rebuilding is what endurance really is. Everything else is detail.

I am the cash machine. You do not store energy in me — you store food, and I hand your legs the only thing they will take.

Meet Kondy

The thing that does the converting is called a mitochondrion. It lives inside your cells. It is small enough that a single muscle fibre holds far more of them than anyone could sit down and count.

That is where our name comes from. Kondya. Kondy is a mitochondrion.

So nobody counts them. What scientists measure is how much of the muscle they fill. Six weeks of training can push that up by more than half.^[lundby]

Here is what one looks like from the inside.

The part that matters is the folded ribbon in the middle. That folded surface is the workbench, and the converting happens on it. More folding means more workbench. More workbench means more energy per second.

That is the difference between a rider who can hold a steady pace while talking, and one who cannot.

Same food, two ways to spend it

Your body has two ways to do the conversion, and they are not equally good.

The first one uses oxygen. It takes a few seconds to get going, but it gets almost everything out of the food.

The second one skips the oxygen. It is instant, which is why you can jump on the pedals before you have even taken a breath. But it barely opens the fuel up. It takes one quick bite and passes the rest along, almost whole.

Nothing is actually thrown away, by the way. What it passes along is lactate. And lactate is not waste: your heart and your slower muscle fibres burn it a few minutes later, using oxygen.^[brooks]

How much better is the oxygen route? Out on the bike, from the same fuel, it rebuilds roughly ten times more ATP.

This is why almost all endurance training is about improving the oxygen route. The fast route is there for the last two hundred metres. The oxygen route is there for everything else.

I only work with oxygen. That is not a weakness. It is the reason you can ride for six hours instead of forty seconds.

Your engine grows, but not quite how you think

Most people picture training as adding more little engines. Under a microscope it turns out to be stranger than that.

In one study, twenty-one men trained for six weeks. Doctors took a sliver of thigh muscle before and after, and photographed it. Afterwards, mitochondria filled about half as much again of the muscle. But in each slice, the researchers did not see more of them than before. The ones already there had grown.^[lundby]

Be careful with that, though, because counting slices is a rough way to count anything. Inside a muscle fibre, mitochondria link up into connected clusters — more like a small power grid than a shelf of batteries.^[glancy] Slice a grid and you see plenty of little pieces, but you have not really counted the grid. (Most of that mapping was done in mice. The one detailed map of human muscle found ours looser and less joined up. Still a network, just a baggier one.^[vincent])

And other studies point the other way. One group trained harder and for longer: twelve weeks of tough intervals, not six weeks of steady riding. There, researchers did find more of them, not just bigger ones.^[ruegsegger]

So here is the honest version. Your engine gets bigger, and your body really does build new mitochondrial material to do it. Whether it mostly grows the parts you already have, or adds new ones, seems to depend on how hard and how long you train. That part is genuinely not settled.

Either way, what matters to you is the same. After a block of training there is more engine than there was before.

Easy hours build it, but not only easy hours

You will read everywhere that easy riding builds mitochondria and hard intervals only raise your ceiling. That is not quite what the research says.

The honest version is simpler:

Easy hours win because they are the only pace you can hold for ten hours a week without falling apart. That is their real advantage. Not that they are magic.

Nothing you feel this week means anything

The instruction to build starts during the session itself. The building happens afterwards, over the following days.

But the part you would actually notice — holding more speed at the same heart rate — takes weeks.

This is the most useful thing on this page. If you judge a training block by how you feel on Wednesday, you will change it before it has had any chance to work.

Nothing I build shows up in three days. Give me a month before you decide it is not working.

What Kondya can actually see

Nobody can see your mitochondria. Not a watch, not a power meter, not us. The only real way to look is a needle in your thigh, and you are not doing that.

What we can see is what they let you do. How much power or pace you hold at a given heart rate. Whether that slips over a long ride. How quickly you come back after a hard effort.

Those are clues, not measurements — and Kondya will always tell you which is which. If it cannot work something out, it says so instead of showing you a number it made up.

The Aerobic Engine card tracks how much you produce at a given heart rate, month after month, from the rides you already synced. If there is not enough history to draw the line yet, it tells you that instead of drawing a flattering one.

Open Training Analysis in Kondya

Or paste this to Kondy

Kondy, is my engine actually getting bigger, or am I just getting more tired?

Going further

Everything above is the version you need. Everything below is the version behind it, for anyone who wants the actual numbers. Nothing here changes the advice — it just shows the working.

Where the ten comes from

An ATP molecule carries three phosphate groups. Snapping the last one off releases the energy that makes a muscle fibre contract, and the leftover is immediately rebuilt into ATP again. You hold very little at any moment; the whole game is the rebuild rate, not the store.

While you are riding, your fuel is mostly the glycogen already stored inside the muscle. Burn one unit of it all the way down using oxygen and you rebuild roughly 31 to 34 ATP. Do it without oxygen and you get 3. That is where ten comes from.

If the fuel is glucose arriving in the blood instead, it is 30 to 32 against 2, which is nearer fifteen. Most articles quote that second pair, because it is the one in the textbook. On a bike, the first pair is the one you are actually living.

One more caveat. Older textbooks say 36 to 38 rather than 30 to 32, because they assumed round numbers for how much each step of the chain yields. The measured values turned out lower.^[hinkle]

A sprint barely touches your ATP

Eight trained cyclists went thirty seconds all out and had a muscle sample taken immediately afterwards. Phosphocreatine — a fast backup store that sits right next to the ATP and refills it — had crashed to about a fifth of its resting level. ATP itself was still at roughly 70%.^[bogdanis]

Your body defends ATP the way a bank defends its vault. It will spend the backup store, flood the muscle with lactate and acid, and cut your power output, all before it lets ATP fall far.

That is also why a second sprint feels worse. Refilling phosphocreatine is quick at first and then stalls: 65% after ninety seconds of rest, and still only 85% after six full minutes.^[bogdanis] Your legs are not tired in the way you assume. They are underfunded.

What the intensity research actually found

The cleanest split in the literature is not easy versus hard. It is that training volume mainly drives how much mitochondrial machinery you carry, while relative intensity mainly drives how well each unit of it works — and the two often move independently of each other.^[granata]

The largest synthesis to date pooled 353 studies and 5,973 participants. Mitochondrial content rose by a statistically similar amount with endurance training (+23%), high-intensity intervals (+27%) and sprint intervals (+27%). Per hour of exercise actually performed, sprint intervals came out about 3.9× more efficient than steady endurance riding, and about 2.3× more efficient than longer hard intervals.^[molmen] A 2025 review looking specifically at the Zone 2 claim concluded that current evidence does not support it as the optimal intensity for building mitochondrial capacity.^[storoschuk]

None of which makes easy riding pointless. You cannot accumulate fifteen hours a week of sprint intervals. Volume is the one thing only easy pace can buy.

There is a dose that makes you worse

Researchers ramped healthy, already-fit riders up to 152 minutes of high-intensity intervals per week. Mitochondrial respiration — how much oxygen each unit of machinery can turn over — fell by around 40%, and their blood-sugar control went with it. One easy week was not enough to bring it back.^[flockhart]

That is not damage. It is the engine de-rating itself. But the direction is unambiguous: there is an amount of hard training above which more makes you slower, and it is closer than most people assume.

1. primaryMeinild Lundby A-K, Jacobs RA, Gehrig S, et al. (2018). Exercise training increases skeletal muscle mitochondrial volume density by enlargement of existing mitochondria and not de novo biogenesis. Acta Physiologica 222(1):e12905 — 21 men, 6 weeks; volume density +55 ± 9%; the count that did not change was mitochondrial PROFILES PER AREA — a 2D slice count, not an organelle count
2. rctRuegsegger GN, Pataky MW, Simha S, et al. (2023). High-intensity aerobic, but not resistance or combined, exercise training improves both cardiometabolic health and skeletal muscle mitochondrial dynamics. Journal of Applied Physiology 135(4):763–774 — 38 adults, 12 weeks, same electron-microscopy approach; mitochondrial volume, NUMBER and perimeter all rose after high-intensity training
3. primaryGlancy B, Hartnell LM, Malide D, et al. (2015). Mitochondrial reticulum for cellular energy distribution in muscle. Nature 523:617–620 — mouse skeletal muscle
4. primaryVincent AE, White K, Davey T, et al. (2019). Quantitative 3D mapping of the human skeletal muscle mitochondrial network. Cell Reports 26(4):996–1009.e4 — in every human and mouse fibre analysed the network was NOT a single reticulum
5. reviewBrooks GA (2018). The science and translation of lactate shuttle theory. Cell Metabolism 27(4):757–785 — lactate is an oxidative fuel shuttled to the heart and slow fibres, not a waste product
6. rctBogdanis GC, Nevill ME, Boobis LH, et al. (1995). Recovery of power output and muscle metabolites following 30 s of maximal sprint cycling in man. The Journal of Physiology 482(2):467–480 — ATP 70.5 ± 6.5% of rest, PCr 19.7 ± 1.2%, by biopsy
7. reviewGranata C, Jamnick NA, Bishop DJ (2018). Training-induced changes in mitochondrial content and respiratory function in human skeletal muscle. Sports Medicine 48:1809–1828
8. meta-analysisMølmen KS, Almquist NW, Skattebo Ø (2025). Effects of exercise training on mitochondrial and capillary growth in human skeletal muscle: a systematic review and meta-regression. Sports Medicine 55(1):115–144 — 353 studies, 5,973 participants
9. reviewStoroschuk KL, Moran-MacDonald A, Gibala MJ, Gurd BJ (2025). Much Ado About Zone 2. Sports Medicine 55:1611–1624
10. rctFlockhart M, Nilsson LC, Tais S, et al. (2021). Excessive exercise training causes mitochondrial functional impairment and decreases glucose tolerance in healthy volunteers. Cell Metabolism 33(5):957–970 — full text not accessible; figures taken from the abstract and two secondary sources
11. reviewHinkle PC (2005). P/O ratios of mitochondrial oxidative phosphorylation. Biochimica et Biophysica Acta 1706(1–2):1–11