Power-to-weight is one division: the watts you push divided by the kilos you carry. It answers a single question, and it is the question that decides your ride the moment the road tilts up.
On flat roads the thing you are beating is the air, and the air does not care what you weigh. Aerodynamic drag rises with the cube of speed, so what counts is the watts you put through the pedals, not the ratio.
That is why time triallists are big: 330 watts is 330 watts whether you weigh 65 kilos or 85.
When the road goes up, the thing you are lifting is you. Work against gravity is proportional to total weight, so two riders with identical watts climb at different speeds if they weigh different amounts.
A 60 kg rider at 240 watts is doing 4.0 W/kg. An 80 kg rider at 300 watts is pushing sixty watts more and sitting at 3.75 W/kg: ahead on the flat, behind at the summit.
W/kg is not one number. Over five minutes you produce a far higher ratio than over an hour, because systems are working that you cannot sustain for an hour.
This calculator uses FTP, roughly the hour, because it is the reference training plans are built on and the only one you can compare with another rider without fooling yourself. Comparing your five-minute ratio against someone else’s FTP ratio is not comparing the same thing.
There is no trick to the formula: W/kg = watts ÷ kilos. What changes the answer is which watts you put into it. A three-month-old FTP and this morning’s weight describe a rider who does not exist. Update both together.
"Is my W/kg any good?" has no absolute answer, because it depends on your age and your sex. It stops being unanswerable once you look at riders who actually train.
We took the most recent FTP of every cyclist training with us who also has a usable weight on file, kept one figure per rider, and split it by age and sex. These are the same columns the panel above compares you against.
| Group | 25% | 50% | 75% | 90% |
|---|---|---|---|---|
| Men 18–29 | 3.0 | 3.5 | 4.3 | 5.1 |
| Men 30–39 | 2.7 | 3.3 | 3.8 | 4.4 |
| Men 40–49 | 2.7 | 3.3 | 3.7 | 4.2 |
| Men 50–59 | 2.4 | 2.9 | 3.4 | 3.9 |
| Men 60+ | 2.1 | 2.7 | 3.2 | 3.5 |
| Women 18–29 | 2.4 | 3.3 | 4.1 | 4.4 |
| Women 30–39 | 2.0 | 2.8 | 3.4 | 4.2 |
| Women 40–49 | 2.3 | 3.0 | 3.5 | 4.0 |
| Women 50–59 | 1.1 | 2.0 | 2.3 | 3.4 |
The spread inside a single decade is enormous. Among men aged 40-49 the gap from the 25th to the 90th percentile runs 2.7 to 4.2 W/kg. For a 72 kg rider that is 108 watts, far more than separates a forty-year-old from a thirty-year-old.
From 30 to 50 the median does not move: 3.3 W/kg in both decades. The decline everyone expects shows up later, and even then it is about half a point across ten years.
Most of what your number says is about training, not birth date. Which is the useful half of the finding, because the birth date does not respond to training.
Published medians are the part that travels; the spread is the part that answers your question. Knowing the median of your decade tells you almost nothing when a quarter of your peers sit 0.6 W/kg below it and a tenth sit 0.9 above.
Methodology: our cyclists with a recorded FTP and a usable weight, measured on 4 September 2026. One figure per rider, the most recent, so riders who test often do not count more than once. Values outside 50–600 W were dropped as typing errors. Two biases worth stating: these are people training with a structured app, not the general population, and the female groups are small enough that their percentiles should be read as indicative. Women over 60 are absent: the group was too small to publish an honest percentile.
The panel above gives you the gap to the next level in both currencies: watts to add, or kilos to drop. It is worth reading those two numbers side by side, because at a glance the kilos look like the easy road.
An example using our own percentiles. A 45-year-old man at 72 kg with an FTP of 238 watts sits exactly on his group’s median, 3.3 W/kg. Reaching 3.7, the 75th percentile, takes 29 more watts. Or nearly eight kilos less at the same power.
Twenty-nine watts is a serious block of training: six to eight weeks of consistent work, and they stay yours. Eight kilos, for a rider who is already lean, is a different proposition. The denominator has a floor, and getting close to it takes the watts with it: below a certain energy availability power falls along with the weight, and the ratio stops improving while everything else gets worse.
The IOC consensus statement on relative energy deficiency in sport describes exactly that picture, and it is not only about performance: it covers bone health, immune function, hormones and cardiovascular health. This is not a moral argument. It is the reason losing weight works until it does not, and then turns on you.
The honest rule of thumb: treat weight as a slow lever and watts as the main one. If the number has to rise for a June gran fondo, the thing that trains between now and June is the watts.
Worth writing down, because a number you trust too far is worse than no number.
It does not test you. It divides two numbers you typed. If the FTP you enter is stale or wrong, the arithmetic stays perfect and the answer does not.
It does not know what bike you are on. The ratio that actually decides a climb is power over total weight: you, plus the bike, plus the bottles. Two kilos of bike count the same as two kilos of you.
It does not measure the climb. Gradient, altitude, wind and whether you can hold a steady effort all change your time at the same W/kg.
It does not hand you a racing category. The category tables in circulation are conventions from training literature, not measurements, and we have not measured them. What we have is percentiles of real riders, and that is what we publish.
It does not replace judgement about weight. A ratio that rises because power is flat and kilos are falling is not fitness. It is subtraction.
The claims above that come from published work, with the work cited.
Quick answers to the most common questions, so you read your number without making the wrong call.
Divide your power in watts by your body weight in kilograms: W/kg = watts ÷ kg. The figure worth comparing is the one calculated from FTP, the power you can hold for about an hour. A 72 kg rider with a 238 W FTP is at 3.3 W/kg.
Among our male cyclists aged 30 to 50 the median is 3.3 W/kg, and one in ten reaches 4.2 to 4.4. Among women in their thirties the median is 2.8. But the spread inside one decade is wider than the gap between decades, so the comparison that helps is against your own age and sex group, which is what the panel above shows you.
The category tables you see quoted across the web come from training literature, not from measured riders, and we have not measured them, so we do not republish them. What we can tell you is where you sit among cyclists who actually train: enter your numbers above and you get your percentile for your age and sex, plus what it would take in watts or kilos to reach the next level.
Almost always the watts, and that is arithmetic rather than caution. Going from 3.3 to 3.7 W/kg at 72 kg means 29 more watts, which is six to eight weeks of structured work, or nearly eight kilos less at the same power. If you are already lean, those eight kilos do not exist without taking the watts with them. Weight is a slow lever with a floor; watts are the main lever.
For comparing yourself with other riders, no: the convention is body weight, and that is what every table uses, including ours. For working out how fast you will actually climb, yes: gravity lifts all of it, you plus the bike plus the bottles. Two kilos of bike weigh the same as two kilos of you, and cost a great deal more.
Because they are different efforts. Over five minutes your anaerobic system contributes power you cannot hold for an hour, so the ratio rises. That is normal and does not mean your FTP is wrong. Compare like durations: the number on this page is at FTP.
Weight moves fast and FTP does not. Retest every six to eight weeks of consistent training and update your weight when you retest, not daily: a ratio that swings because yesterday’s dinner was salty is not telling you anything about your fitness.
Most W/kg calculators apply a public formula and stop there, or publish a single median per age group. Ours shows the full distribution for your group, the 25th, 50th, 75th and 90th percentiles, and states how it was measured and where it is biased. The median is the part that travels between sites; the spread is the part that answers your question, because a quarter of your age group sits well below that median and a tenth sits well above it.
