Cycling Power to Weight Calculator

Watts per kilogram is the power you can hold for a set time divided by your body weight, and it is the number that compares riders of different sizes.

Two riders can both hold 250 W. At 60 kg that is 4.17 W/kg, and at 85 kg it is 2.94 W/kg.

On a climb, most of your effort goes into lifting your body and bike against gravity. In the 10 km, 7% example below, gravity takes 87% of the rider's power. That is why W/kg is the usual number for climbing, and raw watts matter more on flat roads where air drag takes over.

W/kg says nothing by itself about flat speed, handling or racing skill. Treat it as one measure of one kind of fitness.

The calculation is one division. Everything else on the page is a lookup or a model built on top of it.

• Power to weight

  W/kg = watts ÷ kg

  Use the power for one effort length and your body weight without the bike.

• Pounds to kilograms

  kg = lb × 0.45359237

  The exact definition of the pound. 1 kg is 2.2046 lb.

• Watts needed for a target

  watts = target W/kg × kg

  Rounded up to a whole watt so the target is met.

• Heaviest weight for a target

  kg = watts ÷ target W/kg

  Rounded down to 0.1 so the target is still met.

The table below gives W/kg for common FTP and weight pairs, with no table of levels involved.

Use the best power you have ever held for that exact length, and compare it only with the same column in the table.

Coggan picked 5 s, 1 min, 5 min and functional threshold power to reflect neuromuscular power, anaerobic capacity, VO2max and lactate threshold^[1]. He adds that these are index efforts that correlate with those abilities, and not pure tests of them. A 1 minute effort, for example, is not fully anaerobic.

Functional threshold power is most often described as the highest power you could hold for about an hour. TrainerRoad notes that the definitions in use differ^[3], so check which one your FTP number came from. If you only know a 20-minute test, convert it first with the FTP Calculator.

Coggan stresses that the values must be your very best for each length, or the profile is distorted^[1]. A ride average is not a best effort.

The table is an anchored scale, not a count of real riders, and its levels are read from overlapping brackets.

Coggan anchored the top of each column on a world champion or record holder and the bottom on an average untrained person^[1]. He thought it highly unlikely that coaches would have a large enough rider database, and that estimates from race results needed tenuous assumptions about body mass and CdA.

Anchoring also keeps a "world class" value equivalent across lengths, whether it is 5 s or 1 h^[1].

The printed table has 57 rows per column. To place you, the calculator takes the nearest row, which is his rule when a value falls between rows^[1].

The image brackets nine rows for each level, with three rows shared between neighbours, and prints no numbers for the edges. The levels below use the bold rows in the middle of each overlap as the edge between two levels. That is a reading of the figure and not a published cut-off.

The men's FTP column runs from 1.27 at the bottom row to 6.60 at the top, and the women's from 0.99 to 5.69.

Different sites use different scales, and none of the bands lines up with another.

Garmin rates FTP in five bands. For men they are Superior from 5.05, Excellent 3.93 to 5.04, Good 2.79 to 3.92, Fair 2.23 to 2.78 and Untrained below 2.23^[2]. For women the same bands start at 4.30, 3.33, 2.36 and 1.90^[2].

Garmin says its bands are based on research by Hunter Allen and Andrew Coggan^[2]. Its page does not show how the band edges were derived, so this tool does not mix them with Coggan's table.

The 3.57 W/kg of the worked example reads "Good" on Garmin's men's scale and Moderate on Coggan's. Both names are labels for the same place on different scales.

Racing category labels such as Cat 1 to Cat 5 are also in use online. TrainerRoad publishes a table of them and calls the boundaries approximate^[3]. They are not in Coggan's table image and are not used here.

Your result includes a table of what each level would take at your weight and power, and the answer for any W/kg you type.

The watts needed are the target W/kg times your weight, rounded up to a whole watt. The heaviest weight is your power divided by the target, rounded down to 0.1, so the target is still met.

This is arithmetic. Changing your weight can change your power too, so the two numbers do not move independently, and the page gives no advice on either.

Add a climb and the calculator turns your 5-minute or FTP power into a steady speed and time on that gradient.

It uses the power balance Martin and colleagues tested against power measured on the road. Their modelled and measured power agreed with R² = .97 and a standard error of 2.7 W^[4]. Your power, after drivetrain loss, has to cover air drag, rolling resistance, gravity and wheel bearing friction.

The inputs are the climb length and gradient, the weight of your bike and kit, and CdA and Crr. The defaults are 0.3 m² and 0.0042, and they are starting points and not facts about you.

A review of wind tunnel work reports that drag area for the time-trial position varies by as much as 40% between studies^[6].

The Crr default of 0.0042 is the median of the 184 road tires with a visible high-pressure result in one lab drum-test table. That is 14.1 W per tire at 28.8 km/h and a 42.5 kg load^[5]. It is a tire lab figure and not a measurement of your road.

The calculator also takes the drivetrain efficiency of 97.7% and the wheel bearing formula from Martin and colleagues^[4]. They assumed that efficiency from a 2.36% gap between two power measurements, so treat it as an assumption and not a property of every drivetrain.

The model assumes a steady speed, no wind and standard air density. It does not model descents, drafting or changes of pace, and it has not been checked against a power meter on your bike. For flat rides and conditions, use the Cycling Performance Calculator, and for a whole course the Cycling Race Predictor.

The power you can hold depends on how long you ride. The calculator warns when the climb time is far from the effort length that drove it, using a window of half to twice the length. That window is a fixed setting and not a published threshold.

The table has no age correction, so age is only used to show a separate sample of real users.

If you enter an age with your FTP, the page shows TrainerRoad's median FTP for your age group and sex^[3]. It is a filtered, anonymised sample of TrainerRoad users and the filtering is not described, so it describes that group and not riders in general.

Nothing here predicts your race result. Coggan notes that the best measure of competitive ability is actual race performance, not power output^[1].

Some figures common elsewhere are left out because no source for them could be traced.

• –

  Rider categories

  Cat 1 to Cat 5 labels on W/kg bands are not in Coggan's table image, and the pages that publish them do not show how the bounds were set.

• –

  Pro and Tour W/kg figures

  Coggan's top row is an anchor for a world champion or record holder. It is not a measurement of any named rider, so none is quoted.

• –

  Age-specific standards

  Coggan chose not to build them. TrainerRoad medians are shown only as a separate sample.

• –

  Ideal or target body weight

  No source ties a W/kg to a best body weight, so the page makes no such suggestion.

• –

  Power meter prices and accuracy

  No sourced comparison was found, so none is listed.

• –

  FTP testing and zones

  Use the FTP Calculator to get FTP from a test and the Cycling Power Zones Calculator for zones.

• –

  Drag area and tire resistance

  Measure or estimate CdA with the Cycling CdA Calculator. The VO2 Max Calculator covers aerobic capacity.