Drag force calculator

Work out the air drag on anything from its speed, drag coefficient and frontal area, with the air density for your altitude and temperature. Then see what drag means on a bicycle: the power a rider needs at a given speed, and the speed a given power buys, with rolling resistance, hills and wind. Free, no sign-up, runs in your browser.

It opens on the numbers from Skill-Lync's "flow over a bicycle" challenge (Cd 0.8, A 0.1 m², ρ 1.2 kg/m³, 1 to 50 m/s). The third tab checks a result from your MATLAB or Python script and names the slips Skill-Lync found when it read 728 of these reports.

F = ½ρv²CdA · U.S. Standard Atmosphere 1976 · measured cyclist drag areas · Updated 20 Nov 2026

Defaults: the bicycle challenge's numbers. Type an altitude and a temperature and press the button to use real air instead of 1.2 kg/m³; leave the temperature empty for the standard atmosphere.

Drag force at 10 m/s4.8 N36 km/h; the weight of 489 g
Power against the air48 WF × v
Dynamic pressure60 Pa½ρv², Cd·A 0.08 m²
05010001020304050F = ½ρv²CdA, Cd 0.8, A 0.1 m²velocity (m/s)drag force (N)02.557.500.250.50.7511.251.5at v = 10 m/s (36 km/h): a straight linedrag coefficient Cddrag force (N)

    What does the flow over a bicycle challenge ask?

    Write a MATLAB (or Python) program that calculates the drag force on a cyclist and plots it twice: against velocity at a fixed drag coefficient, and against drag coefficient at a fixed velocity. The usual inputs are Cd = 0.8, a frontal area of 0.1 m², an air density of 1.2 kg/m³ and velocities from 1 to 50 m/s. It is one of the first scripting challenges in Skill-Lync's MATLAB for Mechanical Engineers and Python for Mechanical Engineers courses.

    Skill-Lync has 790 public reports with "drag force", "cyclist" or "bicycle" in a programming title (728 distinct once copies are removed). Of those whose code text states the values, 123 of 229 use Cd = 0.8, 90 of 195 use A = 0.1 m², 153 use ρ = 1.2 and 42 use 1.225, and 47 of 135 run the velocity to 50 m/s.

    How do you calculate drag force?

    Drag is the force the air exerts on a body against its motion through it. For a bluff body such as a cyclist, a car or a ball it is

    where ρ is the air density, v the speed of the body relative to the air, CD the drag coefficient (a number that captures the shape) and A the reference area, the frontal area for bluff bodies. ½ρv² is the dynamic pressure, the pressure the air would exert if brought to rest. With the challenge's numbers at the top speed:

    The power needed to keep pushing through the air is force times speed, here 6 kW at 50 m/s. That is about 34 times what a real rider needs to hold 36 km/h (worked out below), and 50 m/s is 180 km/h: Filippo Ganna's UCI Hour Record ride of 8 October 2022 averaged 56.792 km/h on a velodrome.

    Why is drag against velocity a parabola and drag against Cd a straight line?

    Drag depends on v², so the first plot is a parabola: doubling the speed makes the force 4 times larger, at any speed. 36 reports describe how the velocity curve grows (with the square of speed, quadratic, parabolic or exponential); 5 of them call it exponential, which it is not (an exponential would multiply the force by a growing factor for each step in speed). Drag is directly proportional to Cd, so at a fixed speed the second plot is a straight line through the origin with slope ½ρv²A.

    05010001020304050F = ½ρv²CdA, Cd 0.8, A 0.1 m²velocity (m/s)drag force (N)02.557.500.250.50.7511.251.5at v = 10 m/s (36 km/h): a straight linedrag coefficient Cddrag force (N)

    The challenge's two plots with its default numbers: drag against velocity at Cd 0.8 (top) and drag against Cd at 10 m/s (bottom). The dots mark 10 m/s and Cd 0.8: 4.8 N.

    How much drag does a real cyclist face?

    The drag coefficient and frontal area of a rider are hard to separate, so wind tunnels and field tests usually report their product, the drag area CDA. Measured values:

    PositionCDA (m²)Drag at 36 km/hPower against airSource
    Time trial, on aero bars0.21112.9 N129 WDefraeye et al. 2010 (wind tunnel)
    Road bike, hands in the drops0.24314.9 N149 WDefraeye et al. 2010
    Time trial, pedalling, 6 riders0.26916.5 N165 WMartin et al. 1998 (mean of 6)
    Road bike, upright, hands on the top of the bar0.2716.5 N165 WDefraeye et al. 2010
    Mountain bike, seated0.35721.9 N219 WBertucci et al. 2013 (field)
    Challenge default (Cd 0.8 × 0.1 m²)0.084.9 N49 Wthe brief

    Drag and power at 10 m/s (36 km/h) in still air with ρ = 1.225 kg/m³.

    Measurements vary between studies and riders: the same positions measured by Jeukendrup and Martin (2001) gave 0.358 m² upright and 0.307 m² in the drops. Either way, the challenge's rider meets 2.6 to 4.5 times less drag than a real cyclist. That does not change the shape of either curve, which is what the exercise is about, but it does change every number. If you want the plots to describe a real bicycle, use CDA ≈ 0.243 m² and speeds up to about 15 m/s.

    How much power does it take to ride a bicycle at a given speed?

    On a steady ride the rider's power goes into air drag, rolling resistance and climbing, and a little is lost in the chain and gears. The road-cycling model validated by Martin and colleagues (1998) has the same terms:

    with v the ground speed, vw the headwind, Crr the rolling resistance coefficient, m the mass of rider and bike, θ the slope and η the drivetrain efficiency. On the flat at 36 km/h, an 80 kg rider and bike in the drops need about 178 W at the pedals, 86% of it against the air. Because the air term grows with v³, going from 36 to 40 km/h costs 33% more power for 11% more speed.

    Martin's model matched power measured on the bike with R² = 0.97 and a standard error of 2.7 W. Their published example (90 kg, CDA 0.2565 m² plus 0.0044 m² for the spokes, 8.36 m/s into a 2.55 m/s headwind, 0.3% grade) gives 209.6 W with this calculator's terms. Martin's own total, which adds wheel bearings and a small speed-up, is 213.3 W, and the power meter read 218 W.

    020040001020304050total at the pedals · air drag · rollingground speed (km/h)power (W)

    Power at the pedals against speed for the same rider on flat road in still air, split into air drag and rolling resistance. The dot marks 36 km/h.

    What air density should you use?

    The standard sea-level density is 1.225 kg/m³ (15 °C, 101,325 Pa). Higher up, or on a hot day, the air is thinner and drag falls in proportion. In the lower atmosphere the International Standard Atmosphere gives

    with T0 = 288.15 K, p0 = 101,325 Pa, L = 0.0065 K/m, R = 287.05 J/(kg K) and g0 = 9.80665 m/s². For a real day, keep the pressure for the altitude and use the actual temperature: sea level at 35 °C gives 1.145 kg/m³.

    AltitudeTemperaturePressureDensityDrag vs sea level
    0 m15.0 °C101.3 kPa1.2250 kg/m³100%
    1,000 m8.5 °C89.9 kPa1.1117 kg/m³91%
    2,000 m2.0 °C79.5 kPa1.0066 kg/m³82%
    3,000 m-4.5 °C70.1 kPa0.9093 kg/m³74%
    5,000 m-17.5 °C54.0 kPa0.7364 kg/m³60%

    Standard atmosphere values computed by this page; they agree with the U.S. Standard Atmosphere 1976 table to within 0.02%.

    Which slips does the check tab catch?

    • A missing ½: ρv²CdA gives exactly twice the drag force.
    • v instead of v², often from writing v^2 on a vector in MATLAB and then "fixing" the error by dropping the power. Square element by element with v.^2 (in NumPy, v**2).
    • Speed in km/h: the formula wants m/s for newtons; km/h makes the force 3.6² = 12.96 times too big.
    • Power reported as force: F × v is in watts.
    • "Proportional to velocity" or "exponential": the velocity curve is a parabola.
    • An unrealistic rider: CDA far below measured cyclists, and speeds far above what anyone pedals. Only 2 of the 728 reports mention rolling resistance, which is 35% of the effort for the rider above at 20 km/h.

    What this calculator does not cover

    • Cd that changes with speed. Cd is taken as constant. For smooth spheres and cylinders it drops sharply near a Reynolds number of a few hundred thousand (the drag crisis); a cyclist's Cd varies much less.
    • Crosswinds and drafting. Only a head or tail wind along the road is modelled. Side winds change CDA with the yaw angle (Martin's riders measured 0.269 m² head-on and 0.255 m² at 15° yaw), and riding close behind another rider cuts drag by 14% to 27% depending on the position.
    • Accelerating, bearings and tyre details. Steady speed only: no kinetic energy, wheel inertia or bearing friction, and one rolling coefficient for the tyres and road.
    • Humid air. Density is for dry air; humid air is slightly lighter.
    Learn MATLAB and Python for engineering

    A drag script is where many engineers write their first vectorised code and their first honest plot. Skill-Lync's MATLAB and Python courses for mechanical engineers build from it to solvers, curve fitting and simulation.

    Frequently asked questions

    How do you calculate drag force?

    Multiply half the air density by the speed squared, the drag coefficient and the frontal area: F = ½ ρ v² Cd A, in newtons with ρ in kg/m³, v in m/s and A in m². With the bicycle challenge's numbers (ρ = 1.2, Cd = 0.8, A = 0.1 m²) that is 4.8 N at 10 m/s and 120 N at 50 m/s.

    Is drag force proportional to velocity or velocity squared?

    To velocity squared, for anything moving fast enough to leave a turbulent wake, which includes cyclists, cars and the challenge's bicycle. Double the speed and the drag becomes 4 times larger; the power needed to push through the air becomes 8 times larger, because power is force times speed.

    What is the drag coefficient and frontal area of a cyclist?

    Measured values are usually given together as the drag area Cd·A, about 0.21 to 0.36 m² from an aero time-trial position to an upright one (see the table on this page and its sources). The challenge's Cd = 0.8 with A = 0.1 m² gives 0.08 m², far below any real rider.

    How much power does it take to cycle at 36 km/h?

    On flat, still air at sea level, an 80 kg rider and bike in the drops (Cd·A 0.243 m²) on road tyres (Crr 0.0032) need about 178 W at the pedals: 149 W against the air and 25 W against rolling resistance, through a drivetrain that is 97.6% efficient.

    What air density should I use for drag calculations?

    1.225 kg/m³ is the standard sea-level value (15 °C, 101,325 Pa). The bicycle challenge uses 1.2, which is 2% lower and fine for a hand calculation. At 1,000 m the standard atmosphere gives 1.112 kg/m³ and at 2,000 m 1.007, so drag at the same speed falls by 9% and 18%.

    Why is the drag force against drag coefficient graph a straight line?

    Because at a fixed speed, density and area, F = (½ρv²A) × Cd: the drag force is a constant times Cd. The plot is a straight line through the origin whose slope is ½ρv²A, so a steeper line simply means a higher speed or a bigger frontal area.

    Does drafting behind another cyclist reduce drag?

    Yes, a lot. In Blocken and colleagues' 2013 CFD study, validated against wind-tunnel tests, a rider following closely behind another had 27.1% less drag when both sat upright, 23.1% less in the drops and 13.8% less in the time-trial position. The leading rider also gained slightly, up to 2.6%.

    Related on Skill-Lync

    Sources

    • NASA Glenn Research Center, The Drag Equation (D = Cd ρV²A/2, A a reference area; frontal area "a more logical choice" for drag as resistance to flow) and Shape Effects on Drag (flat plate 1.28, downstream wedge 1.14, sphere 0.07 to 0.5 depending on Reynolds number, bullet 0.295, airfoil 0.045, model rocket 0.75; low-speed tunnels, projected frontal area).
    • T. Defraeye, B. Blocken, E. Koninckx, P. Hespel, J. Carmeliet, "Aerodynamic study of different cyclist positions: CFD analysis and full-scale wind-tunnel tests", Journal of Biomechanics 43 (2010): rider and bicycle CDA 0.270 (upright), 0.243 (dropped), 0.211 m² (time trial), with other studies' values in its Table 1 (Jeukendrup and Martin 2001: 0.358, 0.307, 0.240 to 0.269 m²).
    • B. Blocken, T. Defraeye, E. Koninckx, J. Carmeliet, P. Hespel, "CFD simulations of the aerodynamic drag of two drafting cyclists", Computers & Fluids 71 (2013): the trailing rider has 27.1%, 23.1% and 13.8% less drag in the upright, dropped and time-trial positions.
    • J. C. Martin, D. L. Milliken, J. E. Cobb, K. L. McFadden, A. R. Coggan, "Validation of a mathematical model for road cycling power", Journal of Applied Biomechanics 14 (1998): the power model (aerodynamic, rolling, bearing, grade and kinetic terms over a chain efficiency of 97.698%), R² = 0.97 and a standard error of 2.7 W against SRM power, CDA 0.269 ± 0.006 m² for 6 riders in the time-trial position, Crr 0.0032 (mean of 0.0027 to 0.0040 for road clinchers on smooth asphalt, after Kyle 1988), and the appendix bout used as a test case here.
    • W. Bertucci et al., Journal of Sports Sciences (2013): seated mountain bikers CDA 0.357 ± 0.023 m². C. Maier et al., Journal of Sports Sciences 36 (2018): 29-inch cross-country tyres off-road, mean Crr 0.0219 (0.0205 to 0.0237).
    • NOAA, NASA and USAF, U.S. Standard Atmosphere, 1976: sea level 101,325 Pa, 288.15 K, 1.2250 kg/m³, g0 = 9.80665 m/s², −6.5 K per geopotential km, and the Table I densities this page is checked against. UCI, press release of 8 October 2022: Filippo Ganna's Hour Record of 56.792 km.
    • Skill-Lync, 790 public drag force and flow over bicycle reports, read 20 Nov 2026: inputs, velocity ranges, statements about the curves.
    • Skill-Lync, scripts/verify/drag.py: checks the standard atmosphere against the 1976 table, evaluates the brief with NumPy, solves the speed for a power with SciPy's brentq and recomputes each slip from the wrong formula; the page re-checks every number against it when it is built.
    Free counselling

    From a first script to engineering simulation.

    Writing clean MATLAB or Python, checking it against physics and explaining the numbers is what CFD, vehicle and design roles ask for. A Skill-Lync counsellor can tell you which course fits where you are now.

    Talk to a course counsellor

    Free. We call you back, usually the same day.

    By submitting you agree to our Terms and Privacy Policy.

    Talk to a career counsellor

    Talk to a career counsellor

    Free 20-minute call. We'll map the right course to where you want to work.

    By submitting you agree to our Terms and Privacy Policy.