CdA in track cycling: what it is, how to measure it and why it determines your time

At 55 km/h, 87% of a cyclist's applied power is dissipated overcoming air resistance. That percentage rises to 92% at 60 km/h. If you want to gain five seconds in 4 km, the engine is not in the legs, it is in the position. And position is measured by a single number: CdA.

What CdA is (and what it isn't)

CdA is the product of drag coefficient (Cd, dimensionless) times frontal area (A, m²). It is an effective drag area: how many square metres of air you are moving. An elite track cyclist in pursuit position has a CdA between 0.175 and 0.200 m². A well-positioned amateur on a road bike is around 0.300. A tourist, 0.400.

CdA is not your body size. It is not your build. It is how much air your specific combination of body, helmet, skinsuit, bars and bike displaces in the exact position you take up in racing. Changing position on the same bike can move CdA by 0.020, which is 26 W at 57 km/h.

The equation behind it

Aerodynamic power is P = F · v, so P = ½ · ρ · CdA · v³. That is the key: power scales with the cube of velocity. Doubling speed multiplies power by 8. Reducing CdA by 5% saves 5% of power at any speed. In a 4-minute pursuit, that is half a second per lap.

What a CdA is worth in time

Five thousandths in CdA = 1.3-1.5 s in 4 km. Nobody trains to give away seconds.

How to measure CdA without a wind tunnel

There are three methods a coach without specialist facilities can use:

1. Coast-down method (deceleration curve)

The rider reaches a stable speed (e.g. 50 km/h) on a long straight, stops pedalling and their deceleration is recorded with a high-frequency GPS and power meter. Fitting the physical model (mass, Crr, ρ) to the decreasing speed profile solves for CdA. Typical precision: ±0.005-0.010. Requires calm air (wind < 1 m/s) and flat trajectory.

2. Constant power velodrome laps (Chung method)

The track standard. The rider does complete laps holding constant power. Average speed per lap is calculated and the force-balance equation is solved iteratively until integrated energy matches. Precision ±0.003 with good data. Needs a calibrated power meter and well-known velodrome ρ.

3. Iso-power comparison method

Rider measured in two different positions holding the same average power. Speed difference gives the CdA difference directly. Useful for a/b testing of bars, helmet or torso position. Does not give absolute value but delta with ±0.002 precision.

Where CdA sits on a track cyclist

Typical breakdown of total CdA (0.190 m² on an elite pursuiter):

• Body (torso + arms + head): ≈ 0.120 m² (63%)
• Legs and hips: ≈ 0.038 m² (20%)
• Bike, wheels and accessories: ≈ 0.032 m² (17%)

The body dominates. That's why raindrop helmets (skinsuit + speedsuit + long-tail helmet) win more seconds than a bar change. And why obsessing over frame aerodynamics when torso position leaks 25 W is priority mis-ordered.

Corner lean: the effect that doesn't show up in the tunnel

On a 250 m velodrome with 42% banked corners, the rider spends 50% of the time leaned. Underwood (2010) showed that corner lean can raise effective CdA by 4-7% versus upright position. Modelling a pursuit without lean underestimates required average power by 8-14 W. The app lets you toggle this ("Model corner lean") and see the difference.

References: Underwood & Jermy (2010), Procedia Engineering. Chung R. (2005), CdA regression estimation method. Blocken B. et al. (2018), J. Wind Eng. Ind. Aerodyn.. Debraux P. et al. (2011), Sports Biomech..