Quantifying Bicycle Setup Differences on the Road: Field Estimates of CdA, Crr and Power Demand

Individual bicycle configurations can be assessed in situ using power and speed measurements without a velodrome or wind tunnel. We apply an integrated energy-balance workflow to five configurations tested by one rider on a road section. Five nominal power levels, each ridden in both directions, yield 25 paired observations. Kinetic-energy changes are corrected explicitly; gravitational work cancels between spatially matched outward and return sections. A shared effective rolling-resistance coefficient Crr and a separate aerodynamic drag area CdA for each setup are estimated jointly. The fitted drag areas range from 0.2288 to 0.3601m^2, with a common Crr=0.00551 +/- 0.00047. At 40km/h, the estimated differences are 17.4 +/- 5.5W favouring the EZ-covered over the uncovered 80mm rear-wheel configuration and 64.3 +/- 4.0W favouring the aerobar position over the drops on the tested road bike. The helmet swap yields a difference of 3.5 +/- 2.8W. Reported uncertainties combine statistical variability and assumed input uncertainties and represent one standard deviation. The study demonstrates a practical workflow for quantifying individual setup differences and connecting resistance estimates to power-speed predictions under specified reference conditions.