The deflection dynamic load allowance (DLA) of stiff bridges with high piers requires sub-millimeter accuracy. New technologies such as the vision-based optical method and GNSS are not yet recognized for use in DLA measurements due to their smaller SNR. Presently, the scaffolding method is widely utilized for dynamic deflection measurements in dynamic load tests owing to the reliability of employing rigid contact. When scaffolding is not available, engineers have to resort to a suspension hammer system. However, the mass eccentricity of the hammer, stretched-wire length, and wind will decrease the measurement accuracy. To overcome these drawbacks of the suspension hammer method (SHM), a preloaded spring method (PSM) and the related stretched-wire-spring system (SWSS) are proposed in this paper. The dynamic deflection of the coupled vehicle-bridge-SWSS was obtained by vehicle-bridge interaction (VBI) analysis. The sensitivity parameters of the PSM were analyzed and optimized to minimize the measurement error. Indoor experiments and field dynamic load tests were conducted to validate the feasibility and accuracy of the PSM. Additionally, the differences in dynamic deflection measurements between the PSM and SHM in windy environments were compared. The results show that, in a windless environment, the DLAs of the PSM are affected by the spring stiffness, stretched-wire length, and stretched-wire section stiffness, independently of the preload force. When the wind speed is less than or equal to 8 m/s and the pier height is less than 30 m, the maximum deflection measurement error of the PSM is −2.53%, while that of the SHM is −15.87%. Due to its low cost and high accuracy, the proposed method has broad application prospects in the dynamic deflection measurement of stiff bridges with high piers.