AbstractThis study seeks to design a city-level trip speed performance index (CTSPI) providing an alternative aspect in quantifying the traffic performance of an entire city. Another objective is to disentangle the original CTSPI time series into several featured patterns, including a trend pattern, two seasonality patterns, and a remainder pattern. The big data in the form of observations of ride-sourcing trips in Beijing, China were adopted. This study also introduces a state-space model, the TBATS [trigonometric, Box–Cox transformation, auto-regressive moving average (ARMA) errors, trend, and seasonal components] filtering procedure to decompose the CTSPI time series. This study adopts Beijing as a representative example because the city has very typical and complicated traffic performance patterns. The proposed CTSPI directly reflects the average trip speed, normalized by the best performance supplied by the corresponding infrastructure systems. After filtering out fluctuation, noise, and irregular patterns, it reveals a smooth and clear-cut trend in the evolving process of the city’s traffic condition, which was never previously disclosed and is important in understanding the macroscopic long-term tendencies of the city’s traffic performance. The results indicate that the CTSPI is capable of capturing the traffic performance of the city well and can sense the influence of special dates or major events, such as the Beijing 2022 Olympic Winter Games, advising tremendous application of traffic management. City-level macroscopic traffic performance is usually measured as index quantities and used to assess traffic situations in different cities. Most often, it is utilized to provide a quantified impression of the degree of congestion to the public, or as traffic-congestion criteria for ranking cities. This study illustrates the importance of measuring city-level macroscopic traffic performance, especially on a daily basis as is appropriate for gauging the impacts of many macroscopic factors on city-level traffic situations.