Public transport performance is a central concern in metropolitan management and spatial planning because travel-time competitiveness influences modal split, service attractiveness, and the efficiency of network design. This study examines the comparative in-vehicle time (IVT) and travel distance of metropolitan bus services and private cars in Metropolis GZM (Upper Silesian and Zagłębie Metropolis) in Poland. The analysis covers three datasets: all metropolitan lines (AML), main metropolitan lines (MML), and feeder metropolitan lines (FML). Using timetable-based bus travel data and matched Google Maps car routes for a consistent late-evening weekday comparison, the study evaluates cumulative travel from each line origin to successive stops and then applies a full-sample and reduced-set benchmarking procedure to check whether the main ranking is stable after excluding outlying lines. The results show that buses are substantially less time-competitive than private cars, although the scale of the gap differs by line hierarchy. For AML, the average relative IVT is 237.6%, compared with 177.0% for MML and 247.9% for FML. Average relative distance values are materially lower, at 128.4%, 111.0%, and 133.4%, respectively, indicating that the principal disadvantage is temporal rather than geometric. A descriptively pronounced increase appears beyond the 40th stop in AML and FML, while no comparable jump is observed for MML. Benchmarking by synthetic indicator identifies AP as the strongest overall line, M107 as the strongest feeder line, and M11, M100, and M108 as the most consequential underperformers. The hierarchy-sensitive ordering is preserved in the reduced-set values, supporting the robustness of the central comparative conclusion. The findings support the managerial value of a hierarchical network structure while showing that competitiveness deteriorates as route length increases. Because the evidence is based on a single late-evening observation window and compares timetable and platform-estimated travel times, the conclusions are best interpreted as structurally informative planning benchmarks rather than full-day causal estimates.