Effect of event classification on the Tsallis thermometer

We analyze identified hadron spectra in pp collisions at s = 13 TeV measured by ALICE within a nonextensive statistical framework. Spectra classified by multiplicity, flattenicity, and spherocity were fitted with the Tsallis-Pareto distribution, and the parameters were studied on the Tsallis thermometer. Multiplicity and flattenicity classes follow a previously observed scaling, while the nonextensivity parameter shows a distinct sensitivity to the spherocity. A data-driven parametrization confirms a proportionality between the Tsallis temperature and mean transverse momentum, offering a simple estimate of the effective temperature. These results highlight the ability of the Tsallis thermometer to capture both multiplicity and event-shape effects, linking soft and hard processes in small systems.