Sunspots host a large variety of oscillatory phenomena, whose properties depend on the nature of the wave modes and the magnetic and thermodynamic structure of the spot. Umbral chromospheric oscillations exhibit significant differences compared to their photospheric counterparts. They show an enhanced power and a shorter dominant period, from waves with an amplitude of a few hundred meters per second in the five-minute band at the photosphere, to amplitudes of several kilometers per second in the three-minute band at the chromosphere. Various models have been proposed to explain this behaviour, including the presence of a chromospheric resonance cavity between the photosphere and the transition region. Jess et al. (2020, Nature Astronomy, 4, 220) claimed the detection of observational evidence supporting this model, obtained from the comparison of spectropolarimetric observations and numerical simulations. Here, it is shown that the observational insight reported by Jess et al. is not a common property of sunspots. More importantly, numerical modelling also shows that it is not an unequivocal signature of an acoustic resonator.