We used a snapshot at redshift 0.1 of the TNG-100 data and selected the galaxies in the data according to their HI mass M_HI and star mass M_star (M_HI>2e8 M_sun or M_star 2e8 M_sun, where the M_sun is the mass of sun), and found that about one in sixteen galaxies has no gas.
In the sub-sample of the main galaxy sample, we selected 4316 luminous red galaxies based on a tanh function and a magnitude threshold (m_r < -17.9, where the m_r is the apparent magnitudes in r−band), and found that more than one third of the galaxies had no gas.
We wonder that if these galaxies that don't contain gas are fake data which should be thrown away, or if these galaxies are real. If they're real, we want to know why they don't contain gas.
Are you selecting only centrals?
if not, you also include satellites, and red satellites will generally be gas poor (and/or contain "no gas"). This is a physical effect due to environmental processes e.g. stripping. See e.g. Ayromlou+21 (Fig 13).
(If you select only centrals, you will likely find a very small fraction that still contain no gas, and the vast majority of these will be backsplash galaxies).