We present a set of nonlocal thermodynamic equilibrium steady-state calculations of radiative transfer for one-year-old Type II supernovae (SNe) starting from state-of-the-art explosion models computed with detailed nucleosynthesis. This grid covers single-star progenitors with initial masses between 9 and 29 M⊙, all evolved with the code KEPLER at solar metallicity and ignoring rotation. The [O I] λλ 6300, 6364 line flux generally grows with progenitor mass, and Hα exhibits an equally strong and opposite trend. The [Ca II] λλ 7291, 7323 strength increases at low 56Ni mass, at low explosion energy, or with clumping. This Ca II doublet, which forms primarily in the explosively produced Si/S zones, depends little on the progenitor mass but may strengthen if Ca+ dominates in the H-rich emitting zones or if Ca is abundant in the O-rich zones. Indeed, Si–O shell merging prior to core collapse may boost the Ca II doublet at the expense of the O I doublet, and may thus mimic the metal line strengths of a lower-mass progenitor. We find that the 56Ni bubble effect has a weak impact, probably because it is too weak to induce much of an ionization shift in the various emitting zones. Our simulations compare favorably to observed SNe II, including SN 2008bk (e.g., the 9 M⊙ model), SN 2012aw (12 M⊙ model), SN 1987A (15 M⊙ model), or SN 2015bs (25 M⊙ model with no Si–O shell merging). SNe II with narrow lines and a low 56Ni mass are well matched by the weak explosion of 9–11 M⊙ progenitors. The nebular-phase spectra of standard SNe II can be explained with progenitors in the mass range 12–15 M⊙, with one notable exception for SN 2015bs. In the intermediate mass range, these mass estimates may increase by a few M⊙, with allowance for clumping of the O-rich material or CO molecular cooling. [ABSTRACT FROM AUTHOR]