Same as the TNS panels of Fig. 4 but fitting to a fiducial SPT-mock data vector using the exact prediction for the SPT kernels and the Eulerian bias (red crosses, and blue crosses for the N5 model). The model parameters are those found in Table 3. The mock data vector is given scatter using a Euclid-like Gaussian covariance, using V = 8.8 h⁻³ Gpc³ and $\overline{n}=6.86\times {10}^{-4}{h}^3{\mathrm{Mpc}}^{-3}$ and the same linear bias as the mock data. The fits are also made using the same covariance. We delimit the k_max which was used to choose the best-fit bias and RSD nuisance parameters when comparing to the ELEPHANT simulations, roughly corresponding to the values found in Table 3. The black triangles indicate a modelling where no f(R) effects are accounted for, i.e. a pure ΛCDM modelling.