From Jarvis 2016 (DES SV shapes) "For NGMIX... we find significantly smaller selection biases when we use(S/N )r to select galaxies for shear measurement, as compared to using (S/N )w ."

This is because "if the PSF is approximately round, as is the case for our data, then more elliptical galaxies will have a lower estimated (S/N )w than round galaxies (holding flux constant). Thus if galaxies are selected according to their measured (S/N )w , the resulting galaxy catalogue will have a selection bias towards round shapes, which will bias the overall mean shear.

One null test we can do is plot vs snr

How to calculate snr_r

1. calculate round r50 Question: should we use T_r to get r50? Judging from this quote we do: "That is, we take the model of the galaxy profile and apply a shear such that its ellipticity becomes zero. We do the same for the PSF, convolve these two profiles together, and then integrate over the pixels. "
2. use the round r50 and [draw a new galaxy](in erin sheldon's ngmix code ngmix/fitting/galsim_results.py) with chosen model (gaussian, esponential, etc). see calc_s2n_r(self, pars) (in k space). Question: what do we do about the weights? from the code, it looks like we keep them the same. do we alter the psf?