Aaron Spring

would this be different from https://esmtools.readthedocs.io/en/latest/api/esmtools.testing.ttest_ind_from_stats.html? I think there are many tests to be added for `xskillscore` in theory, but is the T-test applicable to forecasting and hence `xskillscore`?

I think it’s fine to have themseparate. We have very few tests which are all related to one forecast being better than another here only.

I think we should somehow come up with a pattern/wrapper that translates every skill into a skill score. - Brier Score and Brier Skill Score - RMSE and RMSE Skill...

I think so. In climpred we have a metrics class with attributes min max perfect. Maybe this would be nice to move to xskillscore

this is could be a template: https://github.com/bradyrx/climpred/blob/8a3fc954df2043f998987a1964059a9dc0d2e11c/climpred/metrics.py#L128-L140 were we would need the `perfect` attribute. But many we should overdo things. `xs` is perfect for doing easy functions `xs.metric(a,b,dim)`. skill scores...

Thanks for reporting @matteodefelice Can you please open an issue with this? Conversion from number to xr.dataarray can be easily added then

We have crpss in climpred.

For crpss look into the proper scoring example

I do RPSS from RPS here: https://renkulab.io/gitlab/aaron.spring/s2s-ai-competition-bootstrap/-/blob/master/notebooks/verification_RPSS.ipynb

generalized for perfect-models and hindcasts. https://github.com/pangeo-data/climpred/blob/14f7458c1f2e944990d7008001adab49480fe07d/climpred/reference.py#L106 I create a fake 1-member forecast from groupby(dayofyear).