I've posted the data several times.
It's up to you to determine where the sample size minimum needs to be to make the comp have merit.
Over the years, I have posted, pitcher by pitcher, year by year and also career numbers pitcher by pitcher. The evidence is not only clear, it is overwhelmingly clear. The disparities are consistent and significant in most cases.
We can argue about randomness and other variables being in play, but the facts are clear: most pitchers with large sample sizes with 2 or more Sox catcher do way better with the catcher not names Vaz.