Again, CERA had no value at all, when comparing Vaz to a catcher on another team. That is why no stat geek uses it.
It has very limited and specific value, when you compare how specific pitchers do, one by one, with catcher A v catcher B on their own team. Of course, other factors come into play, like maybe Price faced a mighty line-up when Vaz caught him vs when Leon did, but that can be said about any stat.
The other big criticism is sample sizes, and it is a valid concern. Since several pitchers have one catcher catch them most of the time or even nearly all the time, many sample sizes comps are widely skewed. Also, some catchers catch much better pitchers than the other, so looking at just team CERA each year is near useless.
Only when used correctly, and who does the research, can CERA have any meaning.
Who does the reseacrh?
I do, and I did. Here is what I found, in general, and I have studied this since the days of Varitek and Posada.
1. In cases where both catchers caught the same pitcher during the same year AND over several years for similar sample sizes, Vaz got less with nearly every single pitcher. Only ERod did slightly better with Vaz (career 3.98 v 4.05.
2. In cases where one catcher had a significantly smaller sample size than the other, the trend still showed Vaz got worse or much worse from each pitcher.
3. In cases where bot catchers has small sample sizes, the same was true.
Maybe, each case, one-by-one, may look too small or unbalanced to make a determination, but when you look at all the comps- big and small- together and over 90% all show each pitcher does better or way, way, way better with the back-up catcher, something must be wrong with Vaz. Despite having more time with Vaz, they did better with Leon. Having near equal time- they did better with Leon. Having more time with Leon- they did better with Leon.
Just about every single pitcher with somewhat significant sample sizes or more with both catcher did better with Vaz's back-up.