There's been a lot of talk lately about the mainstreaming of advanced baseball statistics. I'm all for it, but I think we need to recognize that there are inherent cognitive barriers that will slow this process down more than we might expect.
The "stickiest" information we consume is based on our direct observation of the sport itself, either on TV or in person. We see the most obvious things--big throws, stolen bases, athletic fielding, power pitching, power hitting, clutch hitting--but we cannot see the various buckets that comprise UZR, and it's harder to visualize how good defense contributes to bad hitting or good pitching.
Sure, we buy into other stats that aren't directly observable, ratios such as AVG and OBP, but these require far less cognitive energy than BABIP or FIP.
We don't easily separate this information from the context in which the actions occur, and so we fail to value these very important stats. On the other hand, we overvalue traits that aren't all that consistent but resonate with the pictures which stick best in our memories, like clutch hitting. And so long as our hippocampi work in this way, it's going to take a long time for the general public to buy into the new math.