Here's something I've seen more times than I care to think about: a user with a history of posting well-researched, well-written answers drops into a question and posts something lousy, inaccurate, hastily-written, unhelpful or just plain wrong.
Everyone has their off-days and blind spots, and past performance is no guarantee of future results. Apart from the rare instances where someone answers a question on a subject where they are the last word (say Steve Kondik shows up to answer a Cyanogen question or something), it's usually a bad idea to blindly trust someone's answer simply because they've given you good answers in the past.
So this is not likely to ever be implemented on any SE site.