Taking an Occam's Razor approach to Panda, because regardless of the complexity (or lack of it) to the update, there is probably one overarching principle behind it:
- Content farms generally punch above their weight when it comes to SEO (that after all, is there business model), meaning a very skewed direct:search ratio
- Usable and trustworthy websites usually have a pretty healthy direct:search ratio. It's pretty hard to build a healthy direct % with a horrible website (unless you are craigslist :) )
Expanding on this, someone in another thread suggested that Google probably can't work out the source of articles anymore. Direct:search seems to me like it could be a pretty good signal for working this out. For example, ticket A lives on 100 ticket websites, but there is one site with a high direct:search ratio (Ticketmaster); that site is probably the source.
Extrapolating this to the sites "that do not fit the mold", a lot of them were seriously punching above their weight on SEO compared to branded search:
daniweb.com
blogcritics.org
cinemablend.com
digitaltrends.com
[
google.com...]
Digitaltrends was getting almost 100k visits a day, with almost none of that being direct
[
trends.google.com...]
[
trends.google.com...]
This would also explain sites like prnewswire, which would have very little direct traffic / branded search.