I for one am not surprised that when one of the top 3 most-visited-domains in the world goes down, other sites are visited more often, in particular news sites.
It's framed on such a way as to imply that people are more informed without Facebook, and maybe that is true, but that isn't all that is going on there.
- The increases far exceeded what was lost. The dip in the social line was offset three-fold when you add up external, search and direct. They're not to be averaged.
- Unmentioned, but reflected in graph: people using facebook tend to share articles far more than they read them. As in social share count > pageview count. Sometimes by a factor of 100.
After all, what does facebook really do except just repost news from low quality article factories?
It could also just be looking for their content fix.
Most people are probably aware of the names of several major news organisations in their country or city, but it's less obvious what the domain name should be - what is written in full? what is abbreviated?
If you prohibit the use of a lookup system (that's what you did, effectively), some people will be able remember the domain name (it's easier than a telephone number), but it's not hard to guess people will find the system much less useful.
It's a little like saying "if everyone had to use a CLI to start the programs they want to use on their phone, more people would watch TV".
Which is in itself a major news story. Before talking about facebook v. other new sites we mush take account of the greater number of people looking for news content regardless of platform. The increase might more people looking for news rather than people having to go elsewhere than facebook.
There's now an obvious conflict of interest with regard to news sites reporting on Facebook. Their generalized antagonism towards Facebook, and the tech industry has a pretty clear root: they're getting outcompeted, thus the calls for regulation.
Also remember that media companies tried collaborating with Facebook. Remember Pivot to video? Only to discover that Facebook inflated their stats on video views.
Of course, Facebook was sorry.
Furthermore, many of the original landmark copyright decisions regarding, say, Google; were specifically decided to be fair use... partially on the basis that things Google did were sending monetizable traffic back to web publishers. Facebook's actions take the form of these fair uses, but they use their control over their platform to reclaim the value they would ordinarily be creating for publishers. This puts those exceptions to copyright in jeopardy - and that's exactly what we're seeing in, say, the EU.