Can Facebook predict Senate winners?

[Miles]

Article.

You must be logged in to read this quote.

[badgate]

I mean....the predictions in that quoted part you posted all sound like very realistic results.

[Flake]

Well it predicts a higher voter base for the candidate, because the primary reason you like a facebook page is that you like the content/person, and if Candidate A has a larger base than Candidate B, you can trust that Candidate A's voters will go to the polls, while Candidate B's voters may not care enough to vote.

[Miles]

LA would be interesting. Maness is approaching 100K likes while Cassidy and Landrieu combined only have about 30K.

[TJ in Oregon]

I'm curious what the partisan distribution of wrong predictions was. Perhaps it's an outdated stereotype, but I'd suspect Republican voters are more likely to be olds and thus more likely not to be on Facebook.

[Bleach Blonde Bad Built Butch Bodies for Biden]

You must be logged in to read this quote.

Also, Miller has a built-in Tea Party base, so I'm not entirely surprised he's considered a dead heat with Begich.

[Miles]

McConnell's campaign page has 131K likes.

[RogueBeaver]

Cotton has over 90k likes as of December and even got noticed in local media for his extensive use of social media. On the OP, no one's winning because of social media. Though some candidates are better at it than others.

[Miles]

Cotton has over 90k likes as of December and even got noticed in local media for his extensive use of social media. On the OP, no one's winning because of social media. Though some candidates are better at it than others.

Cotton is at 102K...Pryor is only at 10K

I think the article is talking about social media more as a barometer as opposed to a means of winning.

[IceSpear]

A lot of these likes are probably from past races (Joe Miller in particular), so they can they really be compared?

Also, 20/33 isn't exactly a great correlation on the House side.