This article is from the archive of our partner 
.
The folks over at Northwestern's Knight Lab have a handy tool to occupy you until polls close tomorrow night: an app that predicts who your favorite people on Twitter are voting for. There is one caveat however: the machine can't figure out the Andrew Sullivans or Dave Weigels of this world.
The actual app is called Tweetcast and more specifically, it uses an algorithm to predict whether a tweeter will sway Romney/Ryan or Obama/Biden tomorrow. "The system is most accurate — about 80 percent — with users who tweet about political issues. It’s less accurate — about 65 percent — with users who aren’t obviously political" its creator told the folks over at Northwestern's journalism school. And who you mention on Twitter and how you mention also factors in:
Specifically the technology tracks words, @-mentions, hashtags and links in the Twitter feeds of known political supporters and compares those to the same elements in the feeds of other potential voters. The greater the similarity between the two, the more likely they are to support the same candidate, O’Banion said.
80 percent is pretty good (Nate Silver would agree!), but apparently Dave Weigel and Andrew Sullivan are no match for Tweetcast. Here's what happened when we put Weigel in, we got this:
Fine, we know that he calls himself a libertarian and whatnot. But ... but .... look what happens when you put Andrew "talk me off the ledge" Sullivan into the machine:
Danger Will Robinson. Danger. But you know, who knows, maybe the app is actually telling us the truth and Weigel/Sullivan will rip off their masks tomorrow and tell us they voted Romney/Ryan?
