"Trump's election to lose"

[ursulahx]

This is the first thing I've seen which has really troubled me. Those of you who are Political Wire members will have seen the alert about the findings by Predata, who monitor online activity relating to political campaigns and who called the Brexit result against the grain of predictions (Predata claim they don't do predictions, but they're crowing about calling Brexit right. Hmm...).

This is Predata's article: https://medium.com/@predata/the-predata-us-election-special-september-16-2016-in-the-digital-realm-its-starting-to-look-ef3d1a207a4d#.vpt8jdvbu

Can anyone give me rational, convincing reasons why I shouldn't take this seriously? Coupled with John Judis's plausible criticism of Clinton's lack of a positive message (http://talkingpointsmemo.com/edblog/--100769), I'm beginning to think she's throwing away what should have been a slam dunk.

[ursulahx]

There's nothing "new age" about this kind of data analysis, it's already proved surprisingly perceptive in gauging under-the-radar opinion.

I realise you're trolling, but I don't see that Clinton is one state away from losing - that's one of the few things that gives me reassurance. Two states, perhaps.

[ursulahx]

Sorry, but couple this analysis - whether you think it's legitimate or not (I messaged Sam Wang about it, and he seemed unconvinced) - with the polling results showing much greater enthusiasm among Trump supporters (46% Trump vs only 33% Clinton in WaPo/ABC News), and add to it the fact that Clinton's campaign doesn't seem to have a theme, and I'm wondering if the firewall really can hold.

After being so bullish just a few days ago, I'm now actually worried. Only the reassurances of the likes of Mitch Stewart and David Plouffe are reminding me that Clinton's got this covered more than she appears to.