Shared publicly  - 
Can Nate Silver and other techie/stat-minded aggregators nail their presidential predictions?
Kingsley Idehen's profile photoShane Needham's profile photoD E G Gray's profile photoEmber Talent's profile photo
You might as well throw a dart at the board. Predictions and data models are as reliable as surveys and statistics.
Yeah, they can. This is not an art at this point - it is a science. This is all about numbers. In the final days of a campaign there is no new message or event that will change things. At that point it's just up to the statistics and in that respect Silver can give a pretty accurate assessment of the chances of each candidate. He can't say for certain, but he can give the chances and I think he's about right.
Democrats say "Silver = Good", Republicans say "Silver = Baaaaad". From what I've read, he seems the most likely to be right. 
Wait, what? You're going to measure against actual results? What are you, a bunch of liberal hippies? The race is a toss-up because it feels that way. That's all that matters! 
Any model can be wrong in many ways. Whether or not this particular model delivers the correct projection, poll-tracking statistical models like 538 have a bright future.

If I were Romney I wouldn't be ordering drapes.
Let's put it this way - we'll have a good idea in a few more hours if his projections were correct. I hope they are.
Add a comment...