« A grim, familiar eight minutes | Main | Romney's sagging swing states »

November 05, 2012

Comments

And empirical support for PM:

Princeton Election Exchange shows Ohio is a 98% confirmed for Obama as of noon today:

http://election.princeton.edu/electoral-college-map/

Why do I get the feeling that Obama could be up by ten or twenty points and the media in their desperation to hold onto the tight race narrative would still call it a statistical dead heat?

not sure if it is simply stupidity or deliberate obtuseness. the margins of errors of individual polls are not additive. They are measures of the imprecision of a poll based in part on the size of the number of people polled. When you aggregate multiple polls, you increase the pool of voters polled. Statistically, that means the margin of error of the aggregates is *less* than the margin of error of the individual polls.

In other words, polling 2000 people is more precise than polling 400 people. And that's what an aggregate of polls does.

This is the longer version of Nate Silver's famous tweet that 7 polls all showing Obama with a lead in Ohio is not complicated to interpret.

I don't find it at all odd that CNN should announce a tight race when they make their living covering exciting news. If it isn't exciting then pretend that it will be.

Toss up!!!!! Just keep saying it! People have to watch us past 8pm on Tuesday!

The comments to this entry are closed.

Recent Posts and Archives



  • to P.M. Carpenter's Commentary




  • to P.M. Carpenter's Commentary