Long before you were born,
people had to turn for election coverage to something called "the
mainstream media," which relied on "pundits" whose main
qualification was a unique combination of "ignorance" and
If you want a sense of those halcyon days, you had only to tune into Monday's Morning Edition on NPR. Announcer David Greene talked to analyst/ignorant blusterer Cokie Roberts, humbly asking her to help him make sense of the oh-so-confusing state of the presidential election. Roberts pointed to the recent Gallup poll putting Romney six points ahead, but noted that it was an outlier, and suggested that, in fact, the race is in a "dead heat"—as it has been, she said "from the beginning." She concluded from this that Monday night's debate would be very important, and then went on to speculate excitedly about the possibility of an electoral college/popular vote split, which, she said, may favor Obama.
Unfortunately, virtually all of that is either wrong or misleading. In fact, the race has not been in a "dead heat" from the beginning. On the contrary, Obama has held a small lead for basically the entirety of the campaign. There has been some up and down movement, certainly, but even after his strong first debate performance, Romney has remained slightly behind in the majority of polls. In Nate Silver's aggregate polling model, Obama continues to have about a 66-33 percent lead. Similarly, Sam Wang gives Obama a slight Electoral College lead. The Huffington Post poll model puts Romney very, very slightly ahead, but it also shows that Obama has been winning for most of the race. I trust experts like Silver and Wang more than HuffPo… but even if Roberts trusts HuffPo, there's simply no excuse for saying that the race has always been tied. It hasn't, and a five-minute Internet search can tell you it hasn't. What possible excuse, then, can Roberts have for saying it has?
Similarly, there's no real reason to think that Monday's debate will be especially important. Third debates rarely move the polls much, and Obama's solid win in the last debate didn't cause much shift. Certainly, the debate could matter if one or the other participants says something especially stupid, and/or exposes themselves on camera, or if something else odd happens.
As for the Electoral College/popular vote split, that's something that pundits love to contemplate, because it makes election night exciting and promises conflict and controversy. Whether it'll actually happen is another story. Silver has the chances quite low, though Gallup's President Frank Newport seems to think it's likely. I think Silver is more reliable than Newport, but either way, it’s better to point out that this is a possible scenario, rather than presenting it as something to worry about.