According to Gallup, among registered voters Kerry leads in both Pennsylvania and Ohio, one of which Bush must win. And yet the collective MSNBCNNFOX narrative is that it’s desparation time for Kerry because tonight’s debate is his “last chance” to get back in the race.
Go figure.
First of all, when one poll says one thing and another says another, it’s just another indication that all polls (especially of just 800 people) are meaningless and pointless.
That said,
a) First of all, there are two sets of criteria: likely voters, and registered voters. In registered voters, Kerry leads Bush in OH and PA about 49-46/45. Bush leads Kerry in FL 49-44. You’ve used this info to make your point that Kerry leads in OH and PA.
b) But why? Isn’t it logical that “likely voters” means people who will probably actually vote, as opposed to registered voters – some of which will stay home? If you go by their “likely voters” numbers, Bush leads in all three states: 52-43 (FL), 49-47 (OH), and 49-46 (PA)
So how do you (and USA Today, who is drawing the conclusion) reasonably come to the assumption that Kerry’s leading in OH and PA by arbitrarily choosing one set of people over the other, especially when the one they chose to feature is less likely to be indicative of the final outcome?
c) Where do you get the idea that Bush must win OH and PA to win the election? I’m not questioning you, necessarily, I just haven’t heard that prediction before and it wasn’t actually part of the article. Without actually citing a source for that claim, it sounds more like your own analysis – and without explanation it doesn’t make sense.
You’re saying if Kerry wins PA and OH in the end, there’s no way Bush could amass enough Electoral votes to win, regardless of how all the other 48 states vote? Judging by your emphasis on must win it sounds pretty darn critical, with little room for error. I’d just like to know why those are such must-wins.
I wasn’t suggesting that Bush had to win BOTH Ohio and Pennsylvania; I was saying that I don’t see how he gets 270 electoral votes if he doesn’t win at least one of them. Bush pretty well needs to win Ohio, and if Kerry wins both swing states it will be an indication he’s done well elsewhere.
As for the differences between “likely” vs. “registered” voters, I’m becoming a little suspicious of the “likely voter” estimates. We’ve been told there has been a high voter registration in Democratic regions, and I’m not sure the polling models are factoring this in.
Sorry, I misread you – I thought you said both. My bad.