Colin Powell endorses Barack Obama

[Darius871]

Originally posted by KrazyHorse
Given Obama's structural advantage in Electoral College McCain would probably have to win by popular vote by ~1% at this point.

*cough* *cough*

"Those who forget history are condemned to repeat it." - George Santayana

CNN/USA Today/Gallup tracking poll: Gore 51%, Bush 40%

(October 5, 2000)

WASHINGTON (CNN) -- Thursday's CNN/USA Today/Gallup tracking poll indicates that Vice President Al Gore may be opening a solid lead over Texas Gov. George W. Bush, after nearly two weeks of neck-and-neck competition. Today's figures -- 51 percent for Gore to 40 percent for Bush -- represents a significant margin for the vice president.

Originally posted by Darius871
C) the Bradley effect (or for that matter non-racial response biases driven by Obama's popularity) shouldn't be totally ignored, D) the traditional trend of undecideds predominantly leaning toward experience and/or conservative policy out of sheer FUD, and E) I don't trust the methodology of pollsters as far as I can throw it anyway, given that their non-response and coverage inaccuracies made Bush's win in 2000 somewhat surprising. Whatever it is that they're doing wrong (and I mean by objectively bad science, not any kind of bias), it apparently tends to inflate the Democrat's numbers to an extent.

Unfortunately it ain't over until the fat lady sings. Sorry to be the buzzkill but there it is.

[KrazyHorse]

a) Way to choose a single outlier poll from 2000. From what I remember, other polls from around then were much closer
b) Gore ended up winning popular vote, but losing narrowly due to Bush's structural advantage
c) Of course the final popular vote will be closer than currently. That is a well-understood phenomenon

[KrazyHorse]

d) All evidence is that the Bradley effect (if it ever existed) is long gone.

[Ramo]

The lead that Bush had in 2000 (measured by an average of polls) is much smaller than the lead that Obama currently has.

huge swing states like OH & FL are both at 46-48 requiring only a 66% lean for McCain among undecideds (notwithstanding larger spreads in smaller swing states like VA, CO, NM, etc.),

In that qualifier lies the election. Obama could lose several points uniformly in each state, and still win. The VA/CO firewall (NM and IA are off the board) would be pretty hard for McCain to surpass at this point.

[Imran Siddiqui]

McCain has something like a 7% chance to win, or something.

Regardless, has there ever been an endorsement in the general election that really mattered? Eisenhower's lack of endorsement of Nixon did, but that's a bit different than this.

Does anyone really think, say, Zell Miller's endorsement of Bush mattered?

So why exactly is Powell's endorsement of anyone going to matter? Its like treating this like some magical endorsement that actually may do something when others have not.

[Darius871]

Originally posted by KrazyHorse
a) Way to choose a single outlier poll from 2000. From what I remember, other polls from around then were much closer
b) Gore ended up winning popular vote, but losing narrowly due to Bush's structural advantage
c) Of course the final popular vote will be closer than currently. That is a well-understood phenomenon
d) All evidence is that the Bradley effect (if it ever existed) is long gone.

Originally posted by Ramo
The lead that Bush had in 2000 (measured by an average of polls) is much smaller than the lead that Obama currently has.

In that qualifier lies the election. Obama could lose several points uniformly in each state, and still win. The VA/CO firewall (NM and IA are off the board) would be pretty hard for McCain to surpass at this point.

Look, I already said earlier that I think McCain will lose, so I won't spend any time arguing against that, even for fun (that's what Ben Kenobi is for). Hell, even if proof came out that Obama is a post-op transsexual he/she'd probably still win. It's just that I distrust polling methodology enough to not buy any champagne just yet.

[Darius871]

And aside from my comment that overconfidence makes Obama look bad in the eyes of voters, keep another thing in mind: there will be at least some mild Obama supporters who might not bother going to the polls because of the impression that his lead is so big he'll win regardless. The bigger the spread, the bigger the "free-rider effect" gets. I still don't think that, even compounded by other factors, will be enough, but it'll sure as hell be close enough not to hang your hat on a paltry 6% in mid-October.

[Kidlicious]

Obama only has wide leads in the blue blue states. The race is still close in the swing states.

[Arrian]

Valid point. Do not get complacent. I doubt the Obama team will, though.

-Arrian

[KrazyHorse]

Originally posted by Kidicious
Obama only has wide leads in the blue blue states. The race is still close in the swing states.

Any reasonable McCain victory requires that he win FL, OH, IN, MO and NC. That is a given. Those five states together bring him to 247. He needs 23 more EC votes from among VA, CO, NV, PA and NH (a tie is an Obama win unless something insane happens in Congressional vote). These are the only states polling within 10 points currently.

EC votes, current Obama lead:

VA 13 Obama+6
NV 5 Obama+4
CO 9 Obama+6
NH 4 Obama+6
PA 21 Obama+9

Let's get real and take PA off the table.

Now McCain needs to win NV, CO and VA in order to squeak out a 270-268 victory. While also winning every other state that's within 6 points right now.

[Darius871]

Originally posted by KrazyHorse


Any reasonable McCain victory requires that he win FL, OH, IN, MO and NC. That is a given. Those five states together bring him to 247. He needs 23 more EC votes from among VA, CO, NV, PA and NH (a tie is an Obama win unless something insane happens in Congressional vote). These are the only states polling within 10 points currently.

EC votes, current Obama lead:

VA 13 Obama+6
NV 5 Obama+4
CO 9 Obama+6
NH 4 Obama+6
PA 21 Obama+9

Let's get real and take PA off the table.

Now McCain needs to win NV, CO and VA in order to squeak out a 270-268 victory. While also winning every other state that's within 6 points right now.

Wow, that's actually an even rosier picture than I thought. Eeking a slim victory in 8 of the 9 states within 6% doesn't seem that far-fetched to me given the slide that takes place in the last week of all elections. Still no more than 25% odds though, considering any October Surprise would have already happened by now.

[KrazyHorse]

25%?



I'll take 3-1 odds with you any day of the week on that.

And it's not just 8 of the 9 states within 6%. It's a specific 8 of 9. NH doesn't matter.

[KrazyHorse]

I'd love to see what the historic odds of winning every single state that you're behind by 6 points or less with 2 weeks to go are.

[KrazyHorse]

What it works out to is that McCain needs a 7+ point swing nationally in 2 weeks. While being outspent 3-1 or so. With no debates left.

[Ramo]

Yeah, 25% is way too high. Nate Silver's site, which factors in expected tightening of the race, puts the odds of McCain's victory at 7.5%. McCain's problem is that, if anything, the pollsters are probably too conservative in factoring in Obama's turnout operation.