This was a little ditty I wrote for Crikey earlier today.
With five days remaining until the US Election is finally over, the polls paint the picture of a large Republican defeat.
Starting with the national tracking poll data using the brilliant Pollster.com loess regression trend lines, the headline numbers show an ever so slight narrowing in the headline horse race over the last week.
Yet that Obama lead of 5.4% (well it was 5.4% when I wrote the article this morning!) doesn’t even begin to tell the full story of just how far McCain is behind.
In 2004, Bush won 286 Electoral College votes to Kerry’s 252. Currently Obama has the lead in 10 States that Bush won making up a combined 115 Electoral College Votes, pushing Obama out to a projected 367 Electoral College vote win. McCain on the other hand is ahead in exactly zero States that Kerry took in 2004.
Using the Pollster.com State polling analysis to look at the current polling results for those 10 ex-Bush States and putting their respective Electoral College votes in brackets we have:
Iowa(7) 53/41, Virginia(13) 51/43, Ohio(20) 50/43, New Mexico(5) 51/44, Nevada(5) 50/44, Colorado(9) 51/45, North Dakota(3) 45/41, North Carolina(15) 49/46, Florida(27) 48/45 and Missouri(11) 48/47.
Those links are worth hitting, taking you through to the spiffy Pollster.com charts for each State – the visual impact of some of the movements like North Dakota are pretty stunning
With the polls currently having McCain on 171 Electoral College votes and requiring 270 to win, he needs to pick up 99 EVs from the 115 of those ex-Bush States between now and Tuesday. His only other partial alternative is to win Pennsylvania, a 2004 Kerry State with its 21 Electoral College votes – yet Obama has 10 point lead there of 52/42, making it a more dubious proposition than every one of those ex-Bush States with the exception of Iowa, which has gone completely feral on the Republicans.
To make matters worse, if staring down the barrel of a 200 Electoral College Vote drubbing can possibly get any worse, not only does McCain need to come from behind in 100 EVs worth of States we’re he’s trailing by up to 12 points, but he also has to defend Indiana (11 EVs) where he’s currently ahead by half a point but where the trend is running against him, as well as hold on to Georgia (15 EVs) where McCain’s 17 point margin has been slashed to just 2 points over the last 6 weeks.
With all this polling misery for the Republicans, and the betting markets being in even worse shape, the pro-Republican pundits have been having a bit of a tough time trying to keep up morale. Rather than focus on how McCain leads on increasingly obscure polling metrics like better transportation manager, as our own Oz pundits were doing here last year, they’ve taken a different route where there is a special “Effect” at play for every occasion. There’s the dubious Bradley Effect – where white voters tell pollsters they’ll vote for the black guy, but then vote for the white because the shield of the ballot booth protects them from the crazy coercive powers of political correctness. Although strangely, the size of the Bradley Effect always seems to be the same size as Obama’s latest polling margin.
We have the Undecided Voter Effect (where undecideds break to McCain in ever increasing numbers, for increasingly illogical reasons), the Hidden Republican Voter Effect, the Hillary Effect and my personal favourite – Pollster Bias by Office Postcode Effect…. Ah, the pure moonbattery of it all.
Thankfully- this time next week, it will all be over.
11 Comments
“Thankfully- this time next week, it will all be over”.
Then we pick ourselves up, dust ourselves down and start all over again.
Spoken like a true tragic!
read a bit about zogby on 538 – are you a zogby disbeliever as well possum?
This Maths bloke gives The Kid a fair to middling chance of getting the nod on E-Day, but as Ferny sez:
“Hillary would have done better!”
http://politicalwire.com/archives/2008/10/31/simulation_shows_obama_will_win.html
Possum wrote:
President Obama … President Obama … President Obama … I know after I’ve said it and heard it a few times I’ll get used to it.
About 18 months ago on another discussion board, an earnest contributor predicted the rise of Obama to the White House. I more or less laughed in their face, convinced a black man would never even get the nomination, let alone be voted in.
I still feel bad for mocking that person, wouldn’t mind being able to find them again and tell ‘ em they were right all along. Alas, I no longer remember their name, and the board has ceased to exist anyway.
Surprising the things life throws up for us to think about …
I’m not a great fan of Zogby Grog.
It’s hard to imagine McCain will even come close on the polling figures we’re seeing but its interesting wondering whether media speculation on how ‘close’ the election is would be helping one candidate or the other. You’d imagine it would scare Obama supporters enough to get them to the vote, but similarly if McCain voters thought the election was definitely lost you’d imagine they’d be less of an incentive for them to vote.
Small mistake Possum- the Nevada link (above) goes to Pollster’s Kansas graph
Well, back to my old name for the moment .. bit happier now tho
Here’s an interesting article. The amusing bit is, the author thinks he is being critical of Obama, by suggesting that Obama might try to change the US fundamentally. Both economically and culturally. Pushing it more towards Europe (which to me sounds like the US might finally become civilised.) Here’s a choice quote.
“As noted in a 2006 World Bank report, spending in Europe on social-protection programs averages 19% of GDP (85% of it on social insurance programs), compared to 9% of GDP in the U.S. The Obama proposals send the U.S. inexorably and permanently toward European levels of social protection.”
Oh no! (repressed giggle)
http://www.realclearpolitics.com/articles/2008/10/the_true_meaning_of_historic_v.html
Howard’s gone. Australia is heading back to civilisation. Now its time to say bye not just to Bush, but to the people and ideas he represented. Yay!
Oh btw Possum, despite all the modeling, the US election is something I’m not betting on. Polls are one thing. Turnout is another. Had we a model for human behaviour.. For what it’s worth I’m going for 350+ but if there are any big surprises they will be on the + side. Dunno whether you agree with me, but to my mind the Australian election was about generational change. And so too the US.
-moo
Ok, a question for the Poss.
Assuming a uniform population (not a great assumption I know) and a given margin of victory (eg a 10% victory 55/45):
How do we calculate the percentage of the vote counted allows us to predict, with a given level of confidence (eg 90%), that the final result will be the same as the provisional result?