Archive for November 2nd, 2008
To look at the Drudge Report, you’d think McCain has been steadily inching closer to Barack Obama, and is within striking distance of taking the popular vote lead and running the sweep of toss up states necessary to come from behind and win the election. Last week it was a “shock Gallup poll” which showed the two within two points using the ‘traditional model’ for likely voters. By Sunday it was a ten point race in that group again. But no matter, Rasmussen showed it narrow to three points, so that was cited — well within the margin of error! Alas, it expanded back to five points, and Rasmussen declares the race “remarkably stable” with Obama at about an 85% chance for victory.
Then it was the IBD/TIPP poll which has always showed a tighter race. And finally on early Saturday morning Drudge screamed out that “McCain leads in overnight polling!” Wow! He must be zooming back. For the Obama fans, this is their worst case scenario, another defeat snatched from the jaws of victory, an unexpected comeback. For the McCain faithful this plus slightly tightening polls in Pennsylvania and Ohio shows that their come back scenario is on track — they can do it!
Of course, let’s get real. The Zogby tracking poll surveys 1200 people over three days. Gallup and Rasmussen survey 3000. That means that the one day poll is of 400 people, a sample size that suggests a margin of error of 5%, rather than the tracking poll 3% (or 2% for Rasmussen and Gallup, who sample 1000 a night over three nights). Remember, on average one in twenty polls will be outside the margin of errror. Zogby could be detecting a trend, but unless it’s mirrored in other polls, it could also be Drudge and Zogby trying to drum up interest — why else leak one day’s numbers?
Back in 1996 near the election Zogby had a closer race than the others, and a late shock poll put Dole and Clinton even — but only for a day. Moreover, the three night average for Zogby has Obama with a five point lead. That means that the two nights before must have shown a significant Obama lead. Gallup had gone from 5 to 7 to a 10 point lead in the traditional model in the last three days, suggesting they are not observing the trend Zogby claims exists. Given they have 1000 interviews to Zogby’s 400, there is little reason to believe there is a national McCain surge.
Drudge used to be “king of the hill” for online news, but now Huffington Post gets double the traffic, and the blogosphere is leaning leftward, sort of a mirror image of how the right dominated talk radio in the 1990s. Perhaps these kinds of teasers are a rather desperate effort to get hits. I’ve also noticed that Drudge cherry picks only the McCain friendly polls. To be sure, one of the best political websites, Realclearpolitics.com has a conservative bias, but unlike Drudge, you get the sense they are about giving the conservative perspective than promoting propaganda.
In Pennsylvania there is more reason to believe the race is tightening. All polls show it going from double digit to single digit, and Pennsylvania politicians who know the state have said that they could not believe the 12 – 15% point leads polls had been giving them. It now looks like the polls are putting Obama’s lead at about 5, probably still enough to hold the traditionally Democratic state, but definitely tightening. Compared to the polls of 2004, it doesn’t seem particularly ominous.
In fact, looking at the early voting numbers, the widening of the polls in most cases, and the general tenor of the last days of the campaign, things are looking very, very good for Barack Obama. Tomorrow I’ll have my prediction on the outcome (electoral votes, popular vote, and analysis of why I make that call), and then late Monday evening I’ll post my state by state predictions, with a bit about each state on what to watch for, and where one might see signs of either an Obama landslide or a McCain come from behind victory. I’m going to post it in order of poll closing time, which may or may not be easier than alphabetical. And, I’ll blog more if anything else warrants it.
But time is running short, the election is almost here! I think the state of the election is clear. Obama and Biden are buying time in states like Arizona, Montana, Georgia and North Dakota, and traveling a wide variety of states to try to pick up as many states as they can. McCain and Palin are focusing on keeping the core they need: Virginia, North Carolina, Ohio, and Indiana, while they try to flip Pennsylvania and hope enough red states resist the Obama onslaught and go for McCain. (Do they really think they can win Pennsylvania, or do they just want to do whatever they can to keep the Democrats from expanding the map — goad them into making Pennsylvania a ‘battleground)?
This long, strange trip that started in 2006 was dominated in 2007 by the immigration issue, making McCain look defeated early. Hillary was seen as the preemptive nominee, earning Time magazine cover stories and a large mass of early money and superdelegates. Guiliani was the presumed Republican nominee. Then Obama and McCain came on, McCain cinched early, and Obama and Clinton became locked in an historic struggle — either the first woman or the first black nominee from one of the two major parties would emerge. The entire year has been exciting, hardly a week without some twist and turn. And we near the end of this historic, exciting, and intensely emotional but fascinating campaign.
OK, time to start working on my final analysis and predictions…
Update: Obama has a 10 point edge in one night of polling the day after McCain’s one point edge. Simply: there is no surge for McCain.