Polls, pollwatching and OPD

OK it its October 2008 and I am obsessed with the polls, I’ll check 4 or 5 web pages a day to see the latest polls and poll analysis. I know its not healthy, and polls are the opposite of politics and when presented as news they are really a way of avoiding real news. I know this but still I check the polls again and again. I’ve even come up with a psychological diagnosis, OPD obsessive poll-watching disorder. Obama is ahead by six to ten points nationally and according to some maps, he could reach into the mid 300’s with electoral college votes (270 are needed to win), but is it enough.

There are concerns about the Bradly effect, that people whose voting is hampered by racism will be less likely to admit it. The Bradly effect is actually a bit more complex. First the difference between polling and performance in the election seems to largely involve the “undecided” category. With Obama polling above 50% he should have some insulation against this. Secondly the Bradly effect seems to have been more prominent in the 80’s and 90’s than since 2000. There are various explanations for this but it is what the data seems to show. Nonetheless the N, the number of samples in the study is small. Finally we might take solos in the possibility of a reverse Bradly effect, in the Primaries in certain southern states Obama did better than what the polls showed. Again several possible explanations, but that’s what happened. Then there is the much talked about cell phone effect, individuals with cell phones and no land lines are more likely to fit the Obama supporting demographic, but this was supposed to play out in 04 but didn’t appear to

In 2004 I started watching polls more closely, there was a web site where I would check in daily, and I watched with optimism as the indicators looked like Kerry would win. On election day he held a small lead on Bush in Ohio, but many Clevlanders were disenfranchised by polling places with 4 hr lines, and then there was the Republican district where the number of Bush votes exceeded the number of registered voters. In 2000 Bush was appointed by the supreme court, Florida was marred by voter irregularities and again in 2004 Bush stole the election. Steeling an election is most possible when the election is close.

So between the Bradley effect and the possibility of Republican Party chicanery the question is how much of a lead in the polls will translate to a win on November 4th. There is of course a fine line between hope and confidence when it comes to getting out the vote. This uncertainty of polling numbers may help the Obama team keep on its game.

Finally there is the third party issue. Although I have been working on Obama’s Campaign, I am enthused about the Obama Movement (see my first post), and I hope Barack Obama is our next president, I see the limitations of Obama's politics. I am not a Democrat, the political party I affiliate with is the Green Party. If I were completely certain that Obama would win by a “landslide” I would consider voting for the Green presidential candidate Cynthia McKinney. There is a reasonable argument that a clear and decisive outcome will help Obama promote a progressive agenda. So at what point would in the polls would I vote McKinney? I don’t think that Obama is close to safe unless he is polling greater than 50% and more than 10 points above McCain in any given state. But I would need at least a 20 to 30% spread and at least 55% polling for Obama before I would feel comfortable voting for McKinney. At this point Hawaii is looking like safe territory for voting for McKinney.