Take a look at the two Pollster graphs below from a few weeks ago, showing Bush approval/disapproval ratings and support for Obama and McCain, respectively. While the numbers have moved slightly in recent days, they are still in the same ballpark for purposes of this analysis.
(click to enlarge)
Note that since the totals don't add up to 100%, there are three possible choices for each graph. Regarding Bush, there is “Approval,” “Disapproval,” and “No Opinion,” which must add up to 100 percent. Regarding presidential candidate support, there is “McCain,” Obama” and “Undecided,” which also must add up to 100 percent (although “Undecided” also includes supporters of minor party candidates who aren’t shown in the graph)
Every voter will fall into one of nine possible combinations, based on the three choices in each chart, so the combined percentage of the nine combinations will also be 100 percent.
For the sake of this analysis, I’ll make a couple of assumptions regarding the voters who fall into these categories:
1. Voters who still approve of Bush after 7 ½ years of Republican mismanagement are supporting McCain. While there may be a handful of Obama supporters or, more likely, those who are undecided, I’m going to assume the actual number is small enough to round to zero.
2. Of the six percent who have no opinion regarding Bush, I’m going to assume that two thirds of them are people who have no opinion on most things, and consequently overlap with those who are undecided on the presidential race. The remaining third will be split evenly between supporting McCain and Obama.
At this point, like some sort of twisted political version of Sudoku, all of the other categories can be completed, forming the yellow portion of the table below:
Now note that much of the current debate regarding what it will take to win the presidential election involves figuring out a way to appeal to the undecided voters, signified by the two bold numbers in the lower right which make up the 8% of so-called undecided, or swing, voters. However, also recall that some portion of these are current supporters of minor party candidates, or are people who tend not to make decisions at all, so their impact is even less than the original eight percent.
Of more interest, however, are the 17% of voters in the center left who disapprove of Bush, but are supporting McCain! I would love see a poll designed to develop a better sense of how it is that people fall into this category. I want to see some statistics on how these people would answer questions like the following.
In what areas do you expect John McCain to be an improvement over George W. Bush?
What policies do you expect John McCain to enact that would make him a better president for the country than George W. Bush, in areas such as:
• The Economy
• The Environment
• National Security
• Foreign Policy
• Civil and Individual Rights
• Social Issues.
Presumably, if these people are observant enough to have reached the conclusion that they disapprove of Bush, they must have some awareness of these issues, and consequently should be able to articulate some reasons for why they feel McCain is a more desirable choice to lead the country than George Bush. Or they will wake up and realize that McCain is more of the same!
Ultimately, I don’t think the currently undecided voters are the ones who are going to determine this election. Obama will win if he can find a way to get the 17 percent to answer some questions, at least to themselves, about why they prefer McCain to Bush.
Well, that and registering a shitload of new Democrats who aren’t currently reflected in the polls!