Fear of Clowns

"Faith may be defined briefly as an illogical belief in the occurrence of the improbable."
- H. L. Mencken

Sunday, August 22, 2004

FOX News and I are feeling good about November 2 

A day doesn't go by without hearing that this presidential election is neck and neck. In terms of the popular vote as it shows up in preference polls, that's true. But when you look at electoral votes, Kerry is clearly in the lead. In order for things to start looking good for Bush, he's going to have to reverse the trend in several states: over the last two months, Florida, West Virginia, New Hampshire have slid into the Kerry column. If it doesn't make immediate sense how Kerry can clearly be leading with an advantage of only 2-4% nationally, consider how the electoral college works, if it does make sense, jump down this post a bit. Here's an illustrative example of a three state nation that elects their president more or less like we do, with the difference of voting being mandatory. Keep in mind that each state gets two "free" electoral votes regardless ofd population:

Splay that out to find how many people in each state prefer Bush or Kerry:

Add up the votes to see who is ahead in the popular vote nationally and who wins the electoral vote:

When you look at state level polls today, you see something very similar is happening. In the states where Bush is ahead, he's generally ahead by a larger margin than in states where Kerry is leading. Here are a few examples from large states to illustrate:

Looking at only the 132 electoral votes from those four states, we see that across those four states, Bush has 49% of the vote and Kerry has 46% - that's not how we determine the winner of a presidential election. If you get the most votes in a state, you get all the electoral votes: Kerry wins California and Florida for 81 electoral votes and Bush wins Texas and Georgia for 51 electoral votes.

If you do the same thing with every state, you find Kerry has 286 electoral votes and Bush has 233. An easier why of explaining what I just pointed out at length may have been simply to have noted that in 2000, Bush lost the popular vote, but won the electoral vote.

An additional peculiarity I'm anticipating on election day comes from the phenomena of many people desiring "anybody but Bush" to be in the White House. In a typical election, a lot of people are going to not particularly like either of the major candidates and they'll not vote for either, but just stay home from the polls. This election will likely be the same in that way. But also in this election, there are a whole bunch of people that virulently dislike Bush and will be motivated to vote solely to get Bush out of office. There were so many people that showed up to the Democratic caucus in my district and wanted to be delegates for the state convention, special "how to be a delegate" sessions were arranged. In my precinct, people who had been going to the caucuses since they began again during the Vietnam war said there were four times as many people as had showed up ever before. It seems clear to me that the people who are "anybody but Kerry" voters would be voting for Bush anyway.

Pollsters come up with the populations they survey by going off of lists of registered voters - lists which anybody can purchase from secretaries of state. From those lists, they usually poll "likely" voters - a category they come up with by looking at how often the subject has voted in recent elections and asking them a few qualifying questions at the beginning of the interview. If a person hasn't voted since 1992, most pollsters aren't going to consider them "likely", but not pollster uses the exact same criteria. You can see that different pollsters use different criteria by comparing the "likely" numbers to the "registered" numbers - all have Kerry doing better among "registered" voters:

Except for the FOX poll which shows Kerry doing better among "likely" voters:

FOX's poll is the only one that makes sense if you accept the premise that there are a lot of angry voters who will be drawn to the polls by an "anybody but Bush" sentiment (or kept away because they only want to vote for a Republican, but not Bush). FOX's presidential polling sticks out for another reason as well: the president rather consistently has a higher approval rating than in other polls taken around the same time - the president - not "Bush" - FOX had Clinton's ratings higher than the other pollsters as well.

To me, it seems reasonable to believe that the anger at Bush will result in high voter turnout among people that typically vote for Democrats, and that Bush's policies will have brought about low voter turnout among people that typically vote Republican: spending money like there's no tomorrow, expanding the welfare state, having a welcoming attitude towards those south of the border and allowing the size of the federal government to mushroom.

Pollsters don't make such evaluations - their determinations of who is likely to vote are based on questions like "How important you think the outcome of the upcoming presidential election is? Very important, somewhat important, not very important or not important at all?" A better question might be "Are you so angry at Bush that you'd never vote for him in a million years or are you so angry you'd even prefer the idiot is inside that Barney the Purple Dinosaur costume?"

Because new voters do not show up on the voter rolls the pollsters go off of, they are neither "likely" not "registered", there is no way of measuring their influence. So add a lot of new "anybody but Bush" votes to the fact that we already have an angry registered electorate, and Kerry could easily win in a few states which conventional wisdom have until recently considered "safe Republican", but are actually pretty close among the population the pollsters are querying: Missouri, Arkansas, Virginia, North Carolina. Maybe later I'll write about the fact that a few reasonably close states have no senatorial or gubernatorial elections will also work to Kerry's advantage - I have more work to do. Bush says he has a lot more work to do too. He does indeed.

Post a Comment



Post a Comment

This page is powered by Blogger. Isn't yours?
Listed on BlogShares