election predictions
A couple weeks ago, the day of the last debate, I made the prediction that Kerry would win the election on November 2nd by 4-6%. I am still standing on that prediction and, in fact, think it may be low. Here's why:
Polls are showing that Bush and Kerry are swapping lead back and forth, both within the margin of error. This shows that the polls are, on average, at a dead heat. If you take all of them, averaging out their responses, you will find that the lead is swaying back and forth between the two with a rather consistent "undecided" number around 7%, Nader getting around 1%, and (when included, which isn't often) the other candidates sharing another 1%. This bodes well for Kerry on several levels (some of which I've mentioned).
First, polling companies are being left behind by changes in technology. All of the ones that we read about rely exclusively on phone-based polling which itself relies on publically listed directories to pull numbers out to call. Currently, however, as many as 7.5 million Americans use wireless phones exclusively. That's 6% of Americans and 14% of young Americans who are left out of the polls -- a figure that is well outside the margin of error of most polls. Add on those that use internet-based phones, have unlinsted numbers, or no phone at all, and you have an even bigger problem. Add on top of that the 4 million expat Aericans that are voting in record numbers from overseas this year and the skew is even more profound.
This issue helps Kerry in that younger voters are more likely to embrace these new communication technologies, as are single people (I would find it hard for a house with children to operate solely wireless) and younger people tend to vote more Democratically (us kids are more liberal than conservative as has been the case throughout history). Foreigners also tend to be more liberal and also -- in this election especially given the circumstances -- surrounded by the teeming masses of un-Americans that hate Bush, seeing a different perspective that we can't get here thanks to our bullshit "news" agencies. In any case, all of this adds up to a huge chunk of Kerry's demographic being highly underrepresented in the polls...I would not be surprised to see this add up to at least a 1-2% "catch-up" to Bush on Kerry's part come election day.
Second, the polling companies are using outdated presumptions of who will turn out on November 2. The Gallup Poll, for instance, uses numbers that suggest that 40% of those that turn out will be Republican and only 33% Democrat. As John Zogby has pointed out himself, this is completely misrepresentative of recent history which suggests a pretty even split of Reps, Dems, and Independents over the last 12 years. This adds up to issues such as my recent finding of a poll that showed Kerry leading by 2 points among registered voters but losing by 7 points among likely voters (I can't remember where I saw it so I can't link...I want to though).
Not only does this fly in the face of history, but I think it overlooks more trends in American society. It is generally presumed (and rightfully so, unfortunately) that liberals are much more apathetic about voting. The Republican demographics of suburban families, the rich, and the evangelical always get out to vote while the Democratic demographics of the young, minorities, and the poor tend to have to psyched up to get out. I think that the polling companies are misjudging how psyched up the Democratic demos are. Fahrenheit 911 911, for instance, brought in $120 million at the box office this summer, five or six times as much as any documentary ever, driven much by a hatred of Bush that is pulsing through this country.
Furthermore, new registrations to vote in elections against an incumbant tend to mean either a general disdain for the occupant of the White House or in an effort to support one of those candidates that galvanizes the country and inspires people to vote when they otherwise wouldn't. I don't think that either of these guys is anywhere close to meeting the criteria to be in that second category, so I can't help but feel that the record numbers of new registrations (many places reporting a 10% increase over 2000) is a sign of people wanting to remove the incumbant and doing what they need to do in that effort. All of this could lead to an extra boost to Kerry on election day.
Third, and lastly, what could be called the Incumbant Rule also favors Kerry. As I mentioned, there are currently something like 7% of voters that remain "undecided" and even Frank Lunz is admitting that many of those will, in the end, head to Kerry as they are more undecided about the incumbant rather than undecided in general...that is, they are looking for a reason to re-elect, but aren't finding it, which tends to lead to a majority of them going to the challenger. As Polling Report shows in their pie chart, the majority of undecideds go to the challenger 86% of the time:
What Kerry can look at then is for around two-thirds of those that remain undecided to fall his way on election day. In other words, with the current 7% remaining undecided, Kerry may very well see a 4.5% increase in his numbers while Bush only sees a rise of 2%. Kerry pulling a 2% "catch-up" here too.
So all in all, I am going to maintain my prediction that Kerry will pull off a 4-6% win in the popular vote (that big a popular vote win undoubtedly leading to an electoral victory, even though Kerry seems to be generally leading there in the polls even without these facts). I would not be surprised either if that lead is even bigger than that.
Oh well, it's all a guessing game at this point anyway so I could very well be wrong. But I gotta put all those stupid poli sci classes I took in college to work sometime, right?
Polls are showing that Bush and Kerry are swapping lead back and forth, both within the margin of error. This shows that the polls are, on average, at a dead heat. If you take all of them, averaging out their responses, you will find that the lead is swaying back and forth between the two with a rather consistent "undecided" number around 7%, Nader getting around 1%, and (when included, which isn't often) the other candidates sharing another 1%. This bodes well for Kerry on several levels (some of which I've mentioned).
First, polling companies are being left behind by changes in technology. All of the ones that we read about rely exclusively on phone-based polling which itself relies on publically listed directories to pull numbers out to call. Currently, however, as many as 7.5 million Americans use wireless phones exclusively. That's 6% of Americans and 14% of young Americans who are left out of the polls -- a figure that is well outside the margin of error of most polls. Add on those that use internet-based phones, have unlinsted numbers, or no phone at all, and you have an even bigger problem. Add on top of that the 4 million expat Aericans that are voting in record numbers from overseas this year and the skew is even more profound.
This issue helps Kerry in that younger voters are more likely to embrace these new communication technologies, as are single people (I would find it hard for a house with children to operate solely wireless) and younger people tend to vote more Democratically (us kids are more liberal than conservative as has been the case throughout history). Foreigners also tend to be more liberal and also -- in this election especially given the circumstances -- surrounded by the teeming masses of un-Americans that hate Bush, seeing a different perspective that we can't get here thanks to our bullshit "news" agencies. In any case, all of this adds up to a huge chunk of Kerry's demographic being highly underrepresented in the polls...I would not be surprised to see this add up to at least a 1-2% "catch-up" to Bush on Kerry's part come election day.
Second, the polling companies are using outdated presumptions of who will turn out on November 2. The Gallup Poll, for instance, uses numbers that suggest that 40% of those that turn out will be Republican and only 33% Democrat. As John Zogby has pointed out himself, this is completely misrepresentative of recent history which suggests a pretty even split of Reps, Dems, and Independents over the last 12 years. This adds up to issues such as my recent finding of a poll that showed Kerry leading by 2 points among registered voters but losing by 7 points among likely voters (I can't remember where I saw it so I can't link...I want to though).
Not only does this fly in the face of history, but I think it overlooks more trends in American society. It is generally presumed (and rightfully so, unfortunately) that liberals are much more apathetic about voting. The Republican demographics of suburban families, the rich, and the evangelical always get out to vote while the Democratic demographics of the young, minorities, and the poor tend to have to psyched up to get out. I think that the polling companies are misjudging how psyched up the Democratic demos are. Fahrenheit 911 911, for instance, brought in $120 million at the box office this summer, five or six times as much as any documentary ever, driven much by a hatred of Bush that is pulsing through this country.
Furthermore, new registrations to vote in elections against an incumbant tend to mean either a general disdain for the occupant of the White House or in an effort to support one of those candidates that galvanizes the country and inspires people to vote when they otherwise wouldn't. I don't think that either of these guys is anywhere close to meeting the criteria to be in that second category, so I can't help but feel that the record numbers of new registrations (many places reporting a 10% increase over 2000) is a sign of people wanting to remove the incumbant and doing what they need to do in that effort. All of this could lead to an extra boost to Kerry on election day.
Third, and lastly, what could be called the Incumbant Rule also favors Kerry. As I mentioned, there are currently something like 7% of voters that remain "undecided" and even Frank Lunz is admitting that many of those will, in the end, head to Kerry as they are more undecided about the incumbant rather than undecided in general...that is, they are looking for a reason to re-elect, but aren't finding it, which tends to lead to a majority of them going to the challenger. As Polling Report shows in their pie chart, the majority of undecideds go to the challenger 86% of the time:
What Kerry can look at then is for around two-thirds of those that remain undecided to fall his way on election day. In other words, with the current 7% remaining undecided, Kerry may very well see a 4.5% increase in his numbers while Bush only sees a rise of 2%. Kerry pulling a 2% "catch-up" here too.
So all in all, I am going to maintain my prediction that Kerry will pull off a 4-6% win in the popular vote (that big a popular vote win undoubtedly leading to an electoral victory, even though Kerry seems to be generally leading there in the polls even without these facts). I would not be surprised either if that lead is even bigger than that.
Oh well, it's all a guessing game at this point anyway so I could very well be wrong. But I gotta put all those stupid poli sci classes I took in college to work sometime, right?
0 Comments:
Post a Comment
<< Home