Do I fish for polls? Sometimes I catch myself doing it, or at least wanting to do it, so I probably do fish sometimes. Does Real Clear Politics fish for polls? It sure seems like it. They include partisan polls such as the recent Rocky Mountain News / Opinion Strategies in their battleground state average of Colorado, but leave out the last three non-partisan Zogby reports in every single state. They always use LV's, and frequently use three-way polls even when Nader isn't on the ballot in a state. They consider New Jersey a battleground state, but do not consider Arkansas or Virginia battlegrounds.
I am sure they have justifications for all of this, as I have justifications for what I do. This post is not about bashing Real Clear Politics, which often annoys the hell out of me but which I would take over CNN any day of the week when it comes to election analysis. This post is also not about re-opening debates over LV's vs. RV's, two-way vs. three-way, and telephone vs. Internet polls. What this post is about is demonstrating that even Real Clear Politics currently shows Kerry beating Bush in the Electoral College. Of course, not only won't they tell you that, they seem unwilling to tell themselves that either.
I have written fairly often about the Incumbent Rule, which will begin to take effect in a little over one week. The basic tenant of the incumbent rule, that incumbents below fifty are in trouble because undecideds break toward the challenger, is well known. There are, however, less well known aspects of the Incumbent Rule. One of these is that in Presidential races, the Incumbent Rule is actually even more of a "rule" than usual. In fact, while incumbents in all races from 1992-2002 managed 34% of undecideds, from 1976-1996, incumbent Presidents managed only 14% of undecideds.
The reasoning for this is actually quite simple. Most undecideds break toward the challenger because opinions about the incumbent are better formed. If someone is undecided about a candidate s/he has known for years, in reality that voter has probably made up her or his mind. The hypothetical undecided voter has decided that s/he does not like the incumbent, and tends to vote in a way that expresses that opinion. When it comes to incumbent Presidents, this is amplified because the President is the most visible elected official in the country. If, after nearly four years, someone still says they are not sure if he or she will vote for the sitting President, s/he has, in reality, almost certainly decided to vote against the sitting President.
An incumbent President, if he is extremely lucky and doubles the historical average, will mange 30% of the undecided vote in a given state. In their battleground polls, Real Clear Politics only has Nader above 2% in Maine, the second most third-party happy state in the nation. In their battleground polls, Real Clear Politics also never has the total number of undecideds surpassing 7.0% (Minnesota is right at 7.0%). With 30% of the undecided vote going to the incumbent, 7% of the electorate undecided, and 2% of the electorate going third-party, an incumbent--Bush--needs to be at 47.0% in order to have any chance at all to win the state. That is, an incumbent at 47.0% can still hope for a perfect storm, but even then is an extreme longshot.
With this in mind, I see that RCP lists Iowa, Maine CD-2, Minnesota, New Mexico, Pennsylvania and Wisconsin as toss ups. This just doesn't pass the smell test. Here are their own averages of where Bush stands in these and other "battleground" states:
Incumbent Dead Zone = < 47.0
IA: 46.8
NM: 46.5
WI: 46.3
PA: 46.0
OR: 45.0
ME: 44.5 (translates to 45.6 2nd CD)
MN: 44.3
WA: 44.0
NH: 43.8
MI: 43.7
NJ: 43.0
Gee wiz, that's not too good for Bush, is it? Even Real Clear Politics, which is really fishing for pro-Bush polls, doesn't give Bush any real chance to win any Gore states. Further, they also give Bush no chance to win New Hampshire as well. Of course, they do not seem to realize this, as they still list Minnesota, where they have Bush at 44.5 and dead as a doornail, as a toss-up. In reality, because of the Incumbent Rule, they give Kerry 264 without breaking a sweat.
With Kerry at 264, he would only need six more electoral votes to win. Where could he possibly get those from? Here are some more RCP averages:
FL: 48.7
WV: 48.0
OH: 47.2
With 86% of the undecideds, Kerry would win all three of these states according the current RCP average. However, Ohio really sticks out. In a state without Nader (pushing the incumbent dead zone to 47.5), Bush is at 47.2, with only Faux News keeping Bush above 47.0. Further, they have Kerry at 47.6, already ahead of Bush even before the undecided break. Bush would need to win 60% of the undecideds in order to take Ohio, more than quadrupling the standard undecided break for an incumbent President.
Real Clear Politics has gone fishing, but they still haven't caught polls that put Bush in the driver's seat. According to their own battleground state poll averages, Kerry will win 316-222 unless Bush completely busts the Incumbent Rule. Incidentally, right now I also forecast a 316-222 victory for Kerry, largely because I allocate 80% of undecideds to Kerry based on the Incumbent Rule. They can pretend September and October Zogby polls do not exist, post Strategic Vision polls on their front page, and imagine that the latest Fox national poll actually means something, but even their own averages show Bush in serious trouble and with little time to turn things around.
|
|
|
Permalink :: 44 Comments :: Post a Comment
|
In order to post a comment, you must be logged in. If you have a member account, please log in to comment.
If not, you can make an account right here. It's quick and free.
