Free Republic

I Wouldn't Touch It With a 10 Foot Poll Same As It Ever Was A lot has been made about the division between Red and Blue states. John Edwards ran his primary campaign around the concept of "two Americas". At times it seems as if Americans are locked into some sort of dysfunctional marriage badly in need of a John Gray book- "Republicans are from Mars, Democrats are from Venus". From Michael Moore and Al Franken to Ann Coulter it certainly can seem as if partisanship is at an all-time high. But it is not. Partisanship was with our...

<p>Apples to apples, funk to funky, you know that we're all poll junkies...</p> <p>My recent articles have touched on the hunt for the likely voter, how elusive a hunt it can be, and how different companies use different methods to try to predict who they are. A major challenge in trying to weigh several polling results across different polling firms is that the methods can be so different that differences in the results can be not merely due to the random fluctuations that occur invariably a part of sampling, and not merely due to any underlying movement in voter sentiment, but also due to the different techniques used in accumulating and presenting the data. While I have confidence in my ability to weigh these factors, it is still a worthwhile exercise to occasionally compare apples only to apples, and look only at multiple results by the same outfit.</p>

I Wouldn't Touch It With a 10 Foot Poll Likely Story Are you a likely voter? Before you answer that question, I want you to notice that I did not ask if you are likely to vote. These are two very different questions. How likely you are to vote is something you can answer on your own, and only you know how accurate an answer it is. But are you a likely voter? To answer this, you would have to know how the designation is defined. And that depends on who is doing the asking. Different polling companies use...

<p>I have a weakness for Jimmy Stewart movies. Sit me down in front of "The Man Who Shot Liberty Valence" and I am a very happy camper. Every Christmas, I am the sentimental sap who does not run screaming at TBS's eight millionth airing of "It's a Wonderful Life". And as a budding conservative, while growing up I loved the idealism and triumph of old-fashioned patriotism and virtue over abusive government and corruption in "Mr. Smith Goes To Washington".</p>

<p>Two months ago, I kicked off the Electoral College Breakdown for 2004 when it had become clear that John F. Kerry was going to be the nominee of the Democrats. Nearly $100M in advertising later, what has happened to the race? Most people, if you asked them, would say that Kerry had been winning the race until lately, when the President had made a comeback and has now surged into the lead; they would hold this opinion because that is pretty much what the national polls have been showing.</p>

<p>The best national poll for my money is the Battleground Poll. Produced by a joint effort between Democrat pollster Celinda Lake of Snell, Lake, Perry and Associates, and Republican pollster Ed Goeas of the Tarrance Group, it avoids the partisanship that sometimes can slip into the sampling methods of other polls. The partisanship can come out in the strategic analysis each does for the respective parties, although the spin presented is usually substantive. This year's springtime Battleground Poll, released this week, is excellent as always.</p>

<p>I conclude my review of the April 2000 Electoral College Breakdown by examining the tossup states. One of the most frequent criticisms I have received while doing the ECB is over my decision to not use the "Slight Advantage" states in calculating each candidate's total. Indeed, as a nod to the fact that people were going to take my numbers and add these states to the candidate having the slight advantage anyway, I have started to include the totals "with tossups", although I give these totals less prominence.</p>

<p>Are Things Leaning Towards Staying The Same?</p> <p>In retrospect, it is pretty obvious that Oregon and New Mexico would eventually become tight contests in 2000 despite April polls showing George W. Bush with double digit leads over Al Gore. In 1996, President Clinton defeated Bob Dole in Oregon 47.15% to 39.06%, with Ross Perot taking in 8.80%. And in New Mexico, Clinton had topped Dole by almost as comfortable of a margin, 49.18% to 41.86%, with Perot scoring 5.80%. What caught my eye, however, was how similar the vote percentage for the Democrat candidate was in each case.</p>

<p>The election cycle is in full gear. The candidates have been chosen. Unless the Democrats decide to bring some conspiracy theories to life, George W. Bush will defend his Presidency against Senator John Kerry. The polls are coming in, and they are painting a picture that meshes with the conventional wisdom- that 2004 will be 2000 redux.</p>

<p>The designations in the ECB2004 are based most significantly on the most recently available poll(s). In the past, I relied strictly on the polls, using my judgement only when there were two competing polls released during the same timeframe. In such cases, I would use my experience in watching opinion polls to give more credence to the poll run by the company or organization with the best reputation and the best methodology. However, while this removed almost all potential bias from my tracking, it allowed unfettered the bias of selectively released polls; if a poll was released by a group, and no other poll for that state was released in that timeframe, by default I would change the designation based upon that poll. Seeing how campaigns run many more polls than are released, and if one runs many polls eventually one will get a surprising result (an 'outlier', or a poll outside of the margin of error), the result was that campaigns would release the polls that would spin the race the way they want and instead of the ECB providing the most accurate picture, it could occasionally paint a distorted one.</p>

<p>This past week, "The Hill" ran an article by a Dr. David Hill titled "Is the election already over?" Rarely have I found a published piece of poll analysis that was so wrong on so many levels. That it was written by a director of a polling firm, and a Republican polling firm at that, left me scratching my head. Surely a professional could not be so falacious without intention.</p>

<p>Right now, most nationwide polls show a very tight race between President Bush and Senator Kerry. The media loves a horse race because it sells copies and drives ratings; the media is in full feeding frenzy mode right now over the sagging poll numbers for Bush and the soaring ratings for Kerry. There are two problems with this conventional wisdom. First, the movement of the polls that we are seeing now is nothing unusual. Second, winning electoral votes on the state level elects Presidents. A race that is close in either popular vote or in electoral vote may not be close in the other. Despite these caveats, there are some interesting numbers in recent polls that point to some of the problems facing President Bush.</p>

<p>This week was a pretty quiet one on the ECB front. The number of polls released this week was small, and most either reinforced previous results or represented very small changes. The one piece of significant movement from previous polling opens the door to the possibility of this election being a two front war, a possibility that demonstrates some vulnerability for the Bush/Cheney re-elect effort.</p>

<p>Since I began the rundown of the states, California had a new poll released.</p> <p>Please, no comments on the colors regarding who is red and who is blue. The map was made for me by SC Swamp Fox using a tool online, and they chose the colors for him. I'll eventually be doing my own map. Also, please note that although I call some states as having a slight advantage one way or another, it would be a mistake to count them for either candidate. They are well within the margin of error, and should be considered anyone's game.</p>

<p>Due to popular demand, I am going to revive my poll tracking and analysis that I did during the 2000 election season. I will be starting things a bit differently this year. Instead of starting off with a comprehensive overview of the entire nation, I am going to start by looking at five states at a time. For this tenth installment, the remaining states: Tennessee, New Hampshire, Wisconsin, Oklahoma, and the great state of Texas.</p>

<p>Due to popular demand, I am going to revive my poll tracking and analysis that I did during the 2000 election season. I will be starting things a bit differently this year. Instead of starting off with a comprehensive overview of the entire nation, I am going to start by looking at five states at a time. For this ninth installment, the random state generator presented me with Iowa, Arkansas, Minnesota, Louisiana, and Illinois.</p>

<p>Due to popular demand, I am going to revive my poll tracking and analysis that I did during the 2000 election season. I will be starting things a bit differently this year. Instead of starting off with a comprehensive overview of the entire nation, I am going to start by looking at five states at a time. For this seventh installment, the random state generator presented me with Hawaii, Oregon, Washington, Nebraska, and West Virginia.</p>

<p>Due to popular demand, I am going to revive my poll tracking and analysis that I did during the 2000 election season. I will be starting things a bit differently this year. Instead of starting off with a comprehensive overview of the entire nation, I am going to start by looking at five states at a time. For this sixth installment, the random state generator presented me with Kansas, Ohio, Pennsylvania, Maryland, and Mississippi.</p>

<p>Due to popular demand, I am going to revive my poll tracking and analysis that I did during the 2000 election season. I will be starting things a bit differently this year. Instead of starting off with a comprehensive overview of the entire nation, I am going to start by looking at five states at a time. For this fourth installment, the random state generator presented me with Wyoming, Delaware, Utah, Virginia, and New York.</p>

<p>Due to popular demand, I am going to revive my poll tracking and analysis that I did during the 2000 election season. I will be starting things a bit differently this year. Instead of starting off with a comprehensive overview of the entire nation, I am going to start by looking at five states at a time. For this eighth installment, the random state generator presented me with Indiana, Arizona, South Dakota, New Jersey, and Connecticut.</p>