Slashdot Mirror

2012 is a long way away, and karma is a bitch.

Let's say the Democrats do manage to get Palin past the primary. Then Obama has a streak of bad luck, like W did.

All it would really take is a terrorist attack, or a market crash, or anything that people could latch on to and gripe about. Doesn't even have to be his fault.

If that happens and Obama's numbers were to slip over the next 2 years, that would mean Palin would have a decent shot at being President.

NO THANK YOU.

Instead I'd much rather Obama ran against someone who is vaguely competent, so in the event that he loses the country doesn't wind up in Caribou Barbie's hands.

Besides - the process is to (hopefully) put the best people forward so the country has the best leadership available. Let's not wreck that. Let the Republicans pick whomever they wish, then we will try to win on our own merits.

Remember - karma is a bitch. The only way to win is to be the Good Guys In The White Hats(tm). Don't tempt fate.

Rasmussen doesn't necessarily have a bias, but just a different methodology how it's doing its polling. It polls likely voters instead of all adults, which is an enormous advantage to conservatives since they tend to be the more politically active population in the U.S.. While it doesn't show an accurate representation of the country's opinion, it likely can have a higher likelyhood of predicting an election then a pollster like Gallup. http://www.pollster.com/blogs/why_is_rasmussen_so_different.php

"It should be noted, that in this election cycle the popular vote currently makes it a statistical tie between McCain and Obama. Obama may be slightly ahead, but not enough to beat the margin of error. However, when you sit down with state based polls and an electoral calculator, it's a complete washout."

what

Weighted averages: Obama 51.5%, McCain 44.3%. Just how big do you think the margin of error is?

I'll admit I'm a bleeding heart liberal political junkie.

on TV I'll be mostly MSNBC & PBS along with Daily Show/Colbert Report, CNN, and Current ( Al Gore's news channel )

on the net:

try DIGG's US Elections 2008 tabs both popular and upcoming.

pollster.com - http://www.pollster.com/ - is a good aggregater of polls it's what the TV guys use

Hufington post updates/changes frequently and has lots of videos http://www.huffingtonpost.com/

a news tab on igoogle with lots of feeds is good to have in another of your firefox tabs ( BBC, newspapers, TV channel websites, tec. )

I like Salon http://www.salon.com/?source=refresh ( I have a long time premium account ) - it updates a lot, but not as fast as huffington

This is a major, important election, I'll be watching on the HD big screen with friends. laptop atop my lap. I'll probably redo my "favotires" settings for the night.

Texas right now is polling for McCain by only 7 points, which isn't "very Red" (Bush beat Gore and Kerry by 61:38% and 59:38% respectively, about 20 points).

Senator John Cornyn is beating his underfunded Democratic challenger by only 10 points, while Cornyn's approval rating is under 50% (usually a signal that an incumbent is going to be defeated on Election Day).

The Texas delegation to the House of Representatives is only 19/32 (59%) Republican.

The Texas state legislature is only 4 switched seats (2.7%) from a Democratic majority.

All those districts have been carved out (gerrymandered) specifically for Republicans to gain the most seats - more so than in any other state. Those delegations come at the end of a period of Republican domination of the government (despite similarly thin margins nationwide) that prioritized overrepresenting Republicans everywhere, especially in big Southern places like Texas (the Bush HQ).

With all those advantages, the best the Republicans could do in Texas was a little better than the 50%+1 required for majority rule. Which majorities in every chamber have decreased steadily from initial large Republican gains. Trending back down towards the longtime Democratic majorities that controlled Texas until Bush Sr was the first Republican elected in the state, in the 1960s. Now that the Bush Dynasty is quickly waning, and Bush Sr himself near the end of his life at the center of a vastly powerful political network, Texas isn't looking nearly as Red as it once was.

By 2010, if not 2008, Texas will probably be as Blue as the rest of the USA.

Moderation 0
    50% Troll
    50% Insightful

TrollMods are a little behind the small, but real and growing, definite majority of Americans with an opinion who already want to impeach Bush/Cheney.

18% is flat out wrong. I would suggest Political Arithmetik for your polling analysis. The writer aggregates the different approval polls into one measurement. According to his Presidential Approval Rating page[1] President Bush stands at 28.9% as of 1 July 2007.

In general, the blog and it's sister blog at Pollster.com are a great source for polls and statistical analysis.

[1] if the link doesn't work you can always go to the home page and click on the chart on the right-hand side.

Actually, Bush is in the running for the...

(wait for the meme!)

Worst President Ever!

No seriously. According to this, his approval rating trend is at 28.9%. Typically, you've got a +/- 5 point margin on these polls, if he pulls an especially low poll, it puts him damn close to Truman's record for the all time lowest approval rating ever achieved by a U.S. president of 22%. It'll become more likely for this to happen if he sinks any lower.

Exit polls within the US have had a shady history at best. The meme that it's only been the last two Presidential elections where discrepancies have appeared is a myth created by conspiracy nuts or people just looking to sell a few more papers. About the only thing at all consistent is the level of inaccuracy. For example Clinton's overestimation in 1992 exit polls was almost exactly the same as Kerry's in '04 but Clinton won so no one really cared that the raw polling data was off by almost 3%. There have also been significant problems with 1990, 1994 and 1998 numbers with regards to Senate and Gubernatorial races but once again, no one really cared at the time so those numbers, which were off 58% of the time (in both directions), were basically thrown down the memory hole.

Historically US exit polling results are all over the map and have only been getting worse. It also doesn't help that poor training leads to improper sampling or that laws in certain areas restrict pollers access to voters.

Outside of the US exit polls are much more accurate but that can be easily explained by differences in polling techniques and voter mentality.

A good source for a little less bias polling info is Pollster.com as opposed to a book co-written by an editor of an progressive magazine run by an admitted socialist.

It's a bad thing because it means that merely relying on a party's base to come out and vote isn't enough. In the most recent election, the Republicans had a higher number of their people turn out than did the Democrats.

However, the large independent vote is what turned the tide. Independent voters turned out in higher numbers than in previous midterm elections and voted overwhelmingly for Democrats.

So yes, high voter turnout is a bad. But only if you're one of the two major parties because it screws with your polling numbers. Witness Karl Rove's pronouncement that the entire media world was wrong when it was saying there would be a Democratic takeover of the House and Senate because he, Karl Rove, was looking at 68 polls a week and he had "THE" math to show that Republicans would hold onto both houses. Partial transcript

Incidentally, Florida's Senate race this year is between Democratic incumbent Ben Nelson and Florida Congresswoman Katherine Harris. Harris is of course the former Florida Secretary of State under the President's brother (Governor Jeb Bush), who knocked tens of thousands of African American names off the voter rolls in 2000.

Harris, who has been very vocal in her view that only Christians should be elected to higher office and that anybody who is not a Christian will "legislate sin", however is down by 25 points in the polls behind Nelson (according to an average of the last 5 polls in the race), and doesn't have a shot in hell of winning... Unless....

Another informative site that combines election-related blogs and maps of the various polls is pollster.com.

Enjoy

Disclaimer: I work for Polimetrix, Inc., which runs a poll called PollingPoint and sponsors pollster.com

Another informative site that combines election-related blogs and maps of the various polls is pollster.com.

Enjoy

Disclaimer: I work for Polimetrix, Inc., which runs a poll called PollingPoint and sponsors pollster.com

Another informative site that combines election-related blogs and maps of the various polls is pollster.com.

Enjoy

Disclaimer: I work for Polimetrix, Inc., which runs a poll called PollingPoint and sponsors pollster.com

They myth of the infallability of exit polls is just that, a myth.

In fact 4 out of the 11 federal elections that have used exit polling have shown signigigant discrepencies (1988, 1992, 200, 2004) as well as at least 2 primaries (and this was from a proponent of exit polls, Steve Freeman). So In the US their barely hitting the 50% mark for accuracy on a federal scale.

Exit polls themselves, while usefull indicators, are not perfect and can be greatly affected by all types of social/political/technical issues. Because of that it's almost impossible to directly compare exit poll results from one country to another as they all use their own techniques.

Check out Pollster.com for a thorough breakdown of the pros and cons of the polling system

There are serious questions about both Freeman's conclusions and his methodology used.

Most his argument is based upon the use of the early release unadjusted numbers and not the final release values, which were much closer. For those not acquainted with exit polling systems, the numbers are adjusted to coincide with the actual voter turnouts and remove disproportionate representation within the sample group. I.E. if your sample is made of 55% but actual voter turnout has a percentage of only 35% women than the polls are adjusted to compensate for the over representation.

He also tries to use mathematical comparisons where they simply don't work. For example, his comparison of response rates in Blue districts to Red Districts. According to his conclusion, if a GWB supporter has a 56% response rate in Texas they should also have a 56% response rate in L.A. As almost any Conservative living within a Liberal bastion can tell you, vocalizing your Conservative beliefs invokes about the same response as insulting someone's mother or worse, and in terms of vocalizing support for GWB, well you'd probably receive a more favourable response if you just hit someone.

And from his own book (thanks to the review at TPM Cafe):

"In addition to the 1988, 1992 and 2004 presidential elections and the 2000 vote for president in Florida, the only other significant unexplained US exit poll discrepancies on record are the Republican presidential primaries in New Hampshire in 1992 and Arizona in 1996."

As the reviewer points out, exit polling has only been used in the US since the 60's, so out of 11 or so federal elections, 4 have produced questionable results, that's hardly a great confidence builder in their accuracy. But even though Freeman clearly states this, he ignores any implications that may have and continues with his "exit polls are always accurate" argument.

The fact his co-author is the editor of a 'Progressive' magazine also does not help his image as a independant arbiter for truth much.

Overall, I'll stick with Mark Blumenthal's work when it comes to impartial polling analysis.