User Panel
Posted: 9/27/2004 5:38:04 AM EDT
|
|
Goddamnit
Too many states in play... when Ohio goes, we go. - BG |
|
On the bright side, this map shows that President Bush can win without Florida.
|
|
Although I don't like the pink and outlined pink states. Has anyone noticed that Kali and Illinois is light blue now? When and how did this happen?
|
|
I've already got my absentee ballot requested. My vote will be cast, by God!
|
|
I sure hope they are right, because that final is a beautiful site |
|
|
Unfortunately, PA will go Kerry. The big cities just carry too much weight for us in the rural area's to overcome.
|
|
Bzz..Wrong PA wil go Bush. SGatr15 |
|
|
Mrs.G... the beauty of that projection is it takes all the polls conducted to date in all the states... Rasmussen, Zogby, Opinion Dynamics, Mason Dixon, Harris, Gallup, Survey USA... etc and plots them all on an XY axis and then performs a regression analysis to predict with probably a 95% certainty where the final results will be. Now there is 5 weeks left, that is plenty of time for something happen to change the outcome. Regression analysis assumes no major deviations and a continuation of trends. If trends continue, than what you see on that map is about a 95% certainty. An October surprise by either side could change things dramatically, and if OH, PA and FL are neck and neck such a surprise could alter the final outcome.
So if you are in those three states, pound the shit out of the pavement and get our leads up! |
|
At least it isn't blue outlined! I assume that all white means a dead heat. |
|
|
You're 0 for 2 today Sarge. Mybe you should just go back to bed.
|
|
As a Pennsylvania Republican this situation is a frustrating one, but one that we just have to live with. The good news is that President Bush can win without PA, but knowing that if he were to win PA his reelection would probably be automatic is frustrating. I've long doubted that Bush will win PA. I'm sorry about that, but it's the way I see it as a lifelong resident of the Keystone State.
There are something like 400,000 more demos in PA than repubs. This means that everything hinges on turnout. There has also been a MAJOR registration effort here this Summer. It's hard to imgine that those efforts have signed up more repubs than demos. The BIG problem is this; we have a little problem here in Pennsylvania called PHILADELPHIA! The ciy of Philadelphia is the 900 lb. gorrilla that determines statewide elections. When the 900 lb. gorrilla is asleep, anyone can win a statewide race for say Governor or President, but when the political issues are highly emotional and polarized, and the machine in Philly gets out the demo vote, PA goes Democratic. And that's what I see happening this year (same as 2000). |
|
Bush is right up the street from my office this afternoon. They are expecting this to be the largest rally of the campaign so far with 40,000-50,000 in attendence.
When I drove past this morning on the way in, they had semi trailers butted end-to-end as barriers around the perimiter of the park. LEO presence everywhere. Ohio is definately in play. The major metro areas are typically Dem swinging except for Cincinnati, which has a huge Republican base. The swing isn't in the cities, it's in the rural towns and counties. Ohio is the Florida of 2004. ETA: CNN Headline News was running a statistic this weekend about Ohio and new voter registration. They had the Democrat registrations up 200+% with GOP registrations up only 25%. Of course, registrations vs. actual vote are two different animals. |
|
|
What in the hell happened to Georgia?! They must have just polled in Cynthia McKinney's district.
I am ashamed that Georgia would ever be anything but solid, dark red. |
|
Rasmussen poll, add three points... still won't change the coloration as it will put it on the cusp of being light red... but Rasmussen uses extremely conservative methodologies that downplay any candidate's lead. |
|
|
PA is screwed by pittsburgh and Philly, plus the fact that there are still a bunch of idiots hoping for the steel mills to open back up
|
|
Check out WA is even hollow blue now. If my state is slipping, sKerry is certainly in trouble!
|
|
i think realistically President Bush will get around 290 electoral votes. |
|
|
I saw the President at the Lake County Rally a few weeks ago. He and his family were less than 20 feet away from us. He was right on target and hit it outada park. Ben is right, everything outside the Two C's is bush country.
|
|
To think Arnie hasn't said shit for W. |
|
|
I have a real problem with the linear regression methodology used on that site to make the final prediction. Because he uses multiple polls, each of which uses different methods, there is NO WAY to know if the regression line is valid. The poll-to-poll variation is too great, and could be masking what is really happening. In my current line of work we would conclude that the measurement system variation is too great for us to draw any statistically valid inferences.
In plain English: Don't get cocky. Keep working hard for W. |
|
Actually the use of multiple polls cuts both ways... yes it factors in differing methodologies... but it also factors out some of the inherent bias represented in some of those methodologies. For example, it is pretty well accepted that Zogby's polls are skewed againt the president. Then there are polls which are skewed toward the president such as Survey USA. By factoring in a variety of polls over time you can actually eliminate some of the bias. Hell even polls done by the same firms don't always ask the same questions... so you can't stick to one firm all the time.
However, we're not talking about the nuances of issue positions here in which people are asked right track/wrong track... approve/disapprove. We are talking about one simple question... if the election were held today, who would you vote for? Because that is the SINGLE question that forms the basis for the regression, you can use multiple polls. True, some polls use likely voters and some registered voters... they might have larger or smaller sample sizes, but the one issue they are asking about in common is, "if the election were held today, who would you vote for for president." I would have a hige problem with plotting say approval of the War in Iraq using multiple polls for the very reason you suggest. Depending on how the question is asked it will skew the results. But once again, we aren't dealing with nuances here... its a straight forward question... Bush, Kerry or other... I so concur, do not fucking get complacent. I don't care if your state is so red its becoming black... get your ass out there and vote, get your friends registered... contact the party and find out which shut-ins need rides to the polls and get them there. Fight Fight FIGHT! |
|
I do not wish to pick nits, but I work with statistics for a living. Anytime we measure something, the act of measurement itself introduces variation. This is known as measurement system variation. Unless we know the magnitude of this measurement system variation (and there are tools for doing this) we cannot know whether what we are seeing is actual variation in the thing measured, or measurement system variation.
The "If the election were held today" question is not as simple as it seems. Different polls use different methods to both collect and calculate the results (weighting, sampling methods, etc). As you already know, Rasmussen, Pew, and Zogby give significantly different results than say Gallup. When I blend all these polls together, I cannot tell whether there is an actual trend (which is what the linear trendline indicates), or whether this is the result of random variation. A better method would be to look at linear trendlines for each of the polls. For example, what is the linear trend prediction for Zogby, for Pew, for Gallup, etc. Then calculate an expected final result, with mean and standard deviation, from the linear projection final results of all these individual polls. The best method would be to use an Auto Regressive Integrated Moving Average (ARIMA) analysis (after testing for auto- and cross-correlation), since we are dealing with time-series data. But again, you would need to account for the measurement system variation before even this would have any validity. God, that sounds really nerdy. So, I will repeat by earlier advice: Don't get cocky. Keep working hard for W |
|
I agree its not perfect, yes you do sound nerdy , but I know exactly what you are talking about. And while this is very unscientific... one of the things we do when formulating campaign strategy in light of a multitude of polls is to go with a gut check. Yup, you heard it here first... we look at the various polls, we take our past experience with the various pollsters, combine it with our observations of the electorate in question, and we make a damn good guess as to what to do... and the campaign that makes the best guesses usually wins.
I believe in polls as snapshots of the electorate's mood. Because the electorate is so diverse, your sample size is never going to be representative enough... even when you weight the sample to reflect what you feel the respresentation should be in terms of demographic make-up. Pure statistical analysis will never be completely accurate... hell I know several statisticians who firmly believe that statistical analysis has no place in politics until after the election as a means to understand what happened... never as a means to predict. And the one issue I disagree with you on Paeber is using the trendlines for the same polls as a predictor. As I mentioned, unless the polls are using the same questions, same sample size, same weighting, etc... this isn't going to be much better. Look at the Pew poll that had the president up 14 after the convention... their next poll had the president up 12... they thought that was too high, so they revamped the questions, came up with a new methodology for their sample, recanvassed... and low and behold, the 12 point lead is now 1 point. Unfortunately, there does not exist a poll that provides the type of polling integrity you and I crave. In light of the games some firms play... I do feel the regression provided at electoral-vote is useful, imperfect as it is, as it will tend to mitigate some of the outliers. But that is just one of those unscientific gut checks that comes from campaign work versus real statistical analysis. What I think is happening is Kerry's people are waiting for the numbers to roll out and then campaigning to them... in other words... if the polls show confidence in the economy is lagging, that is the message they will seize upon in an effort to exploit it. Bush's people are doing the exact opposite, the have staked out their issues and are hitting them hard and watching the polls to see if that issue is indeed resonating... and refining rather than moving from issue to isse... which is why Kerry is seen as indecisive, pessimistic and a flip-flopper while Bush seems to be steady and has greater integrity. NOW LISTEN UP MAGGOTS! YOU DO WHAT PAEBER SAYS AND GET YOUR ASSES TO WORK! |
|
My thoughts exactly! Many here in Pittsburgh think that by holding on to their union wet dreams the mills will magically reappear along the rivers. Sadly that is part of the reason why the mills are gone. Plus in PGH you have a very aged population that still thinks Democrat means FDR and winning WWII. Things have changed, their voting trends haven't. I'm still suprised we have Rick Santorum and Arlen Specter as Senators although Arlen is a RINO. The Bush rallies in PA (he is here every other week it seems) have had a huge turnout and his recent trip here to view flood damage from Ivan might help. Unfortunately I heard a caller to a local talkshow say that he was angry that Iraq got X amount of billions to rebuild while he was only being offered a low interest loan from FEMA. Why does everyone want handouts? |
|
|
I'm hearing that Maryland might be going BUSH!!! I can't believe it first a Repub Governor, whom is doing a great job and is Pro-Gun now a possibility of our 10 votes going to Bush. WHOO-HOO! Who'd have guessed??
|
|
BBQ, I have no explanation for MD... I would love to see it happen... maybe its a bunch of federal employees who may not be Republican but see that Bush is actually trying to protect this country.
|
|
I would suggest any polling out of FL right now cannot be reliable.
|
|
Max, that is an extremely valid point... because of evacuations they can only reach certain parts of the state... that and mass phone service outages put many more out of touch.
|
|
Absolutely correct...As far as Pennsylvania goes, Kerry is hammering in Pittsburgh and Philadelphia, but the entire center of the Commonwealth votes strong BUSH. Look at the classic Red and Blue map from 2000. They seem to find more and more voters in the urban areas that haven't registered yet. |
|
|
The univariate linear model is a joke in these situations. The data that most of the press is using is univariate-single question (with a qualifying question for "likely voters") and tells very little of the actual story. PAEBR332 is completely correct in referencing the need for an ARIMA model. More important, though, is to look at more than just polling data. There are some VERY good models out there that predict the electoral vote count of the incumbent party (and have been accurate 18 of the last 20 elections).
Also, many polls still have a lot of these swing states within the margin of error, which makes the predictions even more sketchy. I personally put this in the "too close to call...still" category. However, the Kerry campaign doesn't have a message, is largely defensive, and can't seem to strike a chord with the common people on a daily basis. Gaffs like "Lambert Field", and his discussion of the proud tradition of Buckeye football (IN ANN HARBOUR MICHIGAN) are indicative of his disconnectedness, and they really do turn people off. Talking about a local situation for an hour at a rally, and not knowing ANYTHING about local culture are bad juju. shooter |
|
True. Something else that the pro-Bush folks from out of state don't understand is this; Pennsylvanians can and do elect Republicans, but Pennsylvanians prefer RINO Republicans like Arlen Specter over more conservative "born again", restrict stem cell research, anti abortion Republicans like President Bush. Tom Ridge was a pro-choice, "Nelson Rockefeller" type Republican. Rick Santorum is really something of an outlier on the data graph. Also, and this is something very frustrating for gun owners and the NRA, PA gun owners are traditionally also very pro-Union Democrats. The 2000 election was very interesting because it forced many PA gun owners between a rock and a hard place; "Do I vote for the "evil" Republicans whom I've NEVER voted for before?", "Do I vote for my guns and against the UNION?" The PA Democrat gun owner has never faced such a quandary before. There are signs that some of these guys may end up making the same decision many Southern voters made years ago when they went over from Democrat to Republican on the grounds of concerns over multiple social issues, but I'm not holding my breath. My guess is that PA demo gun owners probably won't go so far down this route because I don't believe that religion will play as much a part in their decision as down South. Also PA gun owners are primarily hunters. The problem with this is that for these guys, guns are something that they pick up exactly twice a year; once to go to the range to sight-in and then one more time during hunting season to go into the field. Too many of them are really about as commited to the issue of gun rights as I am to NFL football (sure, it's a fun way to get together with old highschool friends, but if the NFL disappeared tomorrow I really wouldn't care that much). I'm rooting for President Bush. And he's doing the right thing to come here and fight in PA. He CAN win this state, I'm just doubtful that he will. |
||
|
2 words: Employed Nerds |
|
|
3 words: Well paid nerds. |
||
|
Glad they pay the nerds well... Republican operatives are poorly paid, unless you are at the upper echelons. Hell, I'm lucky if I can get my clients to pay half the time... local races always seem to squander the campaign money on ad buys and try to fund raise to cover consultant fees AFTER the race has been lost... That's why most of us have day jobs, we figure we won't ever see a living from doing this alone... although the networking becomes invaluable.
So, what are these models that predicted the electoral vote before the election in the past 18 out of 20 races? I'd love to dive into these! |
|
A handful of political scientists (notable Lewis-Beck and Erikson) have developed some really good models. I may have overstated the accuracy of the Lewis-Beck model (called the "Jobs Model"), but it appears to have predicted the outcome of every presidential race since 1952 excepting 1976. The model looks at economic data, popularity of incumbent party, approval ratings, and a couple other independent variables. I can get you the real political science version of them, because the mainstream press doesn't delve into the math of the equation very well. shooter |
|
|
Boy, there is nothing like waiting to see how bad you are losing (and on what issues) to devise strategy. If my students did this on exams my life would be fun. shooter |
|
|
I am a Six Sigma Master Black Belt. I bet that clears things up. |
||
|
Shooter, I'd love to see those models if you stumble across them... no rush.
As for how Kerry's team is operating, I can't explain it any other way. They have gotten a little more disciplined since some of the Clinton folks came on board... at least in terms of Kerry delivering the message and creating sound bites... but I still think it is the wrong message and they are letting the polls determine what they say on any given day. You can note that their campaign is purely reactive, which is a huge mistake... you never allow the incumbent to define the playing field... which Bush has done. In fact, what Bush has done is absolutely brilliant and has only allowed Kerry two choices... well three, but I'll show why its two... and in fact really one. He can agree with Bush on the apporach to the war on terror. This will secure him some of the middle ground and retain some of the democrats who are defecting to Bush. However, in doing so he risks losing the Dean supporters who already view him as suspect and may well vote Nader or not vote if they view Kerry as being pro-war. He can move to the right of Bush and be the uber-hawk... this most assuredly would lose him the left vote, but could pick up some of the Greatest Generation who would find his social agenda appealing as well as his tough stance on terrorism. He wouldn't pull many Republicans from Bush, but he could once again capture some of those moderate and conservative democrats who had defected. He can move to the left and question Bush at every turn on the war... deployments, equipment, rules of engagement, allies... Basically attempt to kill him with a thousand paper cuts. In doing so he will maintain his left... hold the Anybody Bush crowd and probably not lose any more support than he already has from conservative and moderate democrats. It is safe territory in the sense that he need not worry about Bush moving into that realm. Kerry knows he has lost too many votes to Bush on prescription drugs, education, tax cuts... He needs to pick an issue and position that is clearly his... so now we have the anti-war Kerry... quagmire, 1,000 deaths, beheadings, elections threatened, Allawi is a liar... Essentially Bush only left him one area to stake out, which is exactly where Bush wants him... the liberal left New England Gazillionaire who wants to raise taxes, take your guns, give your money to other people, cave to the unions and trial lawyers... Kerry is just too damn dumb to realize that his flip flopping early on has painted him into a corner... now he must defend that position with his dying breath, any deviation will just further erode his numbers. You know, this is going to sound corny... but Bush's political kung-fu is strong... well at least Rove's and Hughes's are. |
|
A bit. We recently hired a six sigma guru to lead the product implementation area (of an insurance company). They make too many mistakes. |
|||
|
Penn has too many Deer hunters (yellow dog democrats) and Unions (Communist) to go with Bush. Penn went with Gore in 2000 and will go with Kerryin 2004.
|
|
Sign up for the ARFCOM weekly newsletter and be entered to win a free ARFCOM membership. One new winner* is announced every week!
You will receive an email every Friday morning featuring the latest chatter from the hottest topics, breaking news surrounding legislation, as well as exclusive deals only available to ARFCOM email subscribers.
AR15.COM is the world's largest firearm community and is a gathering place for firearm enthusiasts of all types.
From hunters and military members, to competition shooters and general firearm enthusiasts, we welcome anyone who values and respects the way of the firearm.
Subscribe to our monthly Newsletter to receive firearm news, product discounts from your favorite Industry Partners, and more.
Copyright © 1996-2024 AR15.COM LLC. All Rights Reserved.
Any use of this content without express written consent is prohibited.
AR15.Com reserves the right to overwrite or replace any affiliate, commercial, or monetizable links, posted by users, with our own.