Warning

 

Close

Confirm Action

Are you sure you wish to do this?

Confirm Cancel
Member Login
Arrow Left Previous Page
Page / 2
Posted: 9/27/2004 4:38:04 AM EST
Link Posted: 9/27/2004 4:39:39 AM EST
Goddamnit


Too many states in play... when Ohio goes, we go.


- BG
Link Posted: 9/27/2004 4:42:49 AM EST
On the bright side, this map shows that President Bush can win without Florida.
Link Posted: 9/27/2004 4:43:40 AM EST
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?
Link Posted: 9/27/2004 4:44:36 AM EST
Look at Cal...look at cal...


SGatr15
Link Posted: 9/27/2004 4:46:14 AM EST
Minnesota is looking a lot better today than it was last week
Link Posted: 9/27/2004 4:47:27 AM EST
I've already got my absentee ballot requested. My vote will be cast, by God!
Link Posted: 9/27/2004 4:47:44 AM EST
Link Posted: 9/27/2004 4:49:06 AM EST

Originally Posted By HiramRanger:
I think Florida is still our's... PA I don't know. Those guys in Florida really need to be pounding the pavement. The regression analysis for all the polls with a final projection still shows Florida and Pennsylvania going to Bush with this being the final projection.

www.electoral-vote.com/fin/sep27p.png

FINAL PROJECTION... Kerry 192 Bush 336


I sure hope they are right, because that final is a beautiful site
Link Posted: 9/27/2004 4:57:04 AM EST
[Last Edit: 9/27/2004 4:57:28 AM EST by cwd10]
Unfortunately, PA will go Kerry. The big cities just carry too much weight for us in the rural area's to overcome.
Link Posted: 9/27/2004 4:58:58 AM EST

Originally Posted By cwd10:
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
Link Posted: 9/27/2004 4:59:28 AM EST
Link Posted: 9/27/2004 5:02:05 AM EST

Minnesota is looking a lot better today than it was last week


At least it isn't blue outlined!

I assume that all white means a dead heat.
Link Posted: 9/27/2004 5:03:51 AM EST
You're 0 for 2 today Sarge. Mybe you should just go back to bed.
Link Posted: 9/27/2004 5:10:44 AM EST
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).
Link Posted: 9/27/2004 5:15:25 AM EST
[Last Edit: 9/27/2004 5:16:54 AM EST by BenDover]
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.
Link Posted: 9/27/2004 5:15:52 AM EST
Link Posted: 9/27/2004 6:00:01 AM EST
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.
Link Posted: 9/27/2004 6:06:23 AM EST
Link Posted: 9/27/2004 6:09:02 AM EST
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

Link Posted: 9/27/2004 6:10:30 AM EST
Check out WA is even hollow blue now. If my state is slipping, sKerry is certainly in trouble!
Link Posted: 9/27/2004 6:19:30 AM EST

Originally Posted By Admiral_Crunch:
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.

don't worry, the President will take GA by at least 8 points.

i think realistically President Bush will get around 290 electoral votes.
Link Posted: 9/27/2004 6:24:41 AM EST
Link Posted: 9/27/2004 6:26:21 AM EST

Originally Posted By sgtar15:
Look at Cal...look at cal...


SGatr15



To think Arnie hasn't said shit for W.
Link Posted: 9/27/2004 6:26:42 AM EST
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.
Link Posted: 9/27/2004 6:53:40 AM EST
Link Posted: 9/27/2004 7:13:25 AM EST
[Last Edit: 9/27/2004 7:14:50 AM EST by PAEBR332]
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
Link Posted: 9/27/2004 7:34:38 AM EST
Link Posted: 9/27/2004 7:40:56 AM EST
SUPPORT FOR KERRY IS WEAKENING IN MICHIGAN!!!

Link Posted: 9/27/2004 7:48:20 AM EST

Originally Posted By RAVYN:
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




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?
Link Posted: 9/27/2004 7:49:31 AM EST
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??
Link Posted: 9/27/2004 8:03:38 AM EST
Haha! Oregon is red today!
Link Posted: 9/27/2004 8:03:40 AM EST
Link Posted: 9/27/2004 8:15:07 AM EST
[Last Edit: 9/27/2004 8:15:42 AM EST by Max_Mike]
I would suggest any polling out of FL right now cannot be reliable.
Link Posted: 9/27/2004 8:31:26 AM EST
Link Posted: 9/27/2004 8:43:27 AM EST
I feel outright dirty in blue. Time to wash this shit up.
Link Posted: 9/27/2004 8:43:35 AM EST

Originally Posted By Max_Mike:
I would suggest any polling out of FL right now cannot be reliable.




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.
Link Posted: 9/27/2004 8:48:11 AM EST
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
Link Posted: 9/27/2004 9:10:10 AM EST

Originally Posted By PAEBR332:
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.

...




Originally Posted By shooter220:
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.

...





NERDS! NERDS! NERDS! NERDS!
Link Posted: 9/27/2004 9:19:29 AM EST

Originally Posted By wgjhsafT:

Originally Posted By RAVYN:
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




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?



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.
Link Posted: 9/27/2004 9:57:29 AM EST

Originally Posted By Nimrod1193:


photos.ar15.com/ImageGallery/Attachments/DownloadAttach.asp?iImageUnq=25525

NERDS! NERDS! NERDS! NERDS!



2 words:

Employed Nerds
Link Posted: 9/27/2004 10:05:08 AM EST

Originally Posted By shooter220:

Originally Posted By Nimrod1193:


photos.ar15.com/ImageGallery/Attachments/DownloadAttach.asp?iImageUnq=25525

NERDS! NERDS! NERDS! NERDS!



2 words:

Employed Nerds



3 words: Well paid nerds.
Link Posted: 9/27/2004 10:19:57 AM EST
Link Posted: 9/27/2004 10:25:51 AM EST

Originally Posted By PAEBR332:



what do you do for a living?
Link Posted: 9/27/2004 10:32:03 AM EST

Originally Posted By HiramRanger:


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
Link Posted: 9/27/2004 10:33:29 AM EST

Originally Posted By HiramRanger:

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!



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
Link Posted: 9/27/2004 10:45:42 AM EST
It looks like OR is heading in the right direction.
Link Posted: 9/27/2004 10:58:51 AM EST

Originally Posted By Red_Beard:

Originally Posted By PAEBR332:



what do you do for a living?



I am a Six Sigma Master Black Belt.

I bet that clears things up.
Link Posted: 9/27/2004 11:06:12 AM EST
Link Posted: 9/27/2004 11:19:07 AM EST

Originally Posted By PAEBR332:

Originally Posted By Red_Beard:

Originally Posted By PAEBR332:



what do you do for a living?



I am a Six Sigma Master Black Belt.

I bet that clears things up.




A bit.

We recently hired a six sigma guru to lead the product implementation area (of an insurance company). They make too many mistakes.

Link Posted: 9/27/2004 11:26:48 AM EST
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.
Arrow Left Previous Page
Page / 2
Top Top