Here are the final 2004 pre-election polls (national and key states)

[WI_DEM]

RCP Average FINAL 50.0% 48.5% 1.0% Bush +1.5
RCP Average 10/27 - 11/1 48.9% 47.4% 0.9% Bush +1.5
Marist (1026 LV) 11/1 49% 50% 0% Kerry +1
GW/Battleground (1000 LV) 10/31 - 11/1 50% 46% 0% Bush +4
TIPP (1041 LV) 10/30 - 11/1 50.1% 48.0% 1.1% Bush +2.1
CBS News (939 LV) 10/29 - 11/1 49% 47% 1% Bush +2
Harris (1509 LV) 10/29 - 11/1 49% 48% 2% Bush +1
FOX News (1200 LV) 10/30 - 10/31 46% 48% 1% Kerry +2
Reuters/Zogby (1208 LV) 10/29 - 10/31 48% 47% 1% Bush +1
CNN/USA/Gallup(1573 LV)* 10/29 - 10/31 49% 49% 1% TIE
NBC/WSJ (1014 LV) 10/29 - 10/31 48% 47% 1% Bush +1
ABC/Wash Post (2904 LV)** 10/28 - 10/31 49% 48% 0% Bush +1
ARG (1258 LV) 10/28 - 10/30 48% 48% 1% TIE
CBS/NY Times (643 LV) 10/28 - 10/30 49% 46% 1% Bush +3
Pew Research (1925 LV) 10/27 - 10/30 51% 48% 1% Bush +3
Newsweek (882 LV) 10/27 - 10/29 50% 44% 1% Bush + 6

Bush leads in 10, Kerry in 2 and 2 are ties

http://www.realclearpolitics.com/bush_vs_kerry.html

by contrast here, less than a week before the election is where we stand today:

RCP Average 10/22 - 10/28 -- -- 49.9 44.0 Obama +5.9
Rasmussen Reports 10/26 - 10/28 3000 LV 2.0 50 47 Obama +3
Gallup (Traditional)* 10/26 - 10/28 2435 LV 2.0 49 46 Obama +3
Gallup (Expanded)* 10/26 - 10/28 2409 LV 2.0 51 44 Obama +7
Diageo/Hotline 10/26 - 10/28 870 LV 3.6 49 42 Obama +7
Reuters/C-SPAN/Zogby 10/26 - 10/28 1179 LV 2.9 49 44 Obama +5
IBD/TIPP 10/24 - 10/28 894 LV 3.0 47 44 Obama +3
GWU/Battleground 10/22 - 10/28 1000 LV 3.1 49 46 Obama +3
ABC News/Wash Post 10/24 - 10/27 1301 LV 2.5 52 45 Obama +7
Ipsos/McClatchy 10/23 - 10/27 831 LV 3.4 50 45 Obama +5
Pew Research 10/23 - 10/26 1198 LV 3.5 53 38 Obama +15

(Obama leads in all polls)

http://www.realclearpolitics.com/polls/

Florida
ARG | 10/30-11/1 600 LV 4.0 48 50 1 Kerry +2
Zogby | 10/29-11/1 601 LV 4.1 48 48 - TIE
FOX News | 10/30-31 700 LV 3.0 44 49 1 Kerry +5
SurveyUSA | 10/29-31 742 LV 3.7 49 48 - Bush +1
Insdr Adv | 10/29-31 400 LV 5.0 48 48 1 TIE
Quinnipiac | 10/27-31 1098 LV 3.0 51 43 1 Bush +8
CNN/USAT/Gallup | 10/28-30* 1138 LV 4.1 47 50 - Kerry +3
Rasmussen | 10/25-31 600 LV 4.0 50 47 - Bush +3
Mason-Dixon | 10/27-29 625 LV 4.0 49 45 - Bush +4

Ohio
RCP Average | 10/25-11/1 - - 48.8 46.7 Off Ballot Bush +2.1
Zogby | 10/29-11/1 601 LV 4.1 49 43 - Bush +6
FOX News | 10/30-31 700 LV 3.0 50 47 - Bush +3
SurveyUSA | 10/29-31 816 LV 3.5 49 47 - Bush +2
CNN/USAT/Gallup | 10/28-31 1111 LV 4.0 46 50 - Kerry +4
Ohio Poll/UofC | 10/27-31 877LV 3.3 50.1 49.2 - Bush +0.9
Mason-Dixon | 10/27-29 625 LV 4.0 48 46 - Bush +2
Rasmussen | 10/25-10/31 600 LV 4.0 50 46 - Bush +4
Clev. Plain Dealer | 10/26-28 1500 LV 2.6 48 45 - Bush +3
(Kerry ahead in only one poll)

Pennsylvania (compare these numbers to the double-digit leads Obama has currently in PA)
RCP Average | 10/25-11/1 - - 47.3 48.2 Off Ballot Kerry +0.9
SurveyUSA | 10/31-11/1 657 LV 3.9 48 49 - Kerry +1
Zogby | 10/29-11/1 601 LV 4.1 46 50 - Kerry +4
CNN/USAT/Gallup | 10/28-31 601 LV 4.1 50 46 - Bush +4
Quinnipiac | 10/27-31 909 LV 3.3 47 47 - TIE
Rasmussen | 10/25-31 600 LV 4.0 47 49 - Kerry +2
Mason-Dixon | 10/27-29 625 LV 4.0 46 48 - Kerry +2

Nevada (not one showing a Kerry lead)
Poll | Date Sample MoE Bush Kerry Nader Spread
RCP Average | 10/27-11/1 - - 51.0 44.7 - Bush +6.3
SurveyUSA | 10/30-11/1 636 LV 4.0 53 45 - Bush +8
Zogby | 10/29-11/1 601 LV 4.1 50 45 - Bush +5
Mason-Dixon | 10/27-29 625 LV 4.0 50 44 - Bush +6

Colorado (not one showing a Kerry Lead)
RCP Average | 10/25-11/1 - - 50.0 44.8 1.0 Bush +5.2
Survey USA | 10/30-11/1 705 LV 3.8 50 47 - Bush +3
Zogby | 10/29-11/1 600 LV 4.1 49 47 - Bush +2
Mason-Dixon | 10/27-29 625 LV 4.0 50 43 1 Bush +7
Rocky Mtn News | 10/25-27 500 LV 4.3 51 42 - Bush +9

New Hampshire
Poll | Date Sample MoE Bush Kerry Nader Spread
RCP Average | 10/26-11/1 - - 47.5 48.5 1.3 Kerry +1.0
ARG | 10/30-11/1 600 LV 4.0 49 48 1 Bush +1
UNH Tracking | 10/29-31 716 LV 3.7 49 50 1 Kerry +1
Mason-Dixon | 10/27-29 625 LV 4.0 46 47 - Kerry +1
Research 2000 | 10/26-28 600 LV 4.0 46 49 2 Kerry +3
http://www.realclearpolitics.com/bush_vs_kerry_sbys.html

[FSogol]

1. Thanks for the post.

It helped me retain my sanity for another day.

:beer:

[brazos121200]

2. I remember being so sure Kerry would win, but seeing these

old polls I don't know now how I could have been so sure. The situation certainly is different today, and save for a catastrophe, there is no way McCain will pull out a surprise. Five more full days of campaigning, not counting election day, to go is not enough time for McCain to turn this thing around legitimately.

[WI_DEM]

3. part of it, I think, is that when an incumbent is so close you think a challenger can pull it off.

Also Kerry had some huge crowds in those last days.

[brazos121200]

4. Yes, and part of it was also that I figured that with 7 or 8 percent

of poll respondents saying they were undecided, most would go to Kerry, or go goes the conventional wisdom. Why would people wait to the last minute to vote for Bush when you already know what he stands for? In fact, most of those undecideds went for the incumbent, the first time that has happened. And Rove's election theft machine didn't hurt, either. That was probably good for 4-6 percent.

[Drunken Irishman]

10. Based on past races, Kerry should have won.

#1 If an incumbent is polling below 50%, they almost always lose. Bush was polling below 50%.

#2 Undecideds typically break from the incumbent, since they clearly have doubts about the leadership and normally those doubts are too much to overcome and they give in, voting for the challenger. Undecideds in many polls made up 8+ points nationally.

#3 Kerry was surging in the final week. He had all the momentum, the big rallies, the polls tightening and the perception of the above two that helped with the belief he'd win.

Now how'd it all unfold?

Well #2 proved to be wrong. Undecideds broke heavily for Bush, which made #1 wrong, too, since he was able to reach 50+1 on election day with the help of those undecideds. #3 didn't matter because undecideds broke in a very close race in both Ohio & Florida.

[varun]

5. Thanks

Now I can relax.

I know that Obama will win next tuesday.

[WI_DEM]

6. As long as we all vote he will.

[texasleo]

7. Thanks

[tclambert]

8. RCP now has Nevada leaning Obama

And that puts him at 311 electoral votes solid or leaning. Only 270 needed to win. I'm sticking by my prediction of Obama winning by about 200 electoral votes, right around 369 to 169.

[Jennicut]

11. That would be great. Actually realized the other day that all Obama needs

is Kerry blue states plus VA and CO or VA and NM. It would be 273 to 265. Important to hold onto PA and glad NH has really come over to Obama.

[geekgrrl]

9. And see "this day in 2004" for more sanity!!!

http://www.electoral-vote.com/evp2004/Pres/Maps/Oct29.html

And breathe...!