Which presidential polls were most accurate?

[DeepModem Mom]

Source: Houston Chronicle

The Pew Research Center and Rasmussen Reports were the most accurate in predicting the results of the 2008 election, according to a new analysis by Fordham University political scientist Costas Panagopoulos.

The Fordham analysis ranks 23 survey research organizations on their final, national pre-election polls, as reported on pollster.com.

On average, the polls slightly overestimated Obama's strength. The final polls showed the Democratic ahead by an average of 7.52 percentage points -- 1.37 percentage points above his current 6.15-point popular vote lead. Seventeen of the 23 surveys overstated Obama's final victory level, while four underestimated it. Only two -- Rasmussen and Pew -- were spot on.

Here is the list --

1T. Rasmussen (11/1-3)**
1T. Pew (10/29-11/1)**
3. YouGov/Polimetrix (10/18-11/1)
4. Harris Interactive (10/20-27)
5. GWU (Lake/Tarrance) (11/2-3)*

6T. Diageo/Hotline [phone])*
6T. ARG (10/25-27)*
8T. CNN (10/30-11/1)
8T. Ipsos/McClatchy (10/30-11/1)
10. DailyKos.com (D)/Research 2000 (11/1-3)

11. AP/Yahoo/KN (10/17-27)
12. Democracy Corps (D) (10/30-11/2)
13. FOX (11/1-2)
14. Economist/YouGov (10/25-27)
15. IBD/TIPP (11/1-3)

16. NBC/WSJ (11/1-2)
17. ABC/Post (10/30-11/2)
18. Marist College (11/3)
19. CBS (10/31-11/2)
20. Gallup (10/31-11/2)

21. Reuters/ C-SPAN/ Zogby (10/31-11/3)
22. CBS/Times (10/25-29)
23. Newsweek (10/22-23)

Read more: http://blogs.chron.com/txpotomac/2008/11/the_list_which_presidential_po.html

[msongs]

1. didn't 538.com have it correct to the tenths??? nt

[DeepModem Mom]

2. Yes, I think they were accurate, but this study apparently doesn't include 538. nt

[Texas Mom]

4. 538 NOT a pollster

It analyzes others' polls and runs statistical probabilities.

[DeepModem Mom]

7. Thanks, Texas Mom -- welcome to DU!!!

[defendandprotect]

3. Actually, I'm inclined to think they underestimated Obama's popularity ...

and overestimated McCain's ...

How did all of this jibe with EXIT POLLS --- ???

[elocs]

5. Thanks, I was wondering about this exact question. Let's remember it for next time. n/t

[alfredo]

6. Rasmussen's daily tracking was excellent.

[Catch22Dem]

8. Nate Silver deserves a lot of credit too

Nate is not a pollster, so obviously shouldn't be included on a list of pollsters, but his analysis of the polls on fivethirtyeight.com should be commended. He tweaked his model early and often and made it extremely accurate. In the end, this is his model vs. the final outcome.

Nate's final map:


Actual outcome:

[DeepModem Mom]

9. Thanks, Catch22Dem! Nate is amazing. nt

[customerserviceguy]

10. I was curious who was most accurate in state-by-state polls

and this seems to have answered the question. Only Indiana' flip from red to blue did not get accounted for by the above model.

Still, it would be fun to see a statistical analysis of the the percentages projected for each state, and an overall ranking of accuracy on that level.

[stanwyck]

11. But Sean and Drudge assured us "the most accurate

is TIPP/IBD". Yep. For the two weeks proceeding the election, Hannity and Drudge kept the drumbeat going that "the most accurate poll TIPP/IBD" had the race within 2 pts.
Damn. Those guys are almost never wrong about anything.