Posted on 11/01/2012 9:18:38 AM PDT by o2bfree
Today the unweighted Real Clear Politics poll average may be more Trick than Treat! RCP reports a tied race when averaging polls having low margins of error with polls having much higher margins of error. However, Romney gets a signficant lead when you weight the polls in a weighted average.
The weighted average seems to show that that Romney is clearly ahead, by almost a point! See it for yourself:
http://goo.gl/xyV5H
How it works: Polls with larger samples have a higher weight than polls with smaller samples.
This is the correct way to do it, except that you need to take out the Network/Newspaper polls. These are notorious for signficant bias and poor methodology. That leaves Rasmussen, Pew, and Gallup. When you do the same methodology, but only with those three polls, you get Romney 49.4, 0bama: 46.5, spread of 2.9 in favor of Romney.
Great comment. I’ll have to modify the spreadsheet I shared on the link to reflect this. Or maybe add a section to the spreadsheet... that could work too!
2008 most accurate:
1. Rasmussen (11/1-3)**
1. Pew (10/29-11/1)**
2. YouGov/Polimetrix (10/18-11/1)
3. Harris Interactive (10/20-27)
4. GWU (Lake/Tarrance) (11/2-3)*
5. Diageo/Hotline (10/31-11/2)*
5. ARG (10/25-27)*
6. CNN (10/30-11/1)
6. Ipsos/McClatchy (10/30-11/1)
7. DailyKos.com (D)/Research 2000 (11/1-3)
8. AP/Yahoo/KN (10/17-27)
9. Democracy Corps (D) (10/30-11/2)
10. FOX (11/1-2)
11. Economist/YouGov (10/25-27)
12. IBD/TIPP (11/1-3)
13. NBC/WSJ (11/1-2)
14. ABC/Post (10/30-11/2)
15. Marist College (11/3)
16. CBS (10/31-11/2)
17. Gallup (10/31-11/2)
18. Reuters/ C-SPAN/ Zogby (10/31-11/3)
19. CBS/Times (10/25-29)
20. Newsweek (10/22-23)
Still Gallup is in 17th last year...beat out by ABC/Post and CBS.
The only reassuring thing is that they over-estimated Obama by a whole 2 points.
Interestingly, almost exactly in line with Rasmussen: 49-47.
Gallup apparently stands alone in its decision to suspend polling because of the effects of Sandy.
It seems Sandy will provide cover to just about all the polling firms. If they miss badly, they can just blame it on Sandy.
RCP Avg. is a waste of time at this point in the cycle.
Stick w Rassmusen for clearest picture at State & Federal level.
Too much garbage in the other polls for them to be useful in establish an “average” of anything.
If you go beyond removing the newspaper polls and change the weighting a bit to even D/R or just D+1, it’s even better.
Honestly, my feeling is Romney is ahead by 4.
I have been looking at the social media tracker on NBC News/MSNBC elections site since that is always interesting, and while it has swung a bit in Obama’s direction in terms of people announcing votes online, Obama still has a significantly more negative social media message being spread online on average.
I don’t see a shift to Obama yet at least from there.
At this point, the polls, as a group, ttell us almost nothing better than our own guesses would have been, all by themselves. You might as well keep looking at the football odds from day to day to determine who is going to win a game.
It is close enough that IMHO it is ALL about turnout.... are the Repubs and Indies more hyped than the Dims and Libs... that is the real polling question.
You are absolutely right. Pretty clear there are media polls in the RCP average that are DESIGNED to boost Obama’s average.
Subtacting polls from RCP with small sample sizes (< 1000) moves the weighted spread up to 1.63.
No doubt there is a SIZABLE lead for Romney at this point.
Now, Sandy concerns me, but no real shift yet, even slight, in terms of the perception of Obama on social media. I would think if there was, that would be improving for him ahead of any poll shift. Nothing yet.
Waiting with baited breath for the first Gallup.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.