Guess Who Was The Most Inaccurate Pollster In 2002

[DemocratSinceBirth]

http://www.ncpp.org/2002SenGovPoll/2002ElectionPolls.html

[RaleighNCDUer]

1. I'm guessing Gallop, Galooph, whatever, before looking at the link.

[DemocratSinceBirth]

2. Nope... I Don't Even Think Gallup Was Included (nt)

[Dragon Turtle]

8. ? Gallup is listed between

Research 2000 and Quinnipiac College.

[demosincebirth]

3. Zogby?

[DCal]

4. Zogby

I remember he missed the Colorado Senate race by about 10 points.

[Gyre]

5. Too much work.

Why don't you read it and tell us the answer?

Gyre

[DemocratSinceBirth]

6. It's In The Link (nt)

[Tesibria]

7. Is Rasmussen good???

... i'm liking his numbers better than Zogby in the battleground states (most of them anyway)

[DemocratSinceBirth]

9. He Missed 00 By Nine Points But He Changed His Methodology....

Look at all the polls ....


Look at the average or median result...


Throw out the best and worst poll for your guy and look at those in the middle...

This is a tie race with an edge to Kerry....


We're doing fine....


Barring a major gaffe by Kerry he's the next pres....

Bank on it...

[NDFan]

10. Um

Zogby? He predicted 5 races wrong. He did polling on 17 races.

Quinnipiac, with 4 polls, predicted all races correctly.

Mason-Dixon had the most polls out, 23, and only got one wrong.

Gallup, with 7 polls, also got one wrong.

[Cha]

11. Zogby!

And they're being inaccurate, again?

[zoeb]

12. I guess I rely way to much on

Zogby! Gallup, one of the best??? I thought they tended to over sample Repukes. :silly:

[DemocratSinceBirth]

13. That's Why You Should Look At All The Polls...

Gallup's little model works best when Democratic voting is depressed....


Democrats "didn't show up" at the polls in 02 and that's why it was a bad year...

[zoeb]

14. Well, I don't think

Gallup should base this year's election on 02 turnouts... haven't they said (news programs) Democrat registration is up this year in many battleground states?

[DemocratSinceBirth]

15. Yep...

The Gallup model will not work this year because the Democrats are motivated...


The Gallup model only works when Democrats stay home...

[Awsi Dooger]

16. There were some very distinct trends in 2002

Almost all the key senate races were in red-leaning or even partisanship states. We were essentially trying to win road games. I have a PAN or Partisan Adjustment Number that indicated virtually every state with a major senate race was likely to understate GOP numbers in the state polls. DUers didn't like to read that, understandable since I was new here and had very few posts. I had one DUer PM me a month later, saying she briefly put me on ignore for a prediction that turned out to be correct:)

Zogby typically leans Dem so he was at a natural disadvantage in 2002, trying to force Dem wins in Republican states. I remember his final poll adjustments were giving Dem candidates a surreal chunk of the undecideds even in GOP states.

The 2002 gov races were uniquely understandable, if you looked at the decisive trend post-election. The state economies had tanked after Bush took over, and the incumbent party was tossed out if there was no incumbent governor to state his case. In something like 22 of 24 gov races minus an incumbent, the party in control was booted out. That accounted for weird takeovers like Dems in Oklahoma, Kansas, Tennessee and Wyoming, with GOP govs in Hawaii, Maryland and Vermont. The one we screwed up when that trend should have been in our favor was Massachusetts.

[pretzel4gore]

17. polling is political warfare

that's all. it's supposed to be 'informative' but the informers are also dedicated to seeing specific results, thus their polls are set up to provide those results, which then legitimize what comes out of elections, which are determined in part by the polls, which are organised by political gangsters, and reported, or not reported, by mediawhores, who want to see certain things, which they mention to the pollster/gangsters, who then run polls, to effect the very elections which inspire the poll in 1st place.