Look at Gallup's final pre-election polls since 1994

[ProSense]

Republicans Appear Poised to Win Big on Tuesday










How on earth did Democrats win in 2006 and 2008?

[DFLforever]

1. That's a very good question.

Doesn't say much for their likely voter model does it?

[qazplm]

2. well I see gaps

bigger than we saw in 94, so not sure why you think this means Gallup's polling is somehow off.

[book_worm]

3. I think because in two big Dem years their model still shows that the GOP was more

engaged and enthusiastic--2006 and 2008.

[rury]

9. The graph shows Repukes leading us in

in the "absoultely certain to vote" and "paying a lot of attention" to the election in 2006 when we WON.
If their LV model is not "off" I don't know what "off" is.

[FBaggins]

4. The answer for 2006 is right there in your post and their historical data.

2008 isn't addressed (this just just midterm elections).

How did we win in 2006? We led the generic LV poll by seven points and only trailed by four in the "how much thought have you given" question. We also only trailed by two in the "certain to vote" category.

The best combination of these three stats in the four elections covered (not counting this year) provided the best results for democrats of the four elections. What part is confusing?

Oh wait... I know... "whatever", right? :-)

[mzmolly]

6. What?

:shrug:

Where did we lead the generic LV by seven points? Republicans are represented in dark green, "we're" the light green marker.

[FBaggins]

10. Neither chart provides the earlier years' generic ballot prediction.

In 2006 we led in Gallup's final LV generic ballot by 51-44%

http://www.gallup.com/poll/File/144131/Final_Midterm_Election_Estimate%20_Nov_1_2010.pdf

We won that year by about that same gap.

[Guggenheim]

16. That chart does not show any of Gallup's generic ballot final polls for any year

The chart in this thread simply shows the data for those "absolutely certain" to vote. That's only a sub-set of the generic poll.

The list of final pre-election generic-polls among likely voters list is here: http://www.gallup.com/poll/24493/election-polls-accuracy-record-midterm-congressional-elections.aspx

[ProSense]

22. Interesting.

2006 Nov 2-5

Likely voters
Democrat: 51
Republican: 44

Dems won 6 Senate seats.


This year:

Likely voters
Democrat: 40
Republican: 55

Why is the gap so wide this year?

The point of the OP charts, which is based on the final pre-election polls, is not to show likely voters, but the potential gap in turnout.

[Uzybone]

5. Even more important, why are GOP always more engaged than dems?

[The Hitman]

7. Fear.

[sohndrsmith]

8. Rachel hit it right on the head (according to the polls). Dems HAVE the votes

needed to win. If they decide not to bother wasting their precious time by casting those votes, they (we) will lose. The numbers are there, hopefully the votes matter to every single one of us on Tuesday. Per RM, our loss or win is very much up to us and, yes, IT IS just that simple.

If only we might keep sanity (or - at least partial sanity) alive for more than a day.

We'll find out Tuesday...

I'm not sure if it's valid or not, but I'm more concerned about getting Dem wins in the key governor races, but I have no solid data to back myself up. I heard someone mention how GOP wins could make 2012 supremely difficult. I don't want it to be any worse than all are predicting anyway...

[Cosmocat]

13. that is more breathless GOP/MSM circle jerk crape ...

about the governors and 2012 ... Them getting gleeful and ahead of themselves ...

Governors have a margin effect on presidential politics, especially NEW governors ...

The issue at hand with governors it districting for the congressional seats, and having the CEO who signs off on them ...

THAT is the prize with the governors racde ...

[DemocratSinceBirth]

11. That's Why You Have To Look At All The Polls And Not Cherry Pick

~

[JCMach1]

12. This looks very very bad actually...

[Guggenheim]

14. In 2006, Gallup predicted Democrats would win by 7%. They won by 8%. That's pretty accurate

http://www.gallup.com/poll/24493/election-polls-accuracy-record-midterm-congressional-elections.aspx

[FBaggins]

17. It isn't just 2006... they have a long track record with this poll.

Note that this isn't reporting the headline gap (what we commonly discuss), but the specific percentage of the 2-party vote that democrats will win.

The source here is RW, but I don't recognize any errors in the data. It's also important to note that this is just midterms. The generic ballot in presidential years is often well off.



On edit - Welcome to DU!

[Guggenheim]

18. That's correct. i think 15% will prove to be too much though

I wish I saw other pollsters with a similar prediction. I don't remember any others approaching 15%. We'll see.

[mzmolly]

19. What's important are the local individual polls. What does 8 percent

translate into, in terms of house seats?

[vaberella]

15. At this point...Don't trust polls, they are run by the corrupted MSM. n/t

[Avalux]

20. They are trying to suppress Dem votes, same meme CNN had this morning...

they stated that the GOP will definitely win the House and that the Senate is a strong possibility. The subsequent discussion looked at how the GOP could win the Senate, no mention of the House since they've already won it!

The MSM has been DICTATING to the American people how this election will turn out for about a year now; they must be right.

[marlakay]

21. Bush

and he has done a great job of hiding the past two years and a bunch of stupid americans who have super short memories have eaten up the baloney they are being fed daily...