This is another post out there for all of you who still think, or at least worry, that Democrats are going to blow this election. Admittedly I am writing this at least in part to allay my own worries, because I admit that in my gut I still have a nagging sense that somehow Republicans will come back in this election. But what I want to point out is that our chances in this election are not like our chances in other recent elections. We are in a much, much better position, and we would have to make much bigger errors than we have made in the past to not pull this off.
I have offered up
warnings about the generic congressional ballot in the past, but at the very least it can give the sense of the general state of an election when compared to where the ballot stood in previous elections. First, here are the most recent polls from the six firms that have commissioned likely voter generic congressional ballot polls since Labor Day (
from Polling Report):
- Fox: Dems 49%, Reps 38% (9/27)
- Zogby: Dems 42%, Reps 33% (9/25)
- CNN: Dems 55%, Reps 42% (9/24)
- Gallup: Dems 48%, Reps 48% (9/17)
- AP: Dems 53%, Reps 39% (9/13)
- Harris: Dems 46%, Reps 35% (9/11)
These polls, the most recent from the is polling firms to publish generic ballots since Labor Day, show an average Democratic lead of 48.9%--39.3%. This is virtually identical to the numbers found among registered voters:
- Hotline: Dems 43%, Reps 33% (9/26)
- CNN: Dems 54%, Reps 41% (9/24)
- CBS: Dems 50%, Reps 35% (9/19)
- LA Times: Dems 49%, Reps 39% (9/19)
- Gallup: Dems 51%, Reps 42% (9/17)
- AP: Dems 51%, Reps 39% (9/13)
- Harris: Dems 45%, Reps 35% (9/11)
- Pew: Dems 50%, Reps 39% (9/10)
The last eight registered voter generic ballots show Democrats ahead by a mean of 49.1%--38.0%. Rasmussen has a generic ballot where Democrats lead 46%-38%, but I am not sure when it was conducted, so it will not be included here. What little difference there is between the likely voter polls and the registered voter polls is contained entirely within the Gallup poll of likely voters. Clearly, Gallup stands alone, even against itself.
So, Democrats have a double-digit lead across large numbers of polls. My point in this post is to show that we simply have not had leads like this in the generic ballot in any other recent election. Not even close. First, here is a handy table to compare where the generic ballot averages stood at the end of September over the past five campaigns. For the 2006 totals, I used the seven likely voter polls plus the three generic ballots that did not come with a likely voter model (LA Times, Hotline, and CBS):
End Of September generic ballot average, last five federal elections
2006: Dems 48.4%--Reps 38.2%. Democratic Margin: 10.2%
2004: Dems 45.3%--Reps 44.3%. Democratic Margin: 1.0%
2002: Dems 44.3%--Reps 42.6%. Democratic Margin: 1.7%
2000: Dems 44.5%--Reps 43.3%. Democratic Margin: 1.2%
1998: Dems 42.5%--Reps 46.0%. Democratic Margin: -3.5%
This is a whole different ballgame. This is nothing like the other recent elections. In the extended entry, I offer more detail on each of those years:
- In 2006, Democrats have never trailed, and have held double-digit leads outside of the margin of error every month from February forward.
- In 1998, the largest generic ballot lead for Democrats at any point in the campaign was five points, recorded in a single poll by the AP in late October. A few other polls had shown Democrats up by four points, but there was never a consistent Democratic lead of any amount at any point during the campaign. At this point in 1998, the previous six generic ballots actually showed an average Republican lead of 1.0%. The two generic ballots completed in late September of 1998 showed an average Republican lead of 3.5%
- In 2000, generic ballots consistently showed a small Democratic lead. However, the largest lead recorded for Democrats during that election was 7 points, and only in two polls (one in July and one in late October). Three other polls, including one by Democracy Corps, showed Democrats leading by either 5 or 6. The rest showed much smaller leads, and nothing ever approached the levels we see in our current election. The four generic ballots completed in September of 2000 showed an average Democratic lead of only 1.3%.
- In 2002, the generic ballot lead was clearly in favor of Republicans during the early part of the summer, but generally shifted back to Democrats for August and September. In September of 2002, nine polling firms conducted registered voter and /or likely voter generic ballots. By the end of September, the average Democratic lead across those nine polls was only 44.3%--42.6%. The best poll for Democrats during the election season showed them ahead by only seven points, and that poll was included in the average I described in the previous sentence. One Gallup poll in late October showed Democrats up 9 among registered voters, but that same poll showed them only up three among likely voters. Pretty much every poll during this cycle showed the race within the margin of error. Republicans often led during this cycle as well--certainly not something we see during this campaign season.
- In 2004, there were some polls favoring Democrats by large margins around the time of the Democratic convention, but they appeared in only three sources and never showed up after mid-August. At this point in September of 2004, the previous six polling firms to release generic ballots showed Democrats only leading 45.3%--44.3%. After Labor Day of 2004, it was just as common to see Republicans leading in the generic ballot as it was to see Democrats ahead.
This just is not like the previous four elections when it comes to Congress and the generic ballot. While Republicans have repeatedly shown the ability to close well, that has primarily been built upon superior money, media and voter targeting. While they still hold advantages in all three of those areas, in 2006 Democrats have significantly closed the financial gap, have a much larger network of progressive media than in the past, and have made improvements both in voter targeting and money spent on GOTV in congressional campaigns. In other words, we have more protection against Republicans closing well than we have had in the past, and they need to make up a lot, lot more ground.
This probably won't stop me from feeling worried, but I hope people aren't discouraged into inaction because of how other recent elections have ended up. Remember this data, and be convinced that Democrats have a very good chance to take back the House in 2006. Then get back to work making it happen.
