GOP Surges to the Finish Line
As predicted, the Republicans are closing strong in the final hours, and suddenly my conservative Senate projections from earlier this month seem pretty likely to be accurate. It would seem that the worries of this blog's readers have reason to be concerned, particularly in the Senate.
It's always brutally difficult to guess how elections are gonna play out with a groundswell of inconsistent data and numerous headfakes orchestrated by both parties. I still believe the Democrats will win the House of Representatives even if the wind at their backs diminishes to a gentle breeze.
I'll say this about Mason-Dixon polls. They usually tilt Republican, and have artificially high numbers of undecideds that most other polls manage to put in one candidate's column or the other. In Minnesota, the Mason-Dixon poll is sponsored by the St. Paul Pioneer Press newspaper and they have always overstated GOP strength. In 2002, they predicted Coleman would beat Mondale by 5 on the Sunday before the election. Coleman won by 2. In 2004, they predicted George Bush would win by 1. John Kerry won by 3. With that in mind, I suspect some of the Mason-Dixon state polls in other states are also outliers. Chafee is not ahead in Rhode Island. The other polls are all reasonably believable, but seem to be a little more pessimistic than everybody else's polls, and all with significant numbers of undecideds that can typically be expected to trend to the challenger rather than the incumbent party (good news everywhere but Maryland).
The only race I didn't expect to be worrying about when I made my Senate predictions that I now am is Montana, where I definitley didn't see a Conrad Burns insurgency coming. Tester's trajectory is comparable to Max Cleland's 11th hour meltdown in 2002, but I'm not quite as pessimistic here for a number of reasons. Daily Kos indicates that Tester has a 21-point lead in early voting tallies, which are quite significant in numbers (as in nearly 20% of the overall vote), and Burns is still narrowly behind in every poll except this Mason-Dixon poll, below 50% in all of them. I'm losing faith in Tester's chances rather fast, but with predicting reputation on the line, I'm inclined to continue giving him the benefit of the doubt.
It's also hard to read whether the verdict in Saddam's trial will have any effect on the election. Considering the virtually everybody realizes this verdict will be appealed, is likely to inflame further sectarian violence, and won't get us any closer to getting out of Iraq than we were on Saturday, I'd say it's a wash. I put too much stock on the effect the bin Laden video would have on the 2004 outcome. I don't think it had much sway, and my gut says the Saddam verdict won't either.
As for today's ABC News/Washington Post poll suggesting the Dems generic ballot advantage shrunk from 14 points to 6 points in two weeks, their sample was disproportionately Republican according to Stu Rothenberg on "Face the Nation" this morning, making it a wash because the major media polls (Newsweek, CBS, AOL/AP) showing generic Democratic advantages of 15-20 points are disproportionately Democratic. The reality is likely to be somewhere between the two, but in all honestly, closer to the six-point margin than the 15-20 point margins.
I never put much faith in the media narratives about sweeping Democratic gains in the House and a Dem takeover of the Senate. The late-breaking momentum for the GOP is no surprise at all for me, but the bad news is that expectations have been raised so high at this point that the final outcome will now likely be spun as a Democratic defeat if the Dems win less than 30 House seats and fail to take back the Senate. Not that it'll stop the storyline manufacturers from similar irrational exuberance next time.
It's always brutally difficult to guess how elections are gonna play out with a groundswell of inconsistent data and numerous headfakes orchestrated by both parties. I still believe the Democrats will win the House of Representatives even if the wind at their backs diminishes to a gentle breeze.
I'll say this about Mason-Dixon polls. They usually tilt Republican, and have artificially high numbers of undecideds that most other polls manage to put in one candidate's column or the other. In Minnesota, the Mason-Dixon poll is sponsored by the St. Paul Pioneer Press newspaper and they have always overstated GOP strength. In 2002, they predicted Coleman would beat Mondale by 5 on the Sunday before the election. Coleman won by 2. In 2004, they predicted George Bush would win by 1. John Kerry won by 3. With that in mind, I suspect some of the Mason-Dixon state polls in other states are also outliers. Chafee is not ahead in Rhode Island. The other polls are all reasonably believable, but seem to be a little more pessimistic than everybody else's polls, and all with significant numbers of undecideds that can typically be expected to trend to the challenger rather than the incumbent party (good news everywhere but Maryland).
The only race I didn't expect to be worrying about when I made my Senate predictions that I now am is Montana, where I definitley didn't see a Conrad Burns insurgency coming. Tester's trajectory is comparable to Max Cleland's 11th hour meltdown in 2002, but I'm not quite as pessimistic here for a number of reasons. Daily Kos indicates that Tester has a 21-point lead in early voting tallies, which are quite significant in numbers (as in nearly 20% of the overall vote), and Burns is still narrowly behind in every poll except this Mason-Dixon poll, below 50% in all of them. I'm losing faith in Tester's chances rather fast, but with predicting reputation on the line, I'm inclined to continue giving him the benefit of the doubt.
It's also hard to read whether the verdict in Saddam's trial will have any effect on the election. Considering the virtually everybody realizes this verdict will be appealed, is likely to inflame further sectarian violence, and won't get us any closer to getting out of Iraq than we were on Saturday, I'd say it's a wash. I put too much stock on the effect the bin Laden video would have on the 2004 outcome. I don't think it had much sway, and my gut says the Saddam verdict won't either.
As for today's ABC News/Washington Post poll suggesting the Dems generic ballot advantage shrunk from 14 points to 6 points in two weeks, their sample was disproportionately Republican according to Stu Rothenberg on "Face the Nation" this morning, making it a wash because the major media polls (Newsweek, CBS, AOL/AP) showing generic Democratic advantages of 15-20 points are disproportionately Democratic. The reality is likely to be somewhere between the two, but in all honestly, closer to the six-point margin than the 15-20 point margins.
I never put much faith in the media narratives about sweeping Democratic gains in the House and a Dem takeover of the Senate. The late-breaking momentum for the GOP is no surprise at all for me, but the bad news is that expectations have been raised so high at this point that the final outcome will now likely be spun as a Democratic defeat if the Dems win less than 30 House seats and fail to take back the Senate. Not that it'll stop the storyline manufacturers from similar irrational exuberance next time.
posted by Mark at 5:57 PM
3 Comments:
Some new polls from USA Today that, for the most part, look better than Mason-Dixon's, but can they be believed?
Missouri--McCaskill 49, Talent 45
Montana--Tester 50, Burns 41
New Jersey--Menendez 50, Kean 40
Rhode Island--Whitehouse 48, Chafee 45
Tennessee--Corker 49, Ford 46
Virginia--Allen 49, Webb 46
sean, Gallup's generic ballot preference model has been all over the place in the past two months. In September, they had a 48-48 tie. Right after the Foley scandal (only like two weeks later), the Dems had a 23-point advantage, the highest of any pollster in the business. I'm still holding out for more numbers to confirm if the GOP's 11th hour resurgence is real, but it's starting to look it might be.
After thinking about the parallels with 1994 that we may see here, I also thought about the unexpected upset in the Georgia Gov race in 2002. I wonder which races (Governor, Senate, House) will see a Georgia Gov 2002-style outcome.
And I mentioned a while back but completely forgot in all the election madness that the Senate challenger I'm most like is Amy Klobuchar. I am one of the "quiet" liberals on the discussion board (Dave Leip's U.S. Election Atlas), and Amy is expected to be a "quiet" liberal senator, like Jack Reed.
Post a Comment
<< Home