D_Miranda_Wrights
Account Disabled
VinylBoy et al.,
Sorry -- Probably overprecise, but I'm confused about what inferences people are drawing based on the posters' histories here, so it's the only way I can be clear.
I'm going to back up here because I don't think we actually disagree. I'm going to explain why I posted what I did, and then state what I'm intending to say. I just came into this thread and read the dude's original post (I don't know the backstory here) and then this, from VinylBoy:
This was kind of a vague post -- I'm not sure if you mean to say Obama's approval ratings will approval as people make the comparison (maybe); or the other candidates' approval ratings will be worse than -18% (probably not); or that referenda elections tend to fail unless a strong alternative is presented (I disagree.) So, I just stated that there's a poor track record for winning re-election with under 40%. You guys then noted that approval ratings change over time, which is true, but doesn't conflict with anyone's post. You also note that Generic Republican tends to do better than most actual Republicans -- true throughout history, true this year, and immaterial to my point about anemic approval ratings resulting in incumbents losing because of referendum elections.
I don't think we're having a substantive disagreement, but I'll lay my cards out on the table:
1. Opponent approval rating does matter. If you look at the Nevada exit polls, the election became a referendum on both Reid and Angle, because both were unusually unpopular candidates.
2. HOWEVER, Reid benefited from netting the majority of ambivalent voters. Reid's approval rating was about 45% and not <40%, and Reid was going against an unusually unpopular opponent. I think, structurally, the national primary system encourages more electable candidates than a one-day state primary like Angle won.
3. As is obvious from the last two notes, there are lots of complex variables going on here, time being the biggest one. Obama is hardly hopeless. If his approval rating is <40% on Election Night, chances are he'll be screwed.
4. I agree with you about "Generic x" on average outpolling actual candidates; there being lots of time left; and weak (but not non-existent) correlation between approval ratings at this stage and election night.
5. If you're arguing that current approval ratings are a weak predictive mechanism of final results, you should do a statistical analysis instead of finding a few results to cherry-pick. That's like finding five methodologically perfect polls outside the MoE and then arguing that polling is an intrinsically poor predictive mechanism. But if you're trying to suggest it's a weak correlation, you're probably right, although your data are inadequate.
phew -- hope that brings us to the same page.
Sorry -- Probably overprecise, but I'm confused about what inferences people are drawing based on the posters' histories here, so it's the only way I can be clear.
I'm going to back up here because I don't think we actually disagree. I'm going to explain why I posted what I did, and then state what I'm intending to say. I just came into this thread and read the dude's original post (I don't know the backstory here) and then this, from VinylBoy:
He can be at 29% right now... but unless someone provides another real alternative he's still getting the votes of many, MANY people. You may want to focus on finding someone who can beat Obama and actually has a real plan for the economy & job creation, instead of constantly talking about how much you dislike him. Because we all know hate is in this season, and there's enough of it to go around on LPSG.
This was kind of a vague post -- I'm not sure if you mean to say Obama's approval ratings will approval as people make the comparison (maybe); or the other candidates' approval ratings will be worse than -18% (probably not); or that referenda elections tend to fail unless a strong alternative is presented (I disagree.) So, I just stated that there's a poor track record for winning re-election with under 40%. You guys then noted that approval ratings change over time, which is true, but doesn't conflict with anyone's post. You also note that Generic Republican tends to do better than most actual Republicans -- true throughout history, true this year, and immaterial to my point about anemic approval ratings resulting in incumbents losing because of referendum elections.
I don't think we're having a substantive disagreement, but I'll lay my cards out on the table:
1. Opponent approval rating does matter. If you look at the Nevada exit polls, the election became a referendum on both Reid and Angle, because both were unusually unpopular candidates.
2. HOWEVER, Reid benefited from netting the majority of ambivalent voters. Reid's approval rating was about 45% and not <40%, and Reid was going against an unusually unpopular opponent. I think, structurally, the national primary system encourages more electable candidates than a one-day state primary like Angle won.
3. As is obvious from the last two notes, there are lots of complex variables going on here, time being the biggest one. Obama is hardly hopeless. If his approval rating is <40% on Election Night, chances are he'll be screwed.
4. I agree with you about "Generic x" on average outpolling actual candidates; there being lots of time left; and weak (but not non-existent) correlation between approval ratings at this stage and election night.
5. If you're arguing that current approval ratings are a weak predictive mechanism of final results, you should do a statistical analysis instead of finding a few results to cherry-pick. That's like finding five methodologically perfect polls outside the MoE and then arguing that polling is an intrinsically poor predictive mechanism. But if you're trying to suggest it's a weak correlation, you're probably right, although your data are inadequate.
phew -- hope that brings us to the same page.
Last edited: