For everyone looking at Donnelly and McCaskill (her luck ran out unfortunately) or Florida, I give you Oklahoma 5, New York 11, Virginia 2, Texas 32, Kansas Governor, etc. There were misses on both sides, which is what you should expect from a low-poll environment with random/natural polling errors. 2016 was a polling error where all the results tilted one way. 2012 was a polling error where all the results tilted the other way. This is how it should look.
It missed the fact that you can't underestimate racist southerners who can't wait to vote against the black men and women running for office. Honestly, if the media hadn't hyped up Gillum and Abrams, the whites wouldn't have been so triggered and turned up in huge repub numbers.
It wasn't just Florida. Democrats are on pace to win ~225-230 House seats. That means a pickup of between 30-35 seats. If we break those pickups down by region, we get: New England/Acela Corridor/Virginia: +2 PA, +3 NY, +1 ME, +3 NJ, +3 VA = +12 Rust Belt/Upper Midwest/Plains: +2 MI, +2 IL, +2 IA, +0/+1 net MN (depends on MN-01), KS +1, OK + 1 (!) = +8/9 Southwest/Pacific Coast: TX +2, AZ +1, CO +1, UT +1, CA +3, WA +1 = +10 (and could be another three in California) Confederacy states/Border States (minus Virginia and Texas): Ga +1, FL +2, SC +1 (!) = 4 Even if you throw Virginia's three in and Texas' two in, the region dramatically underperforms. McGrath lost in Kentucky. Democrats could have won another Georgia suburb. North Carolina's three competitive seats all broke the GOP way. FL-15 broke the wrong way. Ojeda in WV lost handily. The Democrats can't pick up a ton more seats in New England/Acela because they already have all but fifteen of them now. They can pick up more Midwestern seats, but there aren't a ton of them they could pick up. Between the Pacific coast and Southwest (excluding Texas), the GOP only has 26 seats left. The South, from North Carolina to Georgia to Florida to Tennessee, represents the biggest bloc of seats available, and Democrats continue to underperform. For whatever reason, Southern Republicans are highly motivated to turn out, and Southern Democrats continue to fare poorly. We may never be able to pinpoint the reason...
I was speaking more specifically about Fl/Ga in the governor races. I think up you would have seen less repub turnout if the Dems weren't black. Racism is real. Especially down here. Especially very liberal blacks. I am actually ever so slightly hopeful. The margins were narrower than I thought, tbh. I know the polls said different, but I know my state. I knew there was no way a very liberal black man was going to be 4 points ahead of the average empty headed white dude. So the fact that only about 80k votes out of over 7 million was the margin was actually pretty good. I also knew Rick Scott was more than likely to win. I don't think people realize just how much money he has behind him. Before I lost power, for a straight 6 weeks there were Scott ads constantly, and I mean constantly, on the tv. In a very red area. I didn't even see a Nelson ad until about a week before the election.
C'mon, did anyone really think a black woman would be elected governor of Georgia? Heitkamp and Tester were lucky to get elected last time. I'm disappointed by the results, but hardly surprised. As for polling error, a few upsets in both directions, which is what you would expect with about 500 races.
Pretty much. 2016 was odd in that the errors disproportionately swung elections in one way. Re: McCaskill. I'm getting a lot of post mortem chatter today here in STL. What we thought would happen on election day vs what actually happened: In our local wards, we saw an actual bump in voting vs. the 2016 general. Wards were coming in at 100% - 105% of their 2016 tallies. Further afield in the other reliably Dem strongholds and elsewhere in the city, we were looking at 85-95% from as it turns out was a not so random sample. I was getting the good news from different types of places, but not the underwhelming news from the same general areas. Still, overall from 4 counties (STL City, the county, Boone and Jackson) we wound up getting 90.3% of the votes at pt better than 2016, which featured optimal participation + a rock star type challenger in Kander. That part looked good. If you can get those types of participation levels in counties that are actually declining a bit in population, then you're doing about all you can ask for. Where things went wrong....everyone knew turnout would be higher elsewhere as well. 70% of the vote tally vs. a Presidential election is pretty standard elsewhere in the state. Most people were pegging it at 80% of the POTUS cycle and a couple points more GOP did in 2012. The rest of the state did 87.6% of their 2016 participation at 6.2 pts better for the GOP. And that was that. The polling error boiled down to underestimating the rural/fringe suburban vote in the LV models and the GOP margins in those areas where 4-5 pts too light.
I think this thread is premature. Let's at least wait until all the dust settles and every race in the country is called.
Boooo. So anyway, another piece of interesting data from the McCaskill-Hawley race. Hawley came out against three different ballot initiatives: medical marijuana, lobbyist/districting/clean up MO politics in general, and the minimum wage hike. All three won with 61-65% approval. McCaskill supported all three. In other words, there was a healthy subset of the electorate that agreed with McCaskill and disagreed with Hawley on all sorts of positions, but who voted for Hawley anyway. Polling going into the race used models that assumed the ballot initiatives would juice McCaskill turnout and that there would be less of this Hawley-initiative inconsistency. I don't know if this qualifies as polling error, voter error, or polling error of voting error, but this seems like as good a place to put this as any.
This is becoming axiomatic: conservative voters love liberal policies but hate liberals. So they can vote for medicare expansion and marijuana legalization and an increased minimum wage, but hate the people who make that their platform. You’ve got to remember that these are just simple farmers. These are people of the land. The common clay of the new West.
That cuts both ways. There are some conservative initiatives that pass fairly easily, like for example voter ID requirements. Another may be revoking sanctuary cities. These things are a lot more popular in the mainstream than liberals would like to admit. The liberal/conservative divide in America is increasingly cultural, and not so much policy based. And that cultural divide is increasingly manifested in the urban/rural divide. There are certain cultural gateway issues that automatically shut down the conversation for a lot of people. One of those for conservatives would be guns. Another may be abortion. If you don't pass the gateway, your other ideas never even get heard.
WVa and Arkansas also pass very strong anti-abortion propositions. As I understand WVa, it basically says that if Roe is overturned, abortion will be illegal in the state by the constitution.
Trump is president. Liberals understand very well that the common clay of the new west likes racist $&^%. The point is that the same people who like said racist $&^% also like themselves some super-liberal policies when it comes to health care, minimum wage & taxes. One does not obviate the other.
Good call. And a few other factors. Specifically, perhaps . . . It was enough for election night to briefly feel reminiscent of 2016, as polls underestimated Republicans in several key states and races. It raises questions about whether polls remain vulnerable to a 2016-like error in 2020, when the race promises to be tighter and focused on the kinds of predominantly white, working-class states where the polls underestimated Republicans. I'm curious what @American Brummie thinks, if he can get past the NYT paywall... https://www.nytimes.com/2018/11/21/upshot/polls-2018-midterms-accuracy.html
It's hard to say. A lot of the places Democrats underperformed (Ohio, Florida, Indiana, North Carolina) did so for idiosyncratic reasons. Ohio looks like it is becoming more conservative, whereas Florida is Trump's home state, and the economy in North Carolina is booming. If a recession hits in 2020, who knows which states abandon him. If pollsters adapt to account for their errors in those states, the polling may overshoot in the other direction. My guess is that the 2020 election will not be close enough to make this matter. Trump got 45.9% of the vote against the most-maligned Democrat in a generation. Republicans got 45% in 2018 too. That's their ceiling. Third party voting won't be large in 2020 - Libertarians will revert to 1/1.5%, Greens to <0.5%.
I think North Carolina was much more about lacking a glamorous, top-of-the-ticket race to drive overall turnout. Bet you anything it turns hard left in 2020.
This was very much how Johnson beat Feingold here in Wisconsin (well, that and Feingold didn't go out to the country to get support - but the ads were a major component).
Well, the hidden reason why Jones beat Pedophile 10 Commandment in GA is that the Dems had a network to organize and turn out voters. Abrams had developed a similar network in Georgia. The reason Abrams lost was not the racist voters, but Kemp very much stealing the election.
Not wrong, but what changed from 2016 to 2018. https://www.economist.com/graphic-detail/2019/01/05/the-failure-of-gerrymandering