He does not change the Senate races, so much as identify which voters are not likely to support Democratic candidates downballot.
Now maybe you think that Clinton will actually get at or near a majority on election day, and if so, she’ll probably have coattails. But if she is near her husband’s 1992 performance, you can probably expect similar results to 1992 downballot.
Wang is completely mistaken:
http://www.realclearpolitics.com/epolls/other/2016_generic_congressional_vote-5279.html
Average is D+2, and historically Democrats need D+3 to break even in seats. To win the House they need a lot better than that.
Who says it’s over?
The first objection is that Wang’s retrofit did not predict future behavior. If 2012 behaved like 2000 through 2010 then the Democratic win of 1.2% would have been more likely to result in a Democratic house than a GOP, only a 16% chance of exceeding a GOP majority of 17 and hardly any chance of a GOP majority of 33. Yet the latter was what occurred.
So including 2012 adds data that makes the correlation much less reliable.
Mind you the magnitude of the seats loss wasn’t again so far off the line correlating with popular vote in 2014. The polls had been the miss that time, stating a 1 to 2 point D lead in the RCP average for late July and all of August and keeping it to 2 to 3 point advantage R in the week before. The actual result was R +5.7.
As to the generic issue … would you use a generic Senate candidate poll to predict how all contested Senate seats will go?
Put it together. The correlation turned out to not be as good as it had been with 2012 falling way outside. Polling of the generic House candidate is historically fairly volatile and not as good at predicting final outcomes as general election presidential polling is. And each race has its own dynamics just like the Senate races do. Understood that polling at district levels may not happen much but that does not mean that poorer data should be given great credence.
Sure there is still some correlation and I’d be willing to accept that an election day result consistent with the most recent McClatchy-Marist generic poll (D +8) were to occur that a House majority should follow. Maybe by much more than that correlation line would predict, or maybe it would be a coin flip one way or the other. It actually does come down to individual districts and how those votes get distributed.
Addie, RCP does not include the Marist poll for some reason. Go to HuffPo Pollster and you can find a running trendline margin of D +7.5. Not sure which to take too seriously. Wang uses HuffPo and his median approach to come up with D +5 and as it is in-between RCP’s number and HuffPo’s trendline it seems reasonable. Not so completely mistaken I don’t think. FWIW.
What the polls currently show is that most Congressional Republicans are running well AHEAD of Trump in their home states.
Even Kelly Ayotte, who may well lose, is not losing as badly as Trump.
Trump has no apparent positive OR negative coattails. The Democrats still have time to do a lot of negative ads trying to tie Republican Congressional candidates to Trump, but I don’t think that’s going to be devastating to most of those GOP candidates.
Look, John McCain is vulnerable, but how seriously can anyone in Arizona take the argument that he’s Trump’s pal?
It all depends on Republican turnout. Not so much voter preference in the Senate races. I can see a lot of Republicans wanting Senate candidates to win even though they dislike Donald.
The difference between midterm and presidential Republican turnout for elections is roughly a third from what I can tell. Getting numbers and not percentages is actually a pain in the ass or my google-fu is failing me this morning.
So typically, 1/3 of Republicans don’t vote when there isn’t a presidential candidate for which to vote. Within that third, we’re going to have some that don’t vote, some that vote for Donald or Hillary, and some that will just vote for the non-presidential races. If it were to split evenly among the 3 choices and if voting follows 2012, Hillary would win by roughly 9%. Of course, there a ton of caveats to that: Bernie supporters not voting Hillary, split-ticket voters, demographic changes, etc.
The Times today reported that a fair number of Congressional candidates will be issuing anti-Trump ads in September. Anti might mean they say they will stand up to him if he is elected, not that he is what he is.
I’m not surprised there isn’t that much of an effect yet, since it is early. Dems will hold their feet to the fire, and make it hard for them to be neutral. If they support Trump, they risk losing independents and perhaps much minority support. If they disavow him, they risk losing Trumpistas to not voting at all.
But it is a district by district issue.
I’d love to retake the house, but I doubt it because of the gerrymandering in place. So I’m realistic.
I’ve long since given up explaining to people who refuse to hear simple reality, but I’ll repeat w/out further comment: the GOP don’t control the house due to gerrymandering.
Now, the reality is in terms of this election, once the primaries are 100% over any fear of Paul Nehlen type rebels causing trouble will be gone and you probably will see some incumbents start running ads where they disavow Trump. How well that will work will be questionable. In my own backyard, just across the border in West Virginia, a lot of elected officials were conservative Democrats. They all ran against Obama, but have mostly lost in the past 4-6 years. It’s just not that easy to truly get separation from the top of the ticket, especially since ticket-splitting is at historic lows.
Additionally, at least from the early data we’re seeing at polls it looks like the “bad candidate” effect is hitting Republicans much harder than Democrats. Something like 82% of registered Republicans plan to vote for Trump (I’m in the 18% that aren’t), a number which essentially didn’t move after the convention, which is usually when a candidate seals up his party support. There are always some registered party members that vote cross ballot, but you usually are expected to get 90% or so of your own party. While 82% is still a lot, as close as Presidential elections have been this generation in terms of popular vote, only getting 82/100 vs Hillary currently sitting at 91-92/100 of her own party is a serious disadvantage.
Some of those Republicans will stay home. Some will vote for Johnson, and some (like me) will vote for Hillary Clinton despite some serious ideological differences. The effect will almost certainly be deleterious to the historically huge GOP house majority. But I think it’s big enough that an outright loss of it would be highly unlikely. It could get down into the single digits though. However if that’s the case the GOP is still likely to hold it until 2020, and then control will be decided based on whether Hillary wins reelection or not (I think she will win pretty easily in November, but also think for both structural and political reasons she’ll be the weakest Presidential incumbent since the last guy to actually lose reelection, Bush Sr.)
Part of the reason it’ll hold until 2020 is we’re almost guaranteed to win some seats back in 2018–the electorates in Presidential years are far different from the electorates in off-years. For whatever reason, progressive, young people, and other Democratic demographics just don’t like voting when the election isn’t flashy like it is in a quadrennial Presidential election year. The off year electorate is far older and more conservative, and the only times we’ve seen Democrats in recent memory make significant gains in a mid term have been years in which the GOP have essentially imploded. Example being 2006 when the country had reached its breaking point with George W. Bush’s Presidency.
Bull shit. It does not guarantee Republican control, but it sure helps.
Back up it with facts and I’ll listen.
Here’s my facts. Look at the Cook Partisan Voting Index by congressional district. The mean district is about D + 0.25 while the median district is R +2.
Well, yes, they do. Sure, without Gerrymandering they’d have a very good number of Reps. But I think they’d have lost “control” of the House.
Or you could just read Nate Silver and realize that Democrats are geographically concentrated, which means there are a lot of districts that vote 90% Democrat.
I mean, we could just end majority minority districts and have a 99% white House that would be more Democratic. If that’s what you prefer.
The Cook PVI and Nate Silver’s article say the same thing. The distribution of congressional seats is skewed to the Republicans.
It is, but not enough to ensure majorities. The level of gerrymandering is pretty normal.
No. Before 2010, there had never been an explicitly coordinated, computer-assisted plan to bring national–now “dark”–money to state leg. races, for the specific purpose of securing Republican majorities to control the redrawing of Congressional and statehouse districts after a census reapportionment.
The GOP currently controls the House because in 2014 they won the overall vote for House seats by 5.7%, not because of gerrymandering. Gerrymandering and other structural advantages made it larger but they earned control that time fair and square.
In 2012 gerrymandering was minimally part of what gave them a majority despite losing the popular vote by 1.2%.
In a moderately close overall Congressional election the GOP has a structural advantage. The question is how big that structural advantage is … do the Democrats need to get 53% to overcome it? That is very possible to hit. 55%? Less likely. And of course it depends on where that increase occurs.
addie, let’s assume you are correct that there are already lots of districts that vote 90% Democratic … well odds are then that any increase in D share are not going to come from padding those district margins any more, not much room to further pad. If there is an increase from an almost 6% national level loss to a win of the same magnitude where are those additional margins going to occur, particularly in this cycle?
Well the possible Trump House effect is based on his polling less strongly in many traditionally Red states and relatively very poorly with college educated Whites, actually losing that demographic which Romney for example carried by 12%. Some traditional GOP voters will stay home and not vote at all in some traditionally red states while a majority of college educated Whites will be clicking D … and many won’t ticket split. I put that together and see lots of suburban districts that had been red being winnable.
Of course there is also the theory that an anticipated lopsided win on the presidential level motivates voters to vote for the opposite party in Congress to function as a counterweight. Not sure I buy that though.
No idea why you think that your conclusion follows the first two paragraphs.
Ayotte and Hassan had been running pretty much tied. Ayotte had one poll before Trump’s cratering that put her up by 9 with the other results for two months alternating a couple of points one way or the other.
Trump craters, polling 17 down in NH, and Ayotte now polls down 10. Really hard not to see that as Trump’s collapse sucking her down with him. That’s pretty strong negative coattails to drop her that much that fast.
I have not heard anyone make that argument. In fact the negative impact is based more on the knowledge that McCain likely finds Trump to be repulsive … and yet supports his candidacy.
McCain’s brand is built on the perception that he has a spine, is tough, and is all about “straight talk” … endorsing Trump even as Trump continues to do his best to humiliate him … destroys that brand. And as pointed out already, Trump will get a fair number of traditional GOP voters to just stay home … and with McCain’s brand destroyed they won’t come out to just vote for him.
In Pennsylvania, Republicans get less than half the vote but win more than two-thirds of the US House seats. That’s “earned fair and square”?
Pennsylvania, outside of Philadelphia, is basically Alabama as they say.
The mean district in PA is D +1.4 the median district in PA is R +3.5. That’s worse than the country as a whole.
Creating majority minority districts does not require getting every single black person in the state drawn in to as few of districts as possible.
Look at this ridiculous map. This goes well beyond Dems tending to live higher population density areas than Republicans as that does not require districts with fractal borders.
Again, since apparently it was difficult to understand, the GOP won control of the House “fair and square” with a nearly 6% popular vote margin of victory. The exact score was exaggerated some by structural advantages including but not limited to gerrymandering, but they without question deserved control. The bigger blame for 2014 falls not on gerrymandering but shitty turnout on the Democratic side, especially of Millennials.
PA also helps illustrate why all structural advantages are not the result of gerrymandering (again though not disputing that such significantly exists) and addie point is correct this time: results show that Democrats in PA are highly concentrated in the urban districts, in partcular Philly and its suburbs, all of which went to Democratic Representatives odd shpaes or not. If the oddly shaped 17th district was created to concentrate huge numbers of Democrats there it didn’t work: the margin was solid but not outrageous, 57 to 43%. And the not oddly shaped 14th district, the only other district outside of the Philly area that also went D, is Pittsburgh. The other oddly shaped districts? Didn’t matter what shapes they were they still all went red and would have if you had cut them into squares instead.
This is a structural advantage bing illustrated yes but not mostly the result of gerrymandering.