It has been clear for many months that the things the GOP planned to run on in the midterms—the tax cut, undercutting Obamacare, the economy, etc.—either were failing to motivate voters, or were actively hurting the Party. So, with Donald Trump taking the lead, they have gone with his bread and butter: anti-immigrant xenophobia. That was, after all, the issue that launched his presidential campaign. And as we hit the home stretch, the Party in general, and Trump in particular, have really hit the gas pedal on the nativism and the fearmongering.
One sign of the shift into high gear was the Willie Horton-style ad that Trump posted to his Twitter feed on Wednesday. That ad has been lambasted, not only for being racist, but also for peddling a lie. The man it focuses upon, Luis Bracamontes, entered the U.S. while Bill Clinton was president and...was deported. He then re-entered during the George W. Bush years, and committed the murders discussed in the ad. In short, there is zero basis for blaming any specific Democrat for Bracamontes' crimes, much less "Democrats" in general. And that is before we talk about the second ad Trump posted later in the day, which is getting less attention, but portrays the Republican Party as a force for stability and law and order and prosperity, while presenting the Democrats as a bunch of (mostly) dark-skinned hooligans determined to destroy America and commit random acts of violence:
#JOBSNOTMOBS! VOTE REPUBLICAN NOW!! pic.twitter.com/wso9ZHIvyF
— Donald J. Trump (@realDonaldTrump) October 31, 2018
It's called "Jobs not Mobs", though one is left to wonder if they only chose that title because "Birth of a Nation" was taken.
On Thursday afternoon, Trump delivered a speech full of juicy, red, anti-immigrant meat for the base (and, naturally, full of falsehoods and half-truths). He said that he is definitely going to take steps to restrict asylum-seekers, and that he will order the U.S. army to open fire on any members of the immigrant caravan who throw rocks. Trump also spoke approvingly of the plan announced by Sen. Lindsey Graham (R-SC) to introduce legislation that will end birthright citizenship. Congress doesn't have the power to rewrite the Constitution without amending it any more than Trump does, but when you're auditioning for attorney general, you ignore trivial details like that. In any case, of all the provocative things the President said, the one that is getting the most attention is his announcement that to "protect" Americans from the immigrant caravan, he will probably increase the troop deployment to 15,000. Let's put that in a little context:
Situation | No. of People | Year |
U.S. Marines who stormed the "shores of Tripoli" in the Second Barbary War | 2 | 1815 |
British soldiers in the Charge of the Light Brigade | 670 | 1854 |
American troops at the Battle of the Little Bighorn | 700 | 1876 |
Number of Tuskegee Airmen | 932 | 1941-45 |
Number of soldiers in Theodore Roosevelt's "Rough Riders" | 1,166 | 1898 |
People in the immigrant caravan | 3,500 | 2018 |
Troops currently deployed in Iraq | 5,540 | 2018 |
Size of the U.S. Navy during the War of 1812 | 8,000 | 1812-14 |
Troops currently deployed in Afghanistan (an actual war zone) | 9,294 | 2018 |
Size of George Washington's Continental Army, at its height | 11,000 | 1779 |
Size of the entire Nicaraguan army | 12,000 | 2018 |
Trump's planned deployment | 15,000 | 2018 |
Union troops that participated in the First Battle of Bull Run | 18,000 | 1861 |
Number of Allied troops in the first wave on D-Day | 24,000 | 1944 |
Size of the entire U.S. Army during the Mexican-American War | 60,308 | 1846-48 |
In short, if Trump is planning to wage a good-sized war against seasoned troops, then 15,000 is a fair number. Against 3,500 unarmed refugees, many of them children, it's a tad bit of an overreaction.
The President did not call it a day after his speech, though. He had a rally on Thursday night, as he always does these days, this one in Missouri. There, he repeated most of his anti-immigrant talking points from earlier in the day, and then raised a few eyebrows when he declared that, "We did have two maniacs stop a momentum that was incredible, because for seven days nobody talked about the elections. It stopped a tremendous momentum." There is a bit of truth here, as it appears the pipe bombs/shootings have moved things 1-2 points in the direction of the Democrats. However, Trump is unquestionably the only president crass enough to put things in those terms. Not surprising, though, as it gives him yet another direction to shift the blame if the midterms go poorly.
Ultimately, there is no question that fearmongering is an effective political tactic, though it is one with a limited shelf life. Nativists, xenophobes, red baiters, and the like often rise to power, but they generally don't keep power for long. And so, while Trump may be deploying the "best" available strategy for the GOP, such as it is, the question is whether it will continue to be a successful strategy. And a pair of studies released earlier this week suggest that it will not be. While the Republicans are likely to do fine, Senate-wise, that is primarily a product of structural factors that break in their favor. As to the House, however, the two studies show that the "battleground" House districts are largely prosperous, and are not being negatively affected by globalization (or immigration). So, it is probable that Trump's words will fall on deaf ears in those places, and that nothing he does this week will help save the lower chamber for the GOP. (Z)
The Washington Post and Schar School of George Mason University conducted a poll Oct. 25-28 in 69 House districts considered to be battlegrounds. The poll finds that 50% of the voters support the Democrat running in their district and 46% support the Republican. While this is a small edge for the Democrats, in 2016, the Republicans had a 15-point margin in these districts, so the new poll represents a 19-point shift toward the Democrats. Also important is that 63 of the districts are held by Republicans and only 6 are held by Democrats, so if both groups split evenly, the Democrats will pick up 31 or 32 Republican seats and the Republicans will pick up 3 Democratic seats. Such an outcome would flip the House to the Democrats. Historically, close races don't split evenly, though. Usually, one party wins the lion's share.
When looking at the crosstabs, one thing jumps out, as shown here:
As you can see, Republicans are winning with white male and female non-college voters along with white college men, but they are getting crushed with white college women and non-whites. Thus, to a large extent, the results of the election depend on whether large numbers of white college-educated women and non-whites show up to vote. The former group generally is pretty good about voting but the latter one is not so good.
The poll also showed that among 18-39 year-olds, the Democrats lead 58% to 37%, while older voters split evenly. Younger voters also have a pretty miserable turnout record. So the Democrats' hopes are largely pinned to one group (college women) with a good turnout record and two groups (nonwhites and younger voters) with crummy turnout records (though see below). (V)
Although younger voters have historically had a poor turnout record, there are some promising signs on that front for the Democrats this year. The percentage of early voters under 30 is double what it was in 2014 in Georgia and Texas. Whether this is a harbinger of increased voting by younger voters or merely earlier voting isn't 100% clear, of course. It is entirely possible that the 20% of millennials who vote are so enthusiastic that they couldn't wait and did it already, but the 80% who usually don't vote are still not going to vote.
With that said, while we don't know for certain what's happening in Georgia and Texas, there is some evidence that young voters' enthusiasm is indeed way up this year. A new poll from the Harvard Institute of Politics reveals that 40% of voters between the ages of 18 and 29 say they will "definitely vote" in the midterms. That would be a staggeringly high number, if it comes to pass. In the last 30 years, that group has only broken 20% in a midterm election once: 1994 (21%). And given that Donald Trump's approval rating among young voters is an anemic 26%, it will not be good news for the GOP if there is indeed record-level turnout among young people.
Similarly, there has also been a spike in early voting overall this year. Here are the numbers from seven key states, with the total number of ballots so far, as well as at the same point in 2016 and 2014:
State | 2018 | 2016 | 2014 |
Arizona | 1,117,151 | 1,527,144 | 736,790 |
Florida | 3,345,330 | 5,028,770 | 2,375,144 |
Georgia | 1,506,311 | 1,776.883 | 627,805 |
Montana | 242,103 | 306,110 | 20,786 |
Nevada | 318,174 | 517,030 | 162,946 |
Tennessee | 1,141,857 | 1,321,741 | 234,958 |
Texas | 3,064,679 | 3,165,203 | 982,346 |
We don't know enough to say that this favors the blue team for certain. However, it is clear that turnout this year is far ahead of the last midterm in 2014, and is looking more like the presidential year of 2016. In general, Democrats do much better in presidential years than in non-presidential years. (Z & V)
Consultants in both parties believe that white college-educated women will be the decisive voting bloc next week. They also believe that national security issues, including Donald Trump's kowtowing to Vladimir Putin, the meeting with Kim Jong-Un, and Trump's refusal to accept that the Saudi Royal Family murdered a journalist in cold blood, won't have any effect on the elections. Furthermore, they agree on what the known unknowns are, they just don't know how important each one will be to driving turnout. And of course, there could yet be unknown unknowns. Here, however, are the known unknowns:
In retrospect, we may learn which of these was most important, and which voters were most moved by them. Until Wednesday or later, when the exit polls have been analyzed, these will remain the known unknowns. And again, there could be unknown unknowns as well, things no one now sees as a big deal, but which a lot of voters based their votes on. (V)
The Rust Belt looks brutal for Republicans. Democrats are almost certain to win Senate races in Michigan, Ohio, Wisconsin, and Pennsylvania, all states that Donald Trump won in 2016. Democrats are also very likely to win the governorships of Michigan, Illinois, and Pennsylvania, and have a good chance to win them in Ohio and Wisconsin as well. How come one of Trump's strongest regions is going to be a disaster area for the GOP?
One study shows that the counties that flipped from Obama to Trump are the counties with the most unhappy voters. In other words, the guy who said "the American dream is dead" did best in counties where the American dream really is dead. The people there voted for Trump not because they liked his policies on anything, but because they were fed up with the ways things were and wanted change. In fact, their desire for change is why they voted for Obama in the first place.
The counties that swung hardest to Trump are hotbeds of alienation, dissatisfaction, and opioid use. They have no sense of community, like strong active churches, youth sports leagues, PTAs, etc. If this analysis is correct, these people are not fundamentally Republicans, they are just angry. Many are surely disappointed that Trump hasn't made their lives any better, and may show this by voting for Democrats. Of course, if the Democrats take control (in 2020) and their lives don't improve, then they will take it out on the Democrats. Unless their lives get better, they are going to continue to blame whichever party is in charge indefinitely. (V)
The North Dakota legislature passed a law requiring voters to show up with ID giving their street address. This is a huge barrier for the thousands of Native Americans in the state who live on reservations that do not have street addresses, so they sued to have the law overturned. Yesterday, federal judge Daniel Hovland, a George W. Bush appointee, ruled that it was too late to change the rules a few days before an election.
The decision may seal the doom for Sen. Heidi Heitkamp (D-ND), who was depending on a heavy turnout of Native Americans, most of whom are Democrats. Tribal leaders are scambling to create streets on their reservations and to produce documents giving members IDs with the new streets, but it is doubtful that they will succeed enough to have Native American turnout match that of 2012, when Heitkamp narrowly carried the state. (V)
The Senate Leadership Fund, a super PAC closely aligned with (= de facto controlled by) Senate Majority Leader Mitch McConnell (R-KY), has pulled all its television advertising for Patrick Morrisey (R) in the West Virginia Senate race. That almost certainly means that McConnell is convinced that Sen. Joe Manchin (D-WV) will be reelected, so spending money to prop up Morrisey is a waste. The money is going to races in Florida, Indiana, Montana, Missouri, and Tennessee.
Donald Trump apparently didn't get the memo, so he is going to West Virginia to campaign with Morrisey today. Beating Manchin was always something of a long shot because he is a popular former governor who sides with Trump on environmental issues, especially coal, and voted for both of Trump's Supreme Court nominees. Manchin clearly understands what it takes for a Democrat to win in West Virginia (be conservative on cultural issues, pro-union and populist on economic issues). Also a factor is that Manchin's family has been in West Virginia for generations and the senator was born and raised there, went to college there, and has never lived anywhere else. Morrisey is from New Jersey and once ran (unsuccessfully) for Congress in House district NJ-07. He moved to West Virginia in 2012, so he is basically a carpetbagger. West Virginians notice things like that.
It is not unusual for parties to triage candidates in the final days of an election as their internal polls tell them who is sure to win, who is sure to lose, and who needs help right now. For example, the Democrats just made an unexpected move, pouring $700,000 into the special election in Minnesota, in which appointed Sen. Tina Smith (DFL-MN) is not doing as well as expected against Karin Housley (R). (V)
We have half a dozen polls from Harris Interactive, a British firm, today. Their results seem mostly in line with other pollsters, except putting Rep. Martha McSally (R-AZ) ahead of Rep. Kyrsten Sinema (D-AZ) and giving Sen. Jon Tester (D-MT) a much bigger lead in Montana than anyone else. Those kind of balance out, so there is no reason to exclude them based on fishy numbers. There are always statistical fluctuations.
In our view, the races to watch are Nevada, Arizona, Missouri, Indiana, and Florida. If the polls are accurate, the Republicans will continue to hold the Senate and maybe pick up a seat or two. (V)
State | Democrat | D % | Republican | R % | Start | End | Pollster |
Arizona | Kyrsten Sinema | 43% | Martha McSally | 48% | Oct 25 | Oct 31 | Harris Interactive |
Florida | Bill Nelson* | 46% | Rick Scott | 47% | Oct 25 | Oct 31 | Harris Interactive |
Florida | Bill Nelson* | 49% | Rick Scott | 47% | Oct 24 | Oct 29 | SSRS |
Florida | Bill Nelson* | 49% | Rick Scott | 47% | Oct 30 | Oct 31 | St. Pete Polls |
Indiana | Joe Donnelly* | 42% | Mike Braun | 42% | Oct 25 | Oct 31 | Harris Interactive |
Missouri | Claire McCaskill* | 46% | Josh Hawley | 44% | Oct 25 | Oct 31 | Harris Interactive |
Montana | Jon Tester* | 48% | Matt Rosendale | 41% | Oct 25 | Oct 31 | Harris Interactive |
Nevada | Jacky Rosen | 46% | Dean Heller* | 43% | Oct 25 | Oct 31 | Harris Interactive |
Ohio | Sherrod Brown* | 52% | Jim Renacci | 42% | Oct 30 | Oct 31 | Cygnal |
Pennsylvania | Bob Casey* | 50% | Lou Barletta | 35% | Oct 22 | Oct 28 | Franklin+Marshall Coll. |
Tennessee | Phil Bredesen | 44% | Marsha Blackburn | 52% | Oct 28 | Oct 30 | Emerson Coll. |
Tennessee | Phil Bredesen | 45% | Marsha Blackburn | 49% | Oct 24 | Oct 29 | SSRS |
Texas | Beto O`Rourke | 47% | Ted Cruz* | 50% | Oct 28 | Oct 30 | Emerson Coll. |
West Virginia | Joe Manchin* | 45% | Patrick Morrisey | 40% | Oct 19 | Oct 30 | Research America Inc. |
West Virginia | Joe Manchin* | 47% | Patrick Morrisey | 42% | Oct 28 | Oct 31 | Emerson Coll. |