Facebook
Twitter
Subscribe to RSS
A Labor Day Status Report
Plus, updates from Tuesday night
, August 30th, 2018
We’ve been starting Crystal Ball pieces with a few “key points” summing up the article. As we head into Labor Day weekend and the start of the sprint to Election Day, we thought we’d do something different. Instead of key points from this article, here are some key points about this election so far:
Pluses for Republicans: The economy is good and we’re not in the midst of an unpopular foreign war, two sometimes-predictors of poor midterm performance for the White House party. The map benefits Republicans in both the battle for the House and the Senate: The median House seat by presidential performance (NE-2) voted for President Donald Trump by two percentage points, while Hillary Clinton won the national popular vote by two, making the median House seat about four points to the right of the nation, which is a good shorthand for the generic GOP edge in the House. In the Senate, Democrats are defending 26 of the 35 seats being contested, one of the worst maps any party has had to defend in a midterm. The president’s approval rating, while poor, has not gotten worse in recent months; the House generic ballot, generally showing a Democratic lead of between six-to-eight points, is bad enough to indicate Republicans could lose the House, but not bad enough for that to feel like a certain outcome. Several Democrat-held Senate seats in red states remain very much in play, and Republicans are still very much in the game in Arizona and Nevada, their two hardest Senate seats to defend.
Pluses for Democrats: Trump’s approval rating is perpetually low, and more than the whiff of scandal — one other factor that has hurt the presidential party in past midterms — helps keep it low. Additionally, strong disapproval of the president is routinely much higher than strong approval of the president in polls. The presidential out party usually has an advantage in midterms. The House generic ballot shows a Democratic lead of sufficient size to allow for a House takeover. None of Democrats’ red state Senate incumbents appear to be certain or near-certain losers, although several are endangered. They have a huge number of credible candidates running credible campaigns across the House landscape. Polling generally shows that Democrats are more excited about the election than Republicans, a statistic that is backed up by the bulk of elections conducted since the 2016 election, where Democrats have often run ahead of what one might expect based on recent performance. Republicans are defending far more open House seats than Democrats (42 GOP seats will not have an incumbent on the ballot this fall, compared to just 22 for Democrats). These seats are generally easier to flip. The gubernatorial map favors Democrats, as they are defending only nine of the 36 seats in play, and many of their best targets are open seats.
The unknowns: Future developments in Robert Mueller’s investigation; the impact of tariffs and trade policy, particularly in agriculture-heavy states that otherwise might lean Republican; campaign developments, such as debate mistakes or new opposition research (a big unknown for many Democratic candidates, a good number of whom have not run before); events for which we cannot account even now.
Our best guess in the House right now: Democrats are soft favorites to capture the House majority, but there is a wide range of possible outcomes including both the GOP retaining control with a narrow majority as well as the Democrats winning significantly more than the 23 seats they need to gain control.
Our best guess in the Senate right now: Republicans remain clear favorites to retain the Senate majority, and they continue to have a path to start next year with more seats than they hold now (51, once the late John McCain is replaced by a Republican appointee). Democrats do have a path to win the Senate majority, particularly if they can eventually pull off upsets in a state or two where we currently favor Republicans, like Tennessee or Texas.
Our best guess in the gubernatorial races right now: Democrats will control more governorships at the start of next year than they control now (only 16), and probably more than just a couple more.
The picture should get clearer as we turn the calendar to September. Keep tabs on the race at the Crystal Ball and also at the new Political Atlas we’ve just launched in partnership with the international polling and research firm Ipsos.
KEY POINTS FROM TUESDAY’S PRIMARIES
— The so-called party establishment in both parties struck out in the Florida gubernatorial primary, setting up a battle of strong contrasts in the nation’s most competitive megastate.
— The Oklahoma governor’s race is a sleeper for the fall after businessman Kevin Stitt (R) captured the GOP nomination.
— Rep. Martha McSally (R, AZ-2) got through the Arizona Senate primary comfortably, setting up a very competitive battle with Rep. Kyrsten Sinema (D, AZ-9).
— Unrelated to Tuesday’s primaries, now that Rep.-elect Troy Balderson (R, OH-12) has officially won his hotly contested special election, he enters the fall as a small favorite.
Crystal Ball gubernatorial ratings change
Crystal Ball House ratings change
In Florida, both parties picked gubernatorial nominees that would have seemed very surprising a year ago, but only one picked a nominee whose victory would’ve been surprising a week ago. Republicans selected Rep. Ron DeSantis (R, FL-6), who rode support from President Trump to a big win over one-time clear frontrunner Adam Putnam (R), the state agriculture commissioner and a former U.S. House member. Nothing about DeSantis’ edge was surprising, and his win gives the president another primary victory (Trump’s record is not perfect in primaries, but it’s not far off). Meanwhile, Democrats eschewed seeming favorite Gwen Graham (D), a former House member herself and daughter of former Florida governor and senator Bob Graham (D), in favor of Andrew Gillum (D), the African-American mayor of Tallahassee who ran to Graham’s left with the backing of Sen. Bernie Sanders (I-VT).
Basically, what happened in Florida is that the political professionals who run both party campaigns did NOT get what they wanted there on Tuesday night. Democratic operatives preferred Graham, and non-Trump Republicans generally preferred Putnam even though they were resigned to DeSantis winning long ago. This sets up a real choice for voters in the fall — a Sanders progressive versus a Trump conservative. The stakes are high: Democrats have not won the Florida governorship since 1994, an embarrassing dry spell in a state that leans right, but not as strongly right as Republicans’ dominance of state government in recent years would indicate. DeSantis is as about as closely tied to Trump as one can be, a dangerous place to be in a swingy state (although Trump’s approval generally seems to be a little bit better in Florida than it is nationally). Term-limited Gov. Rick Scott (R-FL) barely won in 2010 and 2014, two good years for Republicans; the national environment will be better for Democrats this year. DeSantis did not kickoff his general election campaign with flying colors, telling Fox News on Wednesday that Florida shouldn’t “monkey this up” by supporting Gillum. Democrats immediately attacked DeSantis for making such a statement in reference to a black candidate. Meanwhile, Gillum too will have to pivot to facing a general electorate, and an FBI investigation of Tallahassee city government still hangs over his campaign, making Gillum a potentially very risky choice for Democrats.
Had this been Graham versus DeSantis, it might have been tempting to move this race to Leans Democratic. Had it been Gillum versus Putnam, it might have been tempting to move it to Leans Republican. But now the race remains firmly in the Toss-up category, along with the state’s Senate clash between Sen. Bill Nelson (D-FL) and Scott.
In Oklahoma, businessman Kevin Stitt (R) defeated former Oklahoma City Mayor Mick Cornett (R) 55%-45% in the Sooner State’s Republican primary runoff for governor. Stitt will now face former four-term state Attorney General Drew Edmondson (D) in the general election. There is little to go on polling-wise, but a mid-July SoonerPoll found Edmondson and Stitt neck and neck, 40%-39% in favor of Edmondson. Notably, the now-eliminated Cornett led Edmondson 43%-35% in the same poll. It’s just one survey, but the long-time mayor of the state’s biggest city seemed like a safer bet for the GOP as it attempts to retain the Oklahoma governorship. Although the state has a strong Republican lean, outgoing Gov. Mary Fallin (R) is one of the most unpopular governors in the country — Morning Consult’s second quarter data for 2018 pegged her with the highest disapproval (74%) among all 50 governors — and the state has seen a number of favorable special election results for Democrats in the past year and a half. Edmondson certainly has a tough path to victory in this conservative state, but Stitt’s nomination on the GOP side probably cracked the door open a bit wider for Democrats, so we are moving the Oklahoma gubernatorial rating from Likely Republican to Leans Republican.
Further west, in Arizona, Rep. Martha McSally (R, AZ-2) ended up winning a decisive victory in her primary against two fringe right-wing candidates, setting up a marquee open seat Senate race with Rep. Kyrsten Sinema (D, AZ-9). This is a must-win race for Democrats if they want to win a Senate majority; in fact, if they don’t win Arizona, they may end up losing ground in the Senate given the other states they have to defend. Down the ballot, both national parties breathed a sigh of relief about the primary outcomes in AZ-2 as ex-Rep. Ann Kirkpatrick (D, AZ-1, now running in AZ-2) and Tucson Hispanic Chamber of Commerce CEO Lea Márquez Peterson (R) won their respective nominations. But neither had an easy time: Kirkpatrick won by about 11 points over 2016 AZ-2 nominee Matt Heinz (D), and had to work to do that, while Márquez Peterson only won by about 4.5 points against no-name opposition. Kirkpatrick enters the fall as a favorite in an open, Clinton-won seat that also is something of a must-win for Democrats (probably no individual House seat is actually must-win for either side, but if Democrats can’t win this seat it would lead one to question how well they are doing elsewhere).
Finally, Rep.-elect Troy Balderson (R, OH-12) officially won his race against Franklin County (Columbus) Recorder Danny O’Connor (D), besting O’Connor by about a percentage point after all the votes were counted late last week. Turnout, while good for a summer special election, should be higher in the fall, which might mitigate the Democratic enthusiasm advantage on display earlier this month and help Balderson, who now possesses a (very weak) form of incumbency. We’re moving the race from Toss-up to Leans Republican for the fall — it should be close again, but if Balderson could win in August he’s probably better than 50-50 to win in November as well.
Forecasting the 2018 Midterm Election Using National Polls and District Information
, August 30th, 2018
| Editor’s Note: In advance of the 2016 election, the Crystal Ball featured several political science forecasts of the presidential race. Released several months in advance of the election, these models included variables such as the incumbent’s approval rating, the economy, and other “fundamental” factors, and they generally did a good job of projecting a very close presidential race. There also are political science models that try to project the midterm outcome for the U.S. House of Representatives, including the one featured below by political scientists Joseph Bafumi, Robert S. Erikson, and Christopher Wlezien. Crystal Ball Senior Columnist Alan Abramowitz released his model last year in the Crystal Ball, with the only variable unknown back then the House generic ballot average. As of Wednesday afternoon, Democrats held leads of six-to-nine percentage points in the polling averages, which translate in Abramowitz’s model to around a 30-seat expected Democratic gain. That projection is very much in line with the Bafumi-Erikson-Wlezien model described below. Both models predict Democratic House gains that, if realized, would give them the majority, but only a small one: Democrats need to net 23 seats to win the House, and there is enough uncertainty in these models that Republicans holding the House is hardly out of the question. In other words, these models track fairly closely with the Crystal Ball’s assessment of the race for the House: Democrats as modest favorites but with Republicans capable of holding on to the majority. This is a summary of an article with the same title that is forthcoming in PS: Political Science and Politics. We’ll be featuring a couple of other models in the coming weeks.
— The Editors |
The Bafumi-Erikson-Wlezien team first applied our model in 2006 and were among the first to predict the return of the Democratic majority. That forecast was off by only two seats. Like most others, our model predicted a Republican takeover in 2010, underestimating the GOP swing by about 10 seats. In 2014 we overshot the Republican surge by just one seat. We now repeat our midterm forecasting exercise for 2018.
Based on information gathered in June, our forecast for 2018 is that the Democrats will gain seats in the House and most likely enough to retake control. The average of our simulations implies that they will win 221 seats to the Republicans’ 214, a slim majority. The predicted seat division may surprise given that we expect the Democrats to win the national vote by a substantial margin (53.6%-46.4% as a two-party share). The translation of votes into seats advantages the Republicans. This largely reflects gerrymandering, some of which is “natural” from the way Democrats and Republicans cluster geographically, and some of which results from redistricting following the 2010 census. There is a related incumbency advantage, though this has shrunk with the growing polarization, which nevertheless makes it difficult for the Democrats to make inroads in Republican-leaning districts.
Our forecast has two parts. First, using past elections as a guide, we model the national vote as a function of the generic ballot polls four-to-five months before the election, plus which party holds the presidency. Even early in the campaign year, generic ballot polls predict well once the president’s party is taken into account. Historically, the trend toward Election Day is for the out-party to make gains. Thus, we expect the Democrats to gain in the polls as the election approaches.
Second, using the most recent election (2016) as a guide, we model how the vote in individual seats varies as a function of the vote for the House and the president in the most recent election. Open seats and incumbent-contested seats are modeled separately. Using the 2016 equations, we substitute the expected national vote in 2018 for the national 2016 vote into the model.
Our simulations incorporate the uncertainty from our modeling. For each of our 3,000 simulations, we draw the national vote taking into account the error variance from our vote model. For each simulation, our congressional district predictions take into account the uncertainty from our modeling of the vote at the district level when the national vote is known. Our seat predictions are represented in Figure 1. There we can see slightly more blue than red, which means that the Democrats are more likely to win control of the U.S. House. Specifically, they have a 54% chance of regaining the majority. A Republican win is only a little less likely (46%) but notice that there is the possibility of a big Democratic win.
The forecast requires an important note of caution. It assumes in effect that parties apply their normal level of effort across all districts. The reports of Democratic enthusiasm about newly winnable races suggest that this assumption may not be correct. For that reason the forecast here may be best considered a lower bound of Democratic prospects.
Figure 1: Bafumi-Erikson-Wlezien House model simulations
| Joseph Bafumi is an associate professor of government at Dartmouth College. He conducts research in American politics and has published in numerous scholarly journals including the American Political Science Review, Journal of Politics, Political Analysis and Quarterly Journal of Political Science. |
| Robert S. Erikson is Professor of Political Science at Columbia University. He is co-author of The Timeline of Presidential Elections (University of Chicago Press), The Macro Polity (Cambridge University Press), and American Public Opinion (Pearson). |
| Christopher Wlezien is Hogg Professor of Government at the University of Texas at Austin. He is co-author of Degrees of Democracy and The Timeline of Presidential Elections. His current research on U.S. elections examines media coverage and electoral preferences and shows that the former can both lead and follow the latter. |
Center for Politics, Ipsos Public Affairs Release New Midterm Election Forecasting Tools
New Political Atlas features expert race ratings, poll-based forecasts, and social media trends in advance of the 2018 election
August 28th, 2018
(WASHINGTON, DC) — The University of Virginia Center for Politics and Ipsos today launched the UVA Center for Politics-Ipsos Political Atlas. The new site combines Sabato’s Crystal Ball race ratings for every House, Senate, and gubernatorial race; Ipsos’ poll-based modeling; and Ipsos’ tracking of social media trends.
Ipsos and the Center for Politics unveiled the Political Atlas, www.political-atlas.com, at an event at the National Press Club on Tuesday morning.
The idea behind the Political Atlas is that election projection is done best by offering readers several different sources of information, which the Atlas provides through the Crystal Ball’s qualitative race rating assessments and the Ipsos’ poll-based model and social media tracking. The three methods provide separate race ratings (Toss-up, as well as Leans, Likely, and Safe Democratic or Republican) to assess the likelihood of either side winning a given race.
“The best practice in forecasting understands that no single approach leads to knowing the future. The most robust predictions use multiple, independent indicators and looks for common conclusions or divergent directions,” said Clifford Young, president of Ipsos Public Affairs.
The Political Atlas will include:
- Daily updates of the main issues affecting citizens in all 50 states.
- Polling and social media indicators for every congressional, senate, and gubernatorial race where available, as well as expert assessments.
“After the 2016 election, we strove to try to find new data sources and perspectives in political forecasting and to present these to Crystal Ball readers and the general public. Our partnership with Ipsos combines into one place, the Political Atlas, several different methods for looking at the possible outcome of the coming election.” said Larry J. Sabato, director of the University of Virginia Center for Politics and editor-in-chief of Sabato’s Crystal Ball.
The Political Atlas will be frequently updated throughout the campaign season. Crystal Ball race ratings are adjusted on an ad hoc but frequent basis; the poll-based Ipsos model will be updated weekly; and the social media data will be updated daily.
The polling that undergirds the Ipsos model will be supplemented throughout the fall by 20 separate state-level polls that it will conduct in conjunction with Reuters and the Center for Politics. The first state-level polls will be released next month.
The Political Atlas represents the latest collaboration between the University of Virginia Center for Politics and Ipsos, an international, independent marketing research firm, over the last year and a half. The Center for Politics and Ipsos released two polls on Americans’ attitudes toward recent presidents, and the Center and Ipsos, along with Reuters, released a poll last month on Americans’ racial attitudes at the time of the one-year anniversary of a neo-Nazi march in Charlottesville, as well as a poll on racial attitudes conducted in the aftermath of that march.
