|Dear Readers: Join us today at 2 p.m. eastern for our latest episode of Sabato’s Crystal Ball: America Votes. We’ll be reacting to the first debate and going over the state of the race, including the many rating changes we made today (see below). We also will be hearing from Chris Jackson of Ipsos, the international polling firm with whom we are collaborating on the Ipsos/UVA Center for Politics Political Atlas, which features Crystal Ball race ratings, polling, and much more. Chris will discuss where Ipsos has the presidential race, how voters reacted to Tuesday night’s debate, online vs. telephone polling methodology, and much more.
If you have questions you would like us to address about the debate, specific races, or other developments in the campaign, just email us at firstname.lastname@example.org.
This webinar series is also available as a podcast at Apple Podcasts, Spotify, and other podcast providers. Just search “Sabato’s Crystal Ball” to find it.
— The Editors
KEY POINTS FROM THIS ARTICLE
— With the first debate now in the books, we have close to 20 rating changes across the Electoral College, Senate, and House.
— Joe Biden is now over 270 electoral votes in our ratings as we move several Midwestern states in his favor.
— Changes in the battle for Congress benefit Democrats almost exclusively. We’re moving two Senate races in their direction, as well as several House contests.
Table 1: Crystal Ball Electoral College rating changes
Table 2: Crystal Ball Senate rating changes
Table 3: Crystal Ball House rating changes
About Tuesday night
Trailing nationally and in more than 270 electoral votes’ worth of states, Donald Trump needed more help from Tuesday night’s debate than Joe Biden did. We don’t think Trump did help himself, and it is possible that he actually made his path to a second term harder by demonstrating the poor behavior that seems to turn off so many voters.
Trump’s performance was so outrageous that it made us ponder whether we should make him a significantly bigger underdog in our ratings than he already has been: We’ve had Biden leading Trump in our Electoral College ratings since early April, and Biden’s been slowly inching up in our ratings ever since, while Trump has been receding. That will continue in our rating changes today, although arguably we could or even should go a lot further.
But we have also been cautious throughout this election cycle, cognizant of an electorate that doesn’t seem to swing that much, even in the face of events that one might expect to change minds.
Electoral College changes
With a stable national lead and a bevy of polling showing him running significantly better with northern white voters than Hillary Clinton performed four years ago, Joe Biden appears to be turning back the clock a bit on the United States’ political transformation.
Namely, after Clinton hemorrhaged white voters in northern small town and rural areas in 2016, Biden appears to be bringing some of those voters back into the Democratic fold while also improving on Clinton’s margins with white suburbanites. If this pattern holds in the actual results, it could pay major dividends for Biden in the Great Lakes region, where American presidential elections are so often won and lost and where the electorates in the competitive states are whiter than the nation as a whole.
We have several Electoral College changes this week, all in this region. We are moving Minnesota from Leans Democratic to Likely Democratic. We are also moving its neighbor, Wisconsin — the decisive state in the 2016 presidential election — from Toss-up to Leans Democratic. And, finally, we’re moving Iowa and Ohio, both of which voted for Donald Trump by margins approaching double digits in 2016, from Leans Republican to Toss-up.
These changes push Biden over the 270 electoral vote mark in our ratings, shown in Map 1.
Map 1: Crystal Ball Electoral College ratings
The changes in Iowa and Ohio come both because of the broad improvements Biden has made with white voters in many different places and also because of more recent polling showing the presidential race very competitive in each state. Last Thursday, we took a detailed look at Ohio, and Biden got two of his best polls in the state of the whole cycle there later that day: Up one in a Quinnipiac University poll and up five in a Fox News poll. This squares with some of what we reported in that story, namely that operatives on both sides in Ohio have found the president struggling mightily in key suburban areas.
In Iowa, the tell may have been that Sen. Joni Ernst (R-IA) has been locked in a very competitive contest with businesswoman Theresa Greenfield (D), and it seems like the presidential race is not much different (though Trump typically polls a little better than Ernst).
Meanwhile, Biden’s lead in Wisconsin has been as good or even better than his lead in Michigan and Pennsylvania, two states we’ve had in the Leans Democratic column since June (Pennsylvania) or since we debuted our Electoral College ratings last year (Michigan). Biden’s leads are in the five-to-seven point range in all three states. Minnesota voted slightly more Democratic than these states in 2016, and the president’s bid to flip the state does not appear to be succeeding. We think it merits being in a less competitive category than the Michigan-Pennsylvania-Wisconsin group: If Minnesota flips, something will have gone seriously wrong for Biden, and Trump would almost certainly be winning a second term.
Now, how might we be wrong about the Midwest?
Simple: It is possible that pollsters across many different methods and firms are just overestimating Democratic support among white voters, and it’s showing up most dramatically in this region.
That said, there are reasons to believe that Trump’s great performances with white voters will be hard to replicate this year. The president has never showed much ability to appeal to a wider audience than the voters who backed him in 2016, and exit poll data suggests that a number of voters took a chance on Trump: He did better than Clinton with voters who had an unfavorable view of both candidates. A small number of these voters may be falling by the wayside: For instance, an ABC News/Washington Post poll that had Biden up nine points in Pennsylvania showed 8% of 2016 Trump voters crossing over to Biden while just 1% of Clinton voters were crossing over to Trump.
It’s easy to think of the Trump electorate as immovable, and much of his backing is rock solid, but not every single one of his 2016 supporters was 100% behind him. In the midst of 2020’s bad environment — a global pandemic and a rocked economy — it would make sense that any incumbent president would struggle to add new voters and retain everyone from his last election. Biden also may simply be a better fit for these voters than Clinton was, and the electorate is not static from cycle to cycle.
We may also be compelled to move Texas and especially Georgia to Toss-up sooner rather than later. At the very least, these states are consistently closer in polls than 2016’s decisive trio of Michigan, Pennsylvania, and Wisconsin. It may be, though, that if Biden is turning back the clock slightly to 2012, the Midwest states, even Iowa and Ohio, are better targets than those emerging battlefields of the Sun Belt. This is also why we continue to hold Arizona, Florida, and North Carolina in the Toss-up category. Of those, Arizona is closest to Leans Democratic.
All told, the president continues to need a significant shift in the numbers — or an even bigger polling error than we saw in 2016 — to bring this race back into true Toss-up territory. The clock keeps ticking to Election Day, and votes are already being cast.
Map 2: Crystal Ball Senate ratings
Last week, the Crystal Ball downgraded the prospects of Sen. Susan Collins (R-ME) — we now rate the four-term Maine senator as an underdog against her Democratic challenger, state House Speaker Sara Gideon. Aside from Collins, the only Republican senator running in a Clinton state this year is Sen. Cory Gardner (R-CO). Colorado, at least in 2016, voted a couple of points more Democratic than Maine, and Gardner hasn’t had decades to cultivate a personal brand — as Collins has — so we’ve had his race at Leans Democratic since February.
The picture for Trump is not good in the Centennial State: as of Wednesday, polling aggregates from FiveThirtyEight give Biden a clean 51%-41% advantage. As one Republican operative summed up in July, “Jesus Christ himself couldn’t overperform Trump by double digits.” Senate polling since then has born this out: while Gardner generally performs better than Trump, he often lags his Democratic challenger, former Gov. John Hickenlooper (D-CO), by high single-digits.
Hickenlooper, who has tried to strike a postpartisan tone, was never a darling of the left. Still in late June, he turned back a primary challenge from former state House Speaker Andrew Romanoff — who ran as a more strident liberal, and actually placed first at party’s (less binding) caucus — by a 59%-41% vote. Though his 2020 presidential bid went nowhere, Hickenlooper did win reelection in a hostile 2014 midterm, and he was the only Democrat to win statewide in Colorado that year. The former governor has underwhelmed in this race against Gardner, but that ultimately may not matter much in a nationalized contest.
Gardner scored a significant legislative win this summer with the passage of his Great American Outdoors Act — but on the campaign trail, he’s very much aligned himself with the president. As a result, Gardner receives near-unanimous support from Republicans, but an early September poll from Morning Consult shows him losing Independents — the largest voting bloc, by registration, in the state — by 29 percentage points. Exit polling from 2014 had Gardner carrying the Independent vote 50%-42%.
The looming Supreme Court confirmation battle could also limit Gardner’s crossover appeal. Though it appears he’ll support Trump’s nominee, judge Amy Coney Barrett, Democrats have worked to define her as an opponent of abortion rights and of the Affordable Care Act. In 2014, Democrats probably overplayed the abortion issue against Gardner — then-Sen. Mark Udall (D) hammered Gardner’s pro-life stances ad nauseam. But with a 6-3 conservative court looking more like a reality, Udall’s former attacks may now seem less abstract. Even before the court vacancy, Gardner’s opposition to the ACA seemed to be hurting his electoral standing. So the coverage of the court hearings may emphasize two issues where Republicans are out of step with the Colorado electorate. This pushes our rating to Likely Democratic and emphasizes, in our ratings, that Gardner is clearly the most vulnerable Republican senator.
If nationalization looms large in Colorado, Alaska is a state that, politically, seems to march to the beat of its own drum — but it’s also a state that Senate Republicans seem to be worried about. First-term Sen. Dan Sullivan (R-AK) is locked in an increasingly competitive race with Al Gross, an independent who is running as the Democratic nominee. In September, the Senate Leadership Fund — a group allied with Senate Majority Leader Mitch McConnell (R-KY) — invested $1.6 million in the Last Frontier, a tangible sign that the Republicans aren’t taking the race for granted.
For Republicans to lose in Alaska, they often have to be on the wrong side of a major local issue. Gross’ campaign is pointing to the state’s controversial Pebble Mine project. In recently publicized tapes, the project’s CEO seemed to brag about his connections to local politicians, including Sullivan. Though Sullivan later voiced his opposition to the project, Gross had already featured the issue in his ads.
Like Gardner in Colorado, the imminent Supreme Court hearings may not necessarily lift Sullivan. Alaska has something of a libertarian streak: though it’s a red state, its residents tend to favor abortion rights, an issue that’s likely to play prominently into both parties’ messaging during the confirmation process. During the last Supreme Court hearing, the Alaska Federation of Natives — in a rare move — announced it opposed the confirmation of now-Justice Brett Kavanaugh. But Sullivan supported Kavanaugh anyway, and another Supreme Court fight could help motivate this key Democratic constituency to turn out (roughly 15% of the state’s population are Alaskan Natives).
The Gross campaign may have suffered a setback when the state’s ballot design was unveiled. He’ll be labeled as the Democratic nominee — this is technically accurate, as he won that primary, but is somewhat at odds with the “independent Alaskan” he frames himself as.
Alaska is notoriously hard to poll, but a pro-Gross internal from late September showed the race essentially tied. Sullivan is still the favorite, but we’re not counting Gross out — we’re moving the race to Leans Republican.
The overall contest for the Senate majority remains highly competitive, but the increasing number of Leans Republican-rated races in our ratings — we now have four, in addition to three GOP-held seats where we favor the Democrats and another two listed as Toss-ups — suggests the possibility that, on a good night, Democrats could make very substantial Senate gains.
Table 4: Crystal Ball House ratings
Today’s rating changes solidify our belief that the Democrats are better-positioned to net seats this year than the Republicans. We’re shifting a dozen races, all but one in the favor of Democrats.
The headline changes include shifting a number of Leans Republican districts to Toss-up: the open seat in VA-5 — a central Virginia district that covers the University of Virginia — as well as those held by Reps. David Schweikert (R, AZ-6), Jim Hagedorn (R, MN-1), and Jeff Van Drew (R, NJ-2). In VA-5, former Campbell County Supervisor Bob Good (R) beat Rep. Denver Riggleman (R) at a nominating convention, and now Republicans are concerned that Good is going to kick away the seat against Cameron Webb (D). These are all Trump-won districts where Democrats nonetheless appear to be running strong challenges.
Schweikert, Hagedorn, and Van Drew all have different problems. Schweikert was hurt by a longstanding ethics investigation that harmed his standing and fundraising. Hagedorn, who only narrowly won a Trump +15 district in 2018 after losing several previous attempts, is locked in a close rematch with veteran Dan Feehan (D), and Hagedorn also has had to deal with a controversy involving taxpayer-funded mailers. Van Drew, who switched parties last year, may see his Trump-won district flip to Biden, and he faces Amy Kennedy (D), who is married to former Rep. Patrick Kennedy (D) of Rhode Island.
Another red-district Democratic target is the at-large seat in Montana, where 2018 nominee Kathleen Williams (D) may hold a small lead on state Auditor Matt Rosendale (R), who unsuccessfully challenged Sen. Jon Tester (D-MT) last cycle. A recent New York Times/Siena College poll actually had Williams up 44%-41%, although the undecideds appeared to be a fairly Republican-leaning group. Still, a Likely Republican rating doesn’t reflect the competitiveness of this seat, so we’re moving it to Leans Republican.
We just moved Iowa to Toss-up in the presidential race. If the 2018 gubernatorial race there is any indication, even if Trump carries the state by a few points, Joe Biden could very well flip three of the state’s four districts: IA-1, IA-2, and IA-3. Most of the GOP’s strength in Iowa is concentrated in the northwestern IA-4, while its three other districts are more Democratic than the state. Help from the top of the ticket could benefit Democrats in these districts, which is part of the reason we’re upgrading Rep. Abby Finkenauer (D, IA-1) from Toss-up to Leans Democratic, though she still faces a strong challenge from state Rep. Ashley Hinson (R). It may be, however, that the open IA-2 is the hardest hold for Democrats. We now rate all three of these seats as Leans Democratic.
Three additional first-term Democratic House members move from Leans Democratic to Likely Democratic. Several weeks ago, we moved Rep. Jared Golden (D, ME-2) from Toss-up to Leans Democratic on account of his huge financial edge over his challenger, former state Rep. Dale Crafts (R), and Biden’s likely improvement in Golden’s sprawling, Trump-won district. Since then, several polls have indicated that Golden is up by double-digit margins, and we’re upgrading him as a result. We’re also upgrading Reps. Lizzie Pannill Fletcher (D, TX-7) and Lauren Underwood (D, IL-14), both of whom occupy formerly Republican suburban districts where Biden should perform well. Fletcher has a strong challenger, veteran Wesley Hunt (R), who, if he loses, perhaps just picked a poor year to run; Underwood faces a perennial candidate, state Sen. Jim Oberweis (R), who Republicans are not enthusiastic about. In another Democratic-trending Chicagoland district, we’re moving Rep. Sean Casten (D, IL-6) to Safe Democratic, and we’re making the same move in the St. Petersburg/Clearwater seat held by Rep. Charlie Crist (D, FL-13). Neither faces strong opposition and there’s not much reason to think either would lose.
The one district we’re moving in favor of Republicans is CA-21, held by Rep. T.J. Cox (D). Cox’s Central Valley district backed Clinton by 15 points, but Cox was lucky to beat well-regarded Rep. David Valadao (R, CA-21) in 2018, and Cox has faced questions about his finances and ties to the district. The Democratic tide in the district very well could save Cox, although Republican polling has Valadao up 11 points (this is an extremely difficult district to poll, and Cox’s performance in the state’s top-two primary gives him some reason for optimism, as we wrote after the California primary). This is a Toss-up for now.
One final note: There is a truly bizarre situation going on in MN-2, a suburban/exurban Twin Cities seat held by Rep. Angie Craig (D) that could impact the rating in her seat.
There is a law in Minnesota that specifies if a major party candidate dies within 79 days of an election, the election is postponed to a later special election. In MN-2, there was a candidate from a pro-marijuana legalization party who died recently. This party was specified as a major party because of the performance of one of its candidates in a 2018 statewide race. So as of now, the November election between Craig and Tyler Kistner (R), a veteran running a credible campaign, has been postponed to a February 2021 special election. Craig is suing, arguing that postponing the election would violate federal law. If in fact the November election is postponed, we are moving this district from Likely Democratic to Toss-up, and for the purposes of organizing the House in January, this would at least temporarily count as a Democratic loss (though not a Republican gain). But because of the truly bizarre circumstances, we are not changing the rating yet as we wait for more clarity.
With these rating changes, we now have 232 seats at least leaning to the Democrats, 188 seats at least leaning to the Republicans, and 15 Toss-ups. Splitting the Toss-ups down the middle 8-7 one way or the other would produce a 4-5 seat net gain for the Democrats compared to the House results two years ago.