Rating Changes: Maine Senate Moves to Leans Democratic

ME-2 electoral vote now a Toss-up; South Carolina Senate narrows

Dear Readers: On Thursday at 2 p.m. eastern, we’ll be devoting our new episode of our Sabato’s Crystal Ball: America Votes webinar series to previewing the debates, which start next week. If you’ve got questions about the debates — both on the matchup this year, and also those from the past — just email us at goodpolitics@virginia.edu.

You can watch live at our YouTube channel (UVACFP). You can find previous episodes of the show there, as well as other recent UVA Center for Politics panels and programming. We’ll include the direct link in the next issue of the Crystal Ball.

Additionally, an audio-only podcast version of the webinar is now available at Apple Podcasts, Spotify, and other podcast providers. Just search “Sabato’s Crystal Ball” to find it.

— The Editors


— The passing of Supreme Court Justice Ruth Bader Ginsburg introduces an explosive new issue into the 2020 contests, although it may not fundamentally alter the race.

— Based on trends in public polling, and with a potential assist from Maine’s ranked-choice voting system, we’re moving that state’s critical Senate race from Toss-up to Leans Democratic.

— The single electoral vote from the sprawling ME-2 is now a Toss-up.

— Sen. Lindsey Graham (R-SC) is in greater danger of an upset, and his race moves from Likely Republican to Leans Republican. The court fight could save him, though.

Table 1: Crystal Ball Senate rating changes

Senator Old Rating New Rating
Susan Collins (R-ME) Toss-up Leans Democratic
Lindsey Graham (R-SC) Likely Republican Leans Republican

Table 2: Crystal Ball Electoral College rating change

State Old Rating New Rating
ME-2 Leans Republican Toss-up

Map 1: Crystal Ball Senate ratings

Map 2: Crystal Ball Electoral College ratings

An uncertain impact from Ginsburg’s passing

The passing of Supreme Court Justice Ruth Bader Ginsburg on Friday elevates the Supreme Court to a major electoral issue for the third election in a row, following Senate Republicans’ refusal to consider Barack Obama’s nomination of Merrick Garland in 2016 and the bitter confirmation battle over Brett Kavanaugh in 2018. The implications for the future of American government are extraordinary. The implications for the election may be as well — or not.

In the aftermath of Ginsburg’s passing, many have speculated as to whether one side might be extra motivated by the vacancy over the other. But was this a sleepy election in need of a jolt? Hardly. It is possible that, despite the pandemic, 2020 could set a modern record for turnout. The battle over the court’s future turns up the heat of American politics, but the temperature was white hot already.

The procedural arguments from the Garland blockade have flipped, with Democrats arguing for the voters to decide through the presidential election, and Republicans arguing for Donald Trump to make the pick. Those making these arguments have shifted, but the source of power in the Senate remains the same now as it was four years ago: Republicans and their majority leader, Mitch McConnell (R-KY), still rule the roost. The argument that prevails in the Senate is whatever McConnell can get 51 votes behind. The Republicans have 53 Senate seats, so it would take four Republican defections both before and after the election, in the lame duck period, to keep the seat open.

Of the eight Republican-held Senate races the Crystal Ball now rates in the Toss-up or Leans categories, just two of them — Colorado and Maine — are very likely to vote more Democratic than the nation as a whole in the presidential election. That makes the Ginsburg vacancy an added burden for Sens. Cory Gardner (R-CO) and Susan Collins (R-ME).

But the other six states — Arizona, Georgia, Iowa, Montana, North Carolina, and South Carolina — seem very likely to, at the very least, vote more Republican than the nation, and Donald Trump might carry all of them even if he is narrowly losing the election, and almost certainly if he is winning. The Republican senators in all of these states, except perhaps for Sen. David Perdue (R-GA), all have had to worry about possibly running behind Donald Trump in their respective states. If the Court battle further polarizes the electorate and reduces ticket-splitting, that could help Republicans in the Senate. But it’s also premature to make any hard and fast projections about what might happen, and it’s not out of the question that Joe Biden could carry half or more of those six states — and bring enough Democratic Senate candidates along with him to get the Democrats over 50 Senate seats.

(A partisanizing vote might also be helpful to Michigan Democratic Sen. Gary Peters, who in recent surveys has led Republican rival John James but often by less than Biden is leading statewide.)

One senator who has said she wants the next president to make the pick is Collins, who once enjoyed immense bipartisan support in Maine but has been partisanized in large part because of her vote to confirm Kavanaugh in 2018. Could Collins cobble together a Republican Gang of Four to keep the seat open? That may be what it would take for her to maintain the level of crossover support she would need to win, but this development also hyper-nationalizes her race at a time when she wants to be localizing it.

Full disclosure: On Friday morning, the Crystal Ball team decided to make three rating changes — the ones described above, moving Collins’ race to Leans Democratic, shifting the ME-2 electoral vote to Toss-up, and downgrading Graham to Leans Republican. The Ginsburg news obviously gave us pause in going through with these changes, but after following developments over the weekend, we have decided to go ahead with these three changes. Here’s why:

Susan Collins’ two big problems

Sen. Susan Collins (R-ME) appears to retain some of her trademark crossover appeal, but between a consistent stream of public polls showing her trailing and the unique electoral structure of her race, she appears to be in an increasingly perilous position in her bid for a fifth term.

As we’ve noted in this space before, the two parties appear to have considerable disagreements on where this race stands. Democrats believe that Collins’ opponent, state House Speaker Sara Gideon (D), is in the lead, whereas Republicans believe that Collins is holding up well.

In this kind of situation, public polls can act as a tiebreaker, and the public polling points to Gideon. This is Collins’ first problem.

Of 16 publicly-released polls of the race that have come out this year (as compiled by FiveThirtyEight’s database) from nonpartisan, Democratic, and Republican sources, Collins has only led in a single one, a Republican internal poll conducted in mid-June. Collins led 45%-38% in that survey. A few others have shown Gideon up only a point. But notice that even in this GOP internal, the longtime incumbent Collins was at just 45% of the vote. Across the 16 polls this year, her average support on the ballot has been just 42%. That is not a strong number for an incumbent who surpassed 60% in her last two reelection bids.

This brings us to Collins’ second problem: The state’s ranked-choice voting system.

When we compiled the average showing Collins at 42% during 2020, we used (when available) polls that included the race’s two other candidates: independents Lisa Savage (a progressive) and Max Linn (a conservative). Savage and Linn appear to be getting enough support that ranked-choice voting — a relatively new voting system that only Maine currently uses at the statewide level — may determine the victor in this race by preventing both Gideon and Collins from getting to 50% outright. That’s what happened in the ME-2 congressional race in 2018, when now-Rep. Jared Golden (D) finished slightly behind then-Rep. Bruce Poliquin (R) in the first round of voting, but moved ahead after the second-choice votes from those who backed third party candidates were taken into account. Collins, as the long-time incumbent, may have a harder time benefiting from this ranked-choice system than the challenger Gideon: Savage is actively encouraging her supporters to back Gideon with their second-place ranking, while some Republicans worry that Linn’s backers may not list Collins second (conservatives are much less supportive of ranked-choice voting in Maine than liberals are, perhaps understandably given the ME-2 result in 2018).

In a poll released Friday, the New York Times/Siena College simulated the ranked-choice vote. Gideon led 44%-40% on the initial ballot, with Savage and Linn at 2% apiece. When the votes for the independents were allocated, Gideon went up 49%-44%. So both candidates gained, but Gideon’s lead went up slightly.

The details of a state’s election procedures matter: In Georgia’s Senate elections this year, we think the state’s general election runoff rule may give the Republicans a little bit of an advantage compared to if there were no runoff. In Maine, the Democrats may get a little bit of a benefit from ranked-choice voting compared to if the state did not have the system.

If we end up being wrong about this rating change, and Collins ends up winning, there might be a fairly obvious explanation: Collins still is going to win a chunk of Joe Biden presidential votes. William Cohen, the former Republican senator from Maine and Bill Clinton-era secretary of defense, is one of those voters, a point he made in an endorsement of Collins late last week.

The three most recent Maine polls (NYT/Siena, Quinnipiac, and AARP) showed Biden leading Trump statewide by 17, 21, and 14 points, respectively, with Gideon up 5, 12, and 1 in the Senate race. If Biden’s lead in Maine ends up being considerably smaller than these polls indicate — and those Biden statewide leads do seem high, although they are three different polls from pollsters we respect, so they are collectively harder to dismiss as outliers than they would be on their own — Collins is demonstrating enough crossover support to hang on.

But we also have to consider the trajectory of this race. Last summer, the AARP poll, then conducted by the GOP firm Fabrizio Ward, found Collins up 17 points on Gideon. That poll, this time conducted jointly by Fabrizio Ward plus the Democratic firm Hart Research Associates, now has her down one.

In last summer’s AARP poll, Collins’ margin was 23 points better than Donald Trump’s in the state (he was down six to Biden in that poll). In this more recent survey, Collins’ margin is 13 points better than Trump’s. That’s still an impressive crossover vote these days, but it’s considerably smaller than what she enjoyed last year.

As the campaign grinds on to conclusion, we’ll have to see if Collins’ ticket-splitting edge widens or narrows. The latter may be more likely, unless Maine swing voters become so convinced of a Biden presidential win that they stick with Collins to serve as a check on a Biden presidency. However, in both the Quinnipiac and NYT/Siena polls, voters said they preferred a Democratic-controlled Senate to a Republican one by a margin of about a dozen points. Generally speaking, the public also seems less confident of a Biden victory than Biden’s polling lead would suggest, a dynamic we explored a couple of weeks ago. And even if Collins spearheads an effort that effectively blocks Trump from making an appointment, that could cost her Republican voters, which could flow to the conservative independent, and perhaps those voters will not rank her second on their ballots. In other words, while the court vacancy introduces a significant wild card into this race, we think it’s likelier to hurt Collins than help her.

Just like in our other Leans Democratic-rated Senate races in states held by Republicans — Arizona and Colorado — there are paths to a Republican comeback. But for now we think Collins is in a difficult enough position that we’re moving the race to Leans Democratic.

Crucial ME-2 electoral vote now a Toss-up

We are making one other change in Maine: We’re moving the single electoral vote in Maine’s 2nd Congressional District from Leans Republican to Toss-up. We already moved the district’s congressional race, in which Golden is seeking a second term, from Toss-up to Leans Democratic a few weeks ago. The NYT/Siena poll found Golden up by 19 points, which bolsters that change. But the presidential race appears to be close in the district. NYT/Siena, Quinnipiac, and AARP found Biden up by two, nine, and four points, respectively, in the heavily white, working-class district. This after Trump carried the district by 10 points in 2016, a massive shift from Barack Obama’s nine-point win there in 2012.

The potential for a Biden victory in ME-2 underscores one of the trends in polling this year: Biden is polling better than Hillary Clinton performed mostly because he is running better with white voters, both with a four-year college degree (a group with which he is doing extremely well for a Democrat) and without a four-year degree (a group he’s losing, but by less than Clinton). Given these shifts, it stands to reason that Biden could be doing markedly better in lily-white Maine than Clinton, who only carried the state by three points overall.

The single ME-2 vote looms large. Our current Electoral College ratings have 269 electoral votes at least leaning to Biden, so ME-2 could hypothetically be vote no. 270 for Biden, although one probably would expect states that were closer in 2016 to flip before it does (namely, the crucial trio of Michigan, Pennsylvania, and Wisconsin).

Graham-Harrison gets closer

During the last Supreme Court confirmation fight, in 2018, Sen. Lindsey Graham (R-SC) emerged as one of the most visible faces on Capitol Hill. This year, he’s set to have an even larger role in any potential hearings, as Chairman of the Judiciary Committee. At home, Graham has been locked in a tightening contest with a well-financed Democratic opponent, Jaime Harrison.

In late April, when the Crystal Ball put South Carolina’s Senate race on the board, moving it from Safe Republican to Likely Republican, we noted that the 2020 race was shaping up to be its most competitive Senate contest since the Bush era. By almost all indicators, the race has gotten closer since then. In polls, Harrison continues to find himself within the margin of error of Graham.

During the summer, the Harrison campaign released some internal polling showing the race within a few points. Public polling since then has confirmed that: in early August, Quinnipiac polling found Graham and Harrison tied at 44% — with a slightly more Republican-leaning sample, and after a likely voter screen, the race remained tied, at 48%, when Quinnipiac surveyed it last week.

While fundraising isn’t everything — over the past few cycles, some Senate candidates have lost to lesser-funded opponents — Harrison’s dominance on that front is hard to ignore. During the month of August he raised over $10 million, a figure more than many senatorial candidates across the country can raise in an entire quarter. Just last week, the Harrison campaign claimed to rake in $2 million over 48 hours, and the Democratic Senatorial Campaign Committee recently announced an investment in the race. In just a little over a day after Ginsburg’s death, the Democratic fundraising site ActBlue reported more than $90 million in donations — some of that has undoubtedly gone to buoy Harrison.

Though he’ll naturally act as one of the White House’s most important allies in any looming court hearings — he’s already signaled his intentions to move forward with the process — Graham has broken somewhat with the president in certain ways. In July he praised Dr. Anthony Fauci. Though he’s become a household name during the pandemic as the nation’s leading infectious disease expert, Fauci has, at times, drawn the ire of Trump. More recently on the campaign trail, Graham has aimed to frame himself as the more transparent candidate, in part by releasing 11 years of tax returns. While calling on Harrison to do the same (Harrison eventually did), he encouraged Trump to release his own — a charge that the president’s Democratic critics often make, though in more pointed terms. Graham has also declined to criticize Joe Biden, his former Senate colleague.

One actually wonders if Graham, who morphed from a Trump critic in the 2016 GOP primary cycle to a major Trump ally, would be better off trying to solidify his standing with Trump voters instead of trying to generate crossover support. Given the notable protest vote against him in June’s primary vote, and Trump’s stronger position in the state — Trump is generally around 50% in state polling and leads Biden — Graham still has work to do solidifying the Trump vote, and we’ll be watching if his standing with GOP partisans strengthens as the Supreme Court confirmation process unfolds. Just as the ranked-choice votes of non-major party candidates could hurt Collins in Maine, the makeup of the South Carolina ballot could present a problem for Graham, and it arguably should incentivize Graham to prioritize outreach to the right. The Constitution Party is slated to be the only third party that will be represented on the ballot — the party’s candidate, Bill Bledsoe, will almost certainly attract conservatives. So minimizing Bledsoe’s share will be key for Graham, while maximizing it is an important part of Harrison’s path to victory.

This is where the court fight could ultimately save Graham, who as the Judiciary Committee chairman is one of the key players in the looming court fight. Still, Graham is among those who are going to look the most hypocritical when comparing his statements when blocking Garland four years ago versus pushing through Trump’s pick now. It’s a certain bet that his Democratic colleagues — and Harrison — will bring up Graham’s past statements, and that could hurt Graham with moderate voters, even if it boosts his credibility on the right (as noted above, the latter ultimately may be more important in a state like South Carolina).

This race could end up like the 2018 Texas Senate contest, where Sen. Ted Cruz (R-TX) received a strong challenge from then-Rep. Beto O’Rourke (D, TX-16). Though O’Rourke came closer than any Democrat in 30 years to winning a Lone Star State Senate seat, the state’s partisanship helped carry Cruz. Another O’Rourke-Harrison commonality is cash: having a prominent opponent seems to open up wallets.

On the surface, South Carolina hasn’t seen the types of massive shifts that the Texas electorate has seen, but there are some common components. O’Rourke came close to winning because of his historic showing in the suburbs. Recently, the Crystal Ball upgraded the prospects of first-term Rep. Joe Cunningham (D, SC-1), in the Charleston metro, and downgraded Rep. Joe Wilson’s (R, SC-2) chances — Wilson’s seat encompasses the Columbia suburbs, which have seen some blue movement. It seems reasonable that Harrison could end up carrying Cunningham’s district, perhaps comfortably, and coming close in Wilson’s — this may be enough to give him a path to victory. With that in mind, we’re moving the race from Likely Republican to Leans Republican, although remember that a Leans Republican rating still means we see Graham as a favorite.