The Senate: Looking Beyond the “Core Four”

Four races are set to decide the Senate, but they are not the only races that matter

Facebooktwitterredditpinterestmail

KEY POINTS FROM THIS ARTICLE

— The focus on the evenly-matched battle for the Senate has in some ways narrowed to four GOP-held seats: Arizona, Colorado, Maine, and North Carolina.

— Practically speaking, Democrats probably have to win all four, and the White House, to win the Senate.

— However, the map may be expanding. Democrats’ best bet among the other targets probably is Montana, but we still see a small Republican edge there.

— We are making two rating changes this week on the periphery of the Senate map: Alaska and South Carolina move from Safe Republican to Likely Republican.

Table 1: Crystal Ball Senate rating changes

Member/District Old Rating New Rating
Dan Sullivan (R-AK) Safe Republican Likely Republican
Lindsey Graham (R-SC) Safe Republican Likely Republican

Map 1: Crystal Ball Senate ratings

Beyond the core four

There is a widespread consensus that, in the battle for the Senate, there are four races that may effectively decide the majority.

Dubbed the “core four” by one operative, the races in Republican-held Arizona, Colorado, Maine, and North Carolina are the ones that the Democrats seem to have the best chance of flipping. (For sports fans, the “core four” term may ring a bell: It describes the four players at the heart of the New York Yankees’ dynasty of the late 1990s and early 2000s.)

The general belief is that in order to win a Senate majority, Democrats have to sweep all four.

The Senate currently features a 53-47 Republican majority, and the seat likeliest to flip this fall is Alabama, held by Sen. Doug Jones (D). If that happens, Democrats need to win at least four currently Republican seats to forge a 50-50 tie that they hope a Democratic vice president would break in their favor (Crystal Ball contributor Seth Moskowitz went over the math last week, and observed that if Democrats were winning these four races, they probably also would be winning the presidency).

Based on the state of play in the core four races, the race for the Senate overall is something of a coin flip. We have the Democrats narrowly favored against Sens. Martha McSally (R-AZ) and Cory Gardner (R-CO), while Sens. Susan Collins (R-ME) and Thom Tillis (R-NC) are in Toss-up races.

Our sense is that Gardner and McSally currently trail in both of their races (there have been recent public polls showing that in Arizona, but Colorado hasn’t had public polling in months).

North Carolina is the likeliest candidate for most expensive Senate race of 2020, and we don’t see a clear favorite now; we don’t see a favorite in Maine either.

One alarming development for Republicans is that the likely Democratic nominees in all four of these races raised substantially more money than the incumbents in the first three months of 2020.

Former astronaut Mark Kelly (D) has almost double the cash on hand of McSally, even though McSally herself has otherwise been one of the cycle’s standout fundraisers (Kelly has just been incredible). Gardner still holds a two-to-one cash-on-hand edge on former Gov. John Hickenlooper (D), his likeliest fall opponent, but Hickenlooper outraised him substantially, about $4 million to $2.5 million, in the first quarter. Maine House Speaker Sara Gideon (D) roughly  tripled Collins’ fundraising this past quarter; Collins retains a small cash edge, but assuming she is nominated, Gideon will inherit a $4 million warchest that Collins’ opponents started raising in 2018 as Collins backed Brett Kavanaugh’s confirmation to the Supreme Court. And former state Sen. Cal Cunningham (D) more than doubled up Tillis this quarter, although Tillis retains a more than two-to-one cash edge. We’ll have to see what the pandemic does to fundraising in the second quarter, and certainly outside groups on both sides will be spending millions on these races.

The advantage for Republicans, if they retain one, is that Democrats need to win all four of these races and the presidency just to get to 50-50, assuming Republicans flip Alabama and no other seats change hands.

But this is also why it’s worth looking at some of these other races to see if the map is truly expanding as Republicans (mostly) play defense across the country.

Our own short answer is that no other Republican-held seat currently matches the vulnerability of the “core four.” However, at least one race could eventually enter that category.

The best Democratic target right now outside the core four is Montana, in our view. Its emergence as a Senate battleground represents the best argument, for Democrats, that they are truly expanding the Senate map beyond the core four.

Gov. Steve Bullock’s (D-MT) late entry into the race last month prevented first-term Sen. Steve Daines (R-MT) from coasting to reelection: We were prepared to move the race to Safe Republican prior to Bullock’s entry; we now list it as Leans Republican.

Based on what we can piece together, the race seems like it’s neck and neck at the moment. But does that actually make it a Toss-up? We are not quite there yet.

As Moskowitz noted in his Senate overview for us last week, it is very rare for an incumbent senator to lose reelection while that senator’s party is winning the state for president. There are only four examples of that happening in the last seven presidential election cycles, and there were confounding circumstances in three of those four races. Donald Trump would really have to crater in order to lose Montana, which he carried by 20 points in 2016.

So just generically, history is really against Bullock winning.

More broadly, we at the Crystal Ball have been very skeptical in recent years of Democratic Senate candidates trying to capture GOP-held seats in red states. That instinct has served us well in recent years: picking Sen. Pat Roberts (R-KS) in 2014 against an independent when others were skeptical of him winning; never moving Senate Majority Leader Mitch McConnell (R-KY) from Likely Republican the same cycle despite his race seeming very competitive at times (he won by 15 points); and always favoring the Republicans in an open-seat race in Tennessee last cycle (now-Sen. Marsha Blackburn ended up winning by 11 points).

In other words, we are really going to need compelling evidence to move Montana to Toss-up. We think Bullock’s chances are decent, but we also think Daines has to be looked at as a small favorite still.

Montana is not as Republican as Trump’s 20-point margin suggested, and the incumbent presidential party often sees their performance sag from first election to second election in Montana, as friend of the Crystal Ball Jacob Smith has documented. Democrats have won the governorship four times in a row, and Sen. Jon Tester (D-MT) has been elected three times in recent years. Prior to Daines’ 2014 victory, the seat he now holds had never voted for a Republican in the popular election era, which is a testament to an ancestral Democratic tradition in the state. Bullock is a reasonably popular governor at a time when governors in general are attracting widespread acclaim for their handling of the coronavirus crisis.

So this is a real race, to be sure. Just one where we continue to see a GOP edge.

Sen. Joni Ernst (R-IA) also appears to retain an advantage in another Leans Republican seat, although the state could hypothetically dip into Toss-up territory at some point. Meanwhile, Sen. Gary Peters (D-MI) has been outraised by 2018 nominee John James (R) in three straight quarters, although their cash on hand is about the same. We rate that race Leans Democratic, mirroring our rating for president (which is more bullish for Democrats in Trump-won Michigan than other forecasters). Joe Biden (and Peters) have generally led recent polls there.

Our other race in the leans category, the Georgia Senate special, is one we may essentially have to set aside until November. The likeliest scenario is that either appointed Sen. Kelly Loeffler (R-GA) or Rep. Doug Collins (R, GA-9) will advance to an early January runoff following an all-party primary and be opposed by the Rev. Raphael Warnock (D). Democrats see Warnock as an electrifying candidate who can rally black turnout in the runoff; Republicans, meanwhile, point to Warnock’s problematic divorce and history that suggests Republicans are favored in a runoff election turnout model.

The eventual winner will have to defend the seat in the 2022 regular election (the same is also true of Arizona).

Beyond these seats in the Toss-up and Leans categories, there are many other races in the Likely category on the periphery of the competitive Senate map. That group gets a little bigger this week as we move two previously Safe Republican seats, Alaska and South Carolina, to Likely Republican.

Alaska and South Carolina

Alaska is a GOP-leaning state in national races but is known for its local idiosyncrasies. Since 2004, Republican nominees for president have carried the Last Frontier by margins ranging from 14 to 26 percentage points.

During that same stretch, no Alaska Senate candidate has won with a majority of the vote.

In 2014 — a disastrous midterm for Democrats across the county — Alaskans didn’t seem to care about partisan loyalty when registering their discontent at the polls: it was the only state that year to oust both a sitting senator, a Democrat, and its incumbent governor, a Republican.

That 2014 Senate race became an expensive contest between incumbent Democrat Mark Begich and Republican challenger Dan Sullivan. Begich was a household name in the state — a theme of his ads was his late father’s service representing the state in the House. But Sullivan tied the incumbent to another leader: President Obama. Despite running especially well with the state’s Native American voters, who make up 15% of the population, Begich lost his seat in a narrow 48%-46% vote.

This year, Sullivan will have the state’s partisan lean on his side, as there’s little doubt Trump will carry the state. During his first term, Sullivan has been a low-profile, party-line conservative — which actually means Sullivan is a bit out of step with the two other members of the state’s all-Republican federal delegation.

Alaska’s other senator, Lisa Murkowski, can almost be seen an independent who caucuses with Republicans. She has a record of breaking with her party on key votes — she voted against the Republican Affordable Care Act repeal effort and against the confirmation of now-Justice Kavanaugh. In her third full term, Murkowski has run a gauntlet of three competitive races, but has cobbled together coalitions that cut across racial and partisan lines.

In the lower chamber, Alaska can claim the Dean of the House: Rep. Don Young (R, AK-AL), who has served since 1973. A colorful figure, he seems most interested in using his seniority to steer resources to the state, while tending to local infrastructure and land issues. Young has faced competitive races in recent cycles, but has held on thanks, in part, to uncommon support from Alaska Natives — a group that will likely be less open to Sullivan. The Crystal Ball rates Young’s race as Likely Republican.

Against Sullivan, national Democrats are again rallying around a candidate with some family political lineage in the state — though this time, with a twist. The Alaska Democratic Party endorsed Al Gross, a surgeon who is running as an independent; his father served as state Attorney General in the 1970s under Gov. Jay Hammond (R-AK). Gross outraised Sullivan last quarter, though Sullivan can claim a better-than two-to-one cash-on-hand advantage. Alyse Galvin, Young’s credible challenger in both 2018 and this year, is also a hybrid independent/Democrat.

In some ways, this race looks reminiscent of the 2014 Kansas Senate contest, mentioned above. In that scenario, Republican Sen. Pat Roberts seemed to be running an underwhelming reelection effort. In what initially appeared to be a three-way scenario, Democrats convinced their candidate to drop out in favor of independent businessman Greg Orman. In a race that appeared headed to photo finish, Orman posted slight leads in most final polls. On Election Day, though, his support seemed to evaporate, as Roberts won by a surprisingly decisive 11-point margin.

Democrats may point to another 2014 race as a template for 2020. Bill Walker, Republican-turned Independent, upset Gov. Sean Parnell (R-AK) that year. As Alaska’s fusion voting system allows, Walker teamed up with a Democratic running mate, Byron Mallott. Parnell entered the election under fire for his handling of abuse in the state National Guard. During the closing weeks of the campaign, the Walker/Mallott ticket got a boost from an unlikely ally: former-Gov. Sarah Palin (R-AK), who criticized Parnell for some of his budgetary decisions (Parnell, formerly the lieutenant governor, took over for Palin after she resigned in 2009). In this wild race, the result turned on factors and personalities that were local — something unlikely to be replicated in a federal race for the Senate.

Alaska is more receptive to third party candidates than many other states are — after all, Murkowski was reelected in 2010 as a write-in candidate after losing the Republican primary. Even so, Kansas offers a cautionary tale for Alaska Democrats, and they would be unwise to rely on a redux of their state’s 2014 gubernatorial contest. We see Sullivan as a clear favorite, but downgrading this race from Safe Republican to Likely Republican seems reasonable given Alaska’s quirkiness and because Gross is at least a credible challenger who should run a real campaign.

Speaking of write-in campaigns, our other ratings change this week comes in a state that also saw a famous victory by a write-in candidate: South Carolina. In 1954, Strom Thurmond won as an “Independent Democrat.” This was six years after he had run as a conservative Democratic “Dixiecrat” alternative for president against Harry Truman, and a decade before he switched from the Democrats to the Republicans.

Thurmond would serve in the Senate from the mid-1950s all the way to 2003, when he was replaced by Sen. Lindsey Graham (R-SC).

Graham, running for a fourth term, may face a more competitive race than we initially thought.

A longtime ally of the late Sen. John McCain (R-AZ), Graham and McCain would frequently fashion themselves as mavericks. The duo irritated conservative elements of their party, especially on issues like immigration and climate change. In his 2014 primary — armed with a $12 million warchest and aided by fractured opposition — Graham polled at 56%. He went on to win the general election 54%-39%.

Since McCain’s death in August 2018, Graham has become a more full-throated advocate of the Trump administration’s priorities after initially being a major Trump critic during the 2016 presidential primary cycle (when Graham himself was a candidate). In a viral clip, Graham defended Kavanaugh during that contentious Supreme Court confirmation hearing. This maneuvering played well with his primary electorate, as did his defense of Trump during the battle over impeachment. An April 2019 poll from Morning Consult showed that 80% of South Carolina Republicans approved of Graham’s work — at the time, it was the highest intraparty approval rating of any GOP senator up for reelection in 2020.

If Graham’s rightward trek strengthened his hand with conservatives, it deadened his prospects for crossover support with Democrats — not exactly a bad trade for the senator, considering Trump carried the state 55%-41% in 2016. Still, Democrats have vowed to give him a more spirited contest than they have in past cycles.

Graham’s likely general election opponent is Jaime Harrison, who chaired the state Democratic Party from 2013 to 2017. Harrison’s fundraising has been excellent. After raising more than $7 million in 2019, he set up a joint fundraising operation with Cunningham, noted above as the Democratic nominee in the Toss-up North Carolina race.

In the first quarter of 2020, Harrison outraised Graham by roughly $7 million to $5.5 million, although Graham still holds a cash edge. The state’s partisanship should be enough to keep this seat in Republican hands. Still, Harrison’s fundraising hauls are an indication that this could be the most competitive senatorial race in South Carolina since the Bush era. Harrison’s internal polling from late March gave Graham a four percentage point lead, though an older public poll from NBC News/Marist College showed the incumbent up 54%-37%. Likely Republican feels like the most appropriate rating.

Democrats are hopeful that Graham will run behind Trump, in part because of some of Graham’s past apostasies and long history in politics, while also losing some usual GOP support in suburban Charleston and Charlotte. Meanwhile, Harrison will try to excite (and expand) the African-American electorate (Harrison is black).

South Carolina remains the most Republican state on the Eastern Seaboard. Harrison’s path is not an easy one, but it is also not an impossible one.

The periphery of the competitive Senate map

With the Crystal Ball’s ratings changes this week, among the biggest differences between our Senate and Electoral College ratings — at least to the extent that differences are warranted — is the number of states occupying the Likely Republican bracket. With the additions of Alaska and South Carolina, seven Senate races fall into the Likely Republican category. In our Electoral College picture, there are zero states in that category.

To a large degree, this is a function of Senate’s staggered terms. Every two years, one-third of the chamber is up for election. The group of states that make up this year’s map, known as the Class II seats, skew towards states in the south and Great Plains — in other words, red states.

This map was last up in 2014, when Democrats lost nine seats and their Senate majority. Ordinarily, that would give the party generous room to rebound the next time those seats were contested. However, seats that flipped Republican in 2014 included Arkansas, Louisiana, South Dakota, and West Virginia. These red-trending states are unlikely to be seriously contested now that they are in Republican hands — Democrats used to hold those states with popular incumbents, good luck, or a combination of the two.

Of the 23 Republican-held Senate seats up for election this time, only two — Colorado and Maine — are in states that Hillary Clinton carried in 2016. Given that imbalance, Democrats’ path to the Senate majority will rely on how they do in states that are not naturally favorable to them.

Alabama, Alaska, Kansas, Kentucky, and South Carolina are Safe Republican for President Trump, but we rate them Likely Republican for Senate. Given factors like candidate quality, fundraising, and local dynamics, it seems like a good bet that Democratic Senate candidates in these states will run ahead of Joe Biden, by now the presumptive Democratic nominee. For example, of that quintet, Democrats are only defending one state, Alabama. It’s likely that Trump will clear 60% of the vote in Alabama, but Sen. Doug Jones (D-AL) generally trails his prospective GOP opponents by smaller margins.

Conversely, the Crystal Ball puts Sens. John Cornyn (R-TX) and David Perdue’s (R-GA) contests in the Likely Republican category, but we have the states at just Leans Republican for president. At this point, we see both Cornyn and Perdue as having the potential to run ahead of Trump because of resource advantages over uncertain Democratic fields.

In 2016, Georgia was one of just 11 states where Trump underperformed Mitt Romney’s 2012 showing. Meanwhile, at the senatorial level, it was roundly considered a recruitment bust for Democrats. Trump still carried the state, 50%-45%, but then-Sen. Johnny Isakson (R-GA) saw a more robust 55%-41% win. While we don’t necessarily expect the gap to be that severe this time, Perdue has an opportunity to run ahead.

In the Trump era, Democrats have certainly made gains throughout the Sun Belt. But the Georgia and Texas races serve as a reminder that partisan re-alignment is slow process, often drawn out through several election cycles. While these states may well be emerging Electoral College battlegrounds, the onus is on Democrats to make the Senate results mirror the presidential results.

Overall, the focus in the race for the Senate is rightly on the “core four” races – Arizona, Colorado, Maine, and North Carolina. But there are many other races that are engaged to at least some degree, and whether any of them become more competitive over the next few months may tell us about the overall direction of the Toss-up battle for the Senate.