The Senate: Ranking the Top Dozen Best Targets

Blogger’s note: One wonders why the Senate Republicans are: A. digging in on support of Trump and opposing big aid package to those hit hardest by pandemic B. Falling behind in more and more Senate races because of A.

Establishment Republicans relieved by Kansas primary result; rating changes in Georgia and Iowa

Facebooktwittergoogle_plusredditpinterestmail

KEY POINTS FROM THIS ARTICLE

— National Republicans breathed a sigh of relief on Tuesday night, as Rep. Roger Marshall (R, KS-1) beat 2018 gubernatorial nominee Kris Kobach (R) in the Kansas Senate primary. Practically speaking, the Kansas Senate race went from being a potentially major Democratic offensive target to one where the Republicans have a very clear edge.

— Kansas remains Likely Republican in our ratings.

— We rank the top dozen Senate seats in order of their likelihood of flipping. Of the 12, 10 are held by Republicans, underscoring the amount of defense that the GOP will need to play in order to hold their majority.

— We have two Senate rating changes, one in favor of each party.

Table 1: Crystal Ball Senate rating changes

Senator Old Rating New Rating
Georgia Special Leans Republican Likely Republican
Joni Ernst (R-IA) Leans Republican Toss-up

Map 1: Crystal Ball Senate ratings   This map is insane; Larry Sabato, what’s going on.

GOP leadership overjoyed by Kansas primary result

In a cycle where the Republicans’ list of defensive responsibilities in the Senate has seemed to get longer and longer, GOP leaders must be extremely happy to be able to effectively cross one off the list. Rep. Roger Marshall (R, KS-1) defeated 2018 gubernatorial nominee and conservative hardliner Kris Kobach (R) Tuesday evening, making it much easier for Republicans to defend the open seat and frustrating national Democrats, who spent real money in Kansas to try to help Kobach win the primary.

Kobach kicked away the Kansas governorship last cycle, losing a very winnable race to now-Gov. Laura Kelly (D). Establishment Republicans were so petrified of Kobach losing a Senate general election that they first implored Secretary of State Mike Pompeo (R) to come home and run for the seat and then tried to get President Trump to back Marshall against Kobach, who Trump endorsed in his very narrow 2018 gubernatorial primary victory. As it was, Trump stayed out, but Marshall won anyway.

Democrats have a respectable nominee, party-switching state Sen. Barbara Bollier (D), but Marshall fits the traditional Kansas GOP mold much better than Kobach. This is the second time Marshall has beaten a further-right Republican in a contentious primary; he also knocked off then-Rep. Tim Huelskamp (R, KS-1) in 2016.

Despite signs of Democratic growth in the Kansas City suburbs and a few other places in the state, Kansas remains a Republican state: The president carried it by about 20 points in 2016. Even if Trump significantly underperforms in the state, he is still very likely to carry it, meaning that Bollier will need to attract at least some crossover support from Trump voters to win. That would have been an easier task against Kobach than Marshall. Kansas also has not elected a Democrat to the Senate since 1932, despite electing many Democratic governors in that same timeframe: A state’s baseline partisanship is often easier to overcome in state races as opposed to federal ones.

We’re keeping the Kansas Senate race as Likely Republican, matching our presidential rating there, but Marshall should be fine.

This is a good development for Senate Republicans, although they still have a lot of defense to play in other states. Speaking of…

The big picture

As we examine the race for the Senate majority, we thought it’d be worthwhile to rank the dozen seats we see as the most competitive from most to least likely to change hands. As we see it right now, 10 of the 12 most vulnerable seats are held by Republicans, even as Democrats are defending the seat likeliest to flip, Alabama.

1. Alabama (D)
2. Colorado (R)
3. Arizona (R)
4. Maine (R)
5. North Carolina (R)
6. Iowa (R)
7. Montana (R)
8. Georgia (Regular) (R)
9. Michigan (D)
10. Texas (R)
11. Georgia (Special) (R)
12. Alaska (R)

Before we explain the rankings (and a couple of rating changes), we wanted to explain how presidential partisanship plays into them. While presidential and Senate results will differ, presidential and Senate outcomes have come further into alignment in recent years.

Table 2 shows the same ranking of Senate seats in terms of likelihood of flipping, but we also added three additional columns.

Table 2: Presidential scenarios in top 12 Senate races

The first is the actual 2016 presidential margin by state, when Hillary Clinton won the national popular vote by two points. The second is a hypothetical scenario in which Joe Biden would win by five points nationally, or three points better than Clinton, and the third is a hypothetical where Biden would win by 10 points nationally, or eight points better than Clinton. A positive number is a Democratic presidential victory in a given state; a negative number indicates a Republican win.

We adjusted the state-level presidential margins to match the hypothetical national change from 2016; this would represent what political scientists might call a “uniform swing,” in which these states’ presidential margins change the same way the national margin does. Reality won’t be so neat and tidy, but this does give us a presidential baseline as we go through our Senate list.

We are not going to say that the situation of Sen. Doug Jones (D-AL) is hopeless, but he has trailed even in Democratic internal polls — when a candidate is behind in even his own party’s polls, he is behind, and likely by more than the party polls show (as nonpartisan surveys have shown). The presidential scenarios show that, even in the event of a Biden national blowout, Jones will need an immense amount of crossover voting to win.

Sens. Cory Gardner (R-CO) and Martha McSally (R-AZ) have generally been behind in their races; there is a little more uncertainty with Gardner given that former Gov. John Hickenlooper (D-CO) suffered through a very bad string of news coverage in advance of his primary a little over a month ago, but we haven’t seen much indication that Gardner has changed the race in a meaningful way. McSally has been behind, and generally not just by a few points, in Arizona, a more frequently polled state. The difference between the two races is the presidential: It’s not hard to imagine Trump winning Arizona, a purple-trending red state, but it is hard to imagine Trump winning Colorado, a blue-trending purple state. So Gardner will need to attract more crossover support than McSally — he likely will win some, but we’d be surprised if he gets enough. Meanwhile, polls show McSally running behind Trump when she may need to run ahead of him.

The presidential factor is also the reason why we see Sen. Susan Collins (R-ME) as slightly more vulnerable than Sen. Thom Tillis (R-NC): Joe Biden seems very likely to carry Maine, and by a bigger margin than 2016, while North Carolina (like Arizona) remains a presidential Toss-up. Again, Collins (like Gardner) probably will get crossover support, but perhaps not enough. Tillis, just like McSally, polls behind Trump.

Beyond Maine and North Carolina, Iowa is now in the Toss-up column. Sen. Joni Ernst (R-IA), like McSally and Tillis, appears to be doing a little worse than Trump in her state. She has a little more wiggle room than the other two — note that Trump still carries Iowa even in this hypothetical scenario where Biden is winning nationally by 10 — but both parties are acting (and spending) like Iowa is a Toss-up.

We continue to rate Montana and Georgia’s regular Senate election as Leans Republican even though good cases can be made that both should be Toss-ups. We have different reasoning for keeping both where they’ve been in our ratings.

In the case of Montana, presidential partisanship is key: Trump seems very likely to carry the state again, albeit by a reduced margin, and it’s historically difficult to dislodge a sitting senator whose party is winning the state concurrently in the presidential race. Additionally, the trajectory of the race may actually be going the way of incumbent Sen. Steve Daines (R-MT); a couple of months ago, we thought Gov. Steve Bullock (D-MT) was leading Daines. Now, based on what we’ve heard and seen, we are not so sure, and Daines may be ahead, slightly.

In Georgia, Sen. David Perdue (R) is locked in a close race with former congressional candidate Jon Ossoff (D), although he generally polls a little bit better than Trump, and he may be able to attract a little bit of crucial crossover support from Trump-skeptical Metro Atlanta suburbanites who aren’t quite ready to abandon the GOP down the ballot. Perdue also has a backstop in his race: a general election runoff if no one gets over 50%. As we explained in a deep dive on Georgia, the runoff scenario could help Republicans in terms of turnout. So Ossoff may need to get over 50% in the November general election to practically be able to win the seat.

The Republicans’ other offensive target on this list, Sen. Gary Peters (D-MI), is honestly closer to being rated Likely Democratic than Toss-up. Both Peters and Biden have consistently posted leads in the state, and Republican pessimism about Michigan at the presidential level seems to be growing, which has to bleed down to the Senate level. John James (R), who is taking a second run at the Senate after losing to Sen. Debbie Stabenow (D-MI) in 2018, has been outraising Peters, but only by relatively small margins.

Texas is kind of like the regular Senate race in Georgia, except that former congressional candidate MJ Hegar (D) doesn’t have the resources that Ossoff does and Texas may vote overall to the right of Georgia for president (as it did in 2016 and has in every presidential election since 1988).

Speaking of Georgia, we are moving the special Georgia Senate race from Leans Republican to Likely Republican for several reasons. First of all, we already mentioned the possibility of a runoff in the other Georgia seat, and that Democrats face certain hardships in Georgia runoffs. A runoff is virtually guaranteed in the special race because it is an all-party primary and there are many candidates on both sides. Additionally, it is not even clear that the Democrats will advance a candidate to the runoff: appointed Sen. Kelly Loeffler (R-GA) and her top GOP challenger, Rep. Doug Collins (R, GA-9), often finish atop polls, while the Rev. Raphael Warnock (D), the choice of national Democrats, sometimes lags behind Matt Lieberman (D), the son of former Sen. Joe Lieberman (D-CT), with former U.S. Attorney Ed Tarver (D) also garnering some support. So Democrats have work to do to just get into the runoff, and if they get there, they have to deal with the same turnout problems that have beguiled them in past runoffs. So the Republicans have a few important backstops in this race.

Finally, there are a few Likely Republican seats that one could put in the final slot. We decided to go with Alaska, where Sen. Dan Sullivan (R-AK) is running for a second term against doctor Al Gross, an independent/Democrat. Others might put Sen. Lindsey Graham (R-SC) in this spot, but despite some close polls, it is just really hard for a Democrat to get a high enough share of the vote to win in such a racially polarized state (Jones has a similar problem in another racially divided Deep South state, Alabama). Alaska’s electorate, though also Republican-leaning, is more fluid, and we see it as a more plausible — though still unlikely — Democratic upset target.

Conclusion

Overall, the battle for the Senate is close, although we would probably rather be the Democrats than the Republicans at the moment. The reason is basically that, of the three decisive Toss-ups in our ratings, we would probably pick the Democrats in at least two of them right now: both Maine and North Carolina are closer to Leans Democratic than Leans Republican. If Democrats win those, as well as Arizona and Colorado (while losing Alabama), they would forge a 50-50 tie, with what they hope is a Democratic vice president breaking ties.

Beyond these top races, the Democrats also have better second-tier targets than the Republicans: namely, the regular race in Georgia as well as Montana. We were prepared to add Kansas to that list, too, but Roger Marshall seems to have spared the GOP that additional headache.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s