KEY POINTS FROM THIS ARTICLE
— Republicans remain on track to win at least 51 Senate seats in November, which would allow them to take control of the chamber.
— Democrats have broadly been holding their own in other key races they are defending, but the upside potential for Republicans to get beyond 51 is clearly there, and the size of an eventual Republican majority, if it materializes, looks to be determined in the Industrial North.
— Republicans have had to play some defense in their own states, with Texas relatively close in polling and Nebraska hosting a surprisingly interesting contest. But we continue to favor them in all the states they currently hold.
Map 1: Crystal Ball Senate ratings
The Senate picture
As we survey the Senate with less than three weeks to go until the election, Republicans clearly remain on course to win at least 51 Senate seats in 2024, which would give them a Senate majority regardless of what happens in the presidential race. The bigger question down the stretch is not whether Republicans will win the majority but rather how big of a majority they can get—an important factor that will have both profound implications for governance in 2025 and shape future battles for the majority. In the closing weeks, the focus of the Senate campaign has landed where competitive American political battles often do: the Midwest/Industrial North, specifically Michigan, Ohio, Pennsylvania, and Wisconsin, all races that Democrats are defending.
To review, the basic path to a Republican majority of 51 seats involves the GOP holding all of their current seats, which we continue to favor them to do, while flipping the open seat in West Virginia and defeating Sen. Jon Tester (D-MT). Since we moved the Montana race to Leans Republican in early September, there’s been little indication that Tester has pulled out of his polling deficit, and he trails businessman Tim Sheehy (R) by roughly half a dozen points In public numbers. While Republicans have had to address a few weaknesses in seats they currently hold, most notably involving Sen. Deb Fischer’s (R-NE) troubles against independent labor leader Dan Osborn in Nebraska and Sen. Ted Cruz’s (R-TX) small polling leads against Rep. Colin Allred (D, TX-32) in Texas, we continue to give the Republicans clear edges in these races (more on them in the second half of this article).
Remember that Democrats face a hellacious defensive assignment on this map. Including the various independents who caucus with them, Democrats are defending 23 of the 34 seats on this year’s map. That includes defending seven Senate seats in states that voted for Donald Trump at least once for president (Montana, Ohio, and West Virginia backed him twice, and Arizona, Michigan, Pennsylvania, and Wisconsin backed him once). It’s not out of the question that all seven could vote for Trump again in 2024—as could an eighth state Democrats are defending, Nevada, which Trump lost in his first two elections—giving the Democrats a tremendous amount of exposure. The Republicans are not defending any states that voted Democratic for president in either 2016 or 2020, and none of these states seem likely to do so in 2024. (Update: This paragraph has been corrected to reflect that Nevada did not vote for Trump in either 2016 or 2020.)
As we reach the final stretch of this campaign, one thing you could say for Democrats is that it’s not as bad for them as it could be.
Yes, West Virginia is gone. Montana may not be quite over, but Tester is definitely an underdog. Based on public data, though, Democrats lead in the other six races to varying degrees. There was a world in which Sen. Sherrod Brown (D-OH) would be in as bad of shape as Tester at this stage of the campaign, but it does not seem like he is even as businessman Bernie Moreno (R) is nipping at Brown’s heels. There also was a world in which Democrats would be trailing in one or more presidential battleground state Senate races, but that doesn’t seem to be the case, either.
Table 1 shows the FiveThirtyEight presidential and Senate polling averages as of Friday morning in the 11 races we rate as something other than safe.
Table 1: Presidential vs. Senate polling
Note: * denotes that there is no Democratic candidate, with independent Dan Osborn running as the challenger to Sen. Deb Fischer (R-NE). In the presidential and Senate columns, a positive number is a Democratic edge and a negative number is a Republican edge. In the difference column, a positive number means a Senate candidate is outperforming his or her party’s presidential candidate, and a negative number means an underperformance. Finally, a positive number in the far right margin column means the Democratic margin is better in the Senate race than the Democratic presidential margin, and a negative number means the Republican Senate margin is better than the Republican presidential margin.
Source: FiveThirtyEight presidential and Senate polling averages as of Friday, Oct. 19
The Democratic Senate margin is better than Harris’s margin in nearly every race, and their share of the vote is also higher than Harris almost across the board, although sometimes the differences in vote share are tiny. Predictably, the only race where the Republican is running ahead of Trump is Maryland, where popular former Gov. Larry Hogan will certainly outrun Trump in a cobalt-blue state, but likely not by enough to win. Republicans are running behind Trump’s share of the vote everywhere else, with the biggest laggard being Fischer in Nebraska.
The belief among Republicans, generally, is that the undecideds in their key targets are disproportionately Trump voters who won’t split their tickets when it’s all said and done. That is not an unreasonable viewpoint, based on history, although there does not have to be a perfect symmetry between the presidential and Senate vote shares (that didn’t happen in 2016 or 2020, even though only a single state, Maine, voted for different parties in its Senate and presidential race in either of those years). There is a consistency in the polling pattern—again, Democrats are just broadly doing better than Harris in almost all of these contests. Is this actually going to happen?
The possibility of it is part of the reason we continue to rate Michigan, Pennsylvania, and Wisconsin as Leans Democratic instead of moving them to Toss-up, even though we freely admit these “Leans” are shaky as Republicans have made up some ground in these races, and the 4-5-point average leads currently enjoyed by the Democrats in those three races are likely still at least a little inflated. But our basic feeling remains the same: Democrats should win these races if Harris carries them for president, and they may have a chance to do so even if Trump does.
The Republicans in these races—businessmen Dave McCormick (PA) and Eric Hovde (WI) and former Rep. Mike Rogers (MI)—likely need Trump to win their respective states, and they are still working to catch up to his share of the vote, as Table 1 shows. Senate Leadership Fund, the monster outside GOP spending group, spelled out some of the dynamics of their races in a recent, candid polling memo uncovered by Politico. Democrats are portraying these races as highly competitive, too: Sens. Tammy Baldwin (D-WI) and Bob Casey (D-PA), as well as Rep. Elissa Slotkin (D, MI-7) have telegraphed their own unease about the state of their races.
One thing that does give us pause about the polling advantage for the Democrats in these races is that we’ve seen Republicans outperform their polls in Senate races held in each state concurrent with the 2016 (Wisconsin and Pennsylvania) and 2020 (Michigan) presidential elections. Back in 2016, Sen. Ron Johnson (R-WI) almost always trailed polls with ex-Sen. Russ Feingold (D) only to come back and win, and now-former Sen. Pat Toomey (R-PA) trailed almost all the polls late in the 2016 race but ended up winning. In 2020, Sen. Gary Peters (D-MI) won a closer-than-expected race against now-Rep. John James (R, MI-10) despite very consistent leads in polls. In other words, some of the polling problems at the presidential level in those states in both years seemed to extend to the Senate races as well. In 2016, Toomey and Johnson ended up running, respectively, 0.7 and 2.6 points ahead of Trump’s margin; in 2020, James ended up running ahead of Trump’s margin by 1.1 points (although he still lost by 1.7 points). We do think the data suggest that to the extent there are differences between the Senate and presidential races in Michigan, Pennsylvania, and Wisconsin this election, the Democrats are likelier to benefit, but we couldn’t blame anyone for looking at recent history and disagreeing with that assessment.
Overall, we want to note this explicitly—the Republicans do really have a chance to break this thing open and grab a sizable Senate edge in this election, which we would categorize as 53 seats or more. The Industrial North—Ohio plus the trio of presidential battlegrounds—will tell the tale: Republicans likely would need to win two of these four to hit that 53-seat mark, barring surprises elsewhere on the map.
Speaking of, Arizona and Nevada seem to have moved more to the sidelines of the intensely competitive map. Both Sen. Jacky Rosen (D-NV) and Rep. Ruben Gallego (D, AZ-3) have more solid leads and neither side views them as being quite as competitive as the Industrial North trio. It is a bit odd, frankly, because it’s also quite possible that Arizona and Nevada will vote to the right of the Industrial North states for president, if only very modestly. This is why we’re holding these races as Leans Democratic, even though one could conceivably make the case for Likely Democratic.
A closer look at the Democratic target list
Moving beyond the seats that are currently held by Democrats, we continue to see Florida, Nebraska, and Texas as Democrats’ only real offensive opportunities, although all three remain in the Likely Republican category.
In the Texas race, Cruz and Allred both took the stage on Tuesday night and participated in what was scheduled to be the sole general election debate of the contest. Each nominee generally stuck to themes that they’ve highlighted in their ads: Cruz, for instance, tried to disqualify Allred by casting him as a partisan liberal, while Allred emphasized abortion rights. While the debate may not move the needle too much in the race, both sides were already circulating clips of it—something that will almost certainly help with fundraising.
Speaking of fundraising, also on Tuesday, we got fresh FEC data aggregated by Rob Pyers, who runs the indispensable CATargetBot. From July to September, Allred raised nearly double what Cruz took in ($30 million compared to $17 million), although Cruz went into the final stretch of the race with about $10 million more in the bank (as of Oct. 1, he had $14 million to Allred’s $3 million, although we also need to remember that fundraising for both has of course been going on for the last couple of weeks, which is not included in this report). Not surprisingly, ad spending in this race so far has cut roughly 2-to-1 in favor of Allred, although Cruz was set to have an advantage in future reservations. The big outside Senate groups on either side are not spending money on independent expenditures in the race, but both the Democratic Senatorial Campaign Committee and National Republican Senatorial Committee have been spending money with their candidates on coordinated advertising.
In public polls, Cruz has generally led by margins slightly larger than his nearly 3-point reelection win from 2018. Referring back to Table 1, Cruz has typically posted margins a few points weaker than Donald Trump’s showing in his state, although Allred had rarely claimed more than a 45% share of the vote. The most recent New York Times/Siena College poll of Texas showed Trump leading Harris 48%-41% while Cruz was up 48%-44%. The aforementioned SLF polling memo showed Cruz up just 1 point, and that’s also a Republican poll, too. So there are some real indications the race is close.
Staying in the Sun Belt, Sen. Rick Scott (R-FL) has led former Rep. Debbie Mucarsel-Powell (D, FL-26) in every poll of the race so far, although his aggregates were perhaps buoyed by a recent NYT/Siena poll that had him up by a larger-than-usual 49%-40%. While Mucarsel-Powell has fundraised competitively, especially for a nominee in a race that we don’t place in either of our more competitive categories, Scott has not been reluctant to use his personal wealth in his campaigns (just in this past fundraising quarter, he self-financed nearly $2 million).
Something keeping us from pushing Florida or Texas into the Leans Republican category is presidential partisanship. Five of our seven presidential Toss-up states have Senate races this year, and we put all of them in the Leans Democratic category, although, as mentioned above, that designation is somewhat tenuous in several cases. But we can see those Democratic incumbents—or, in the case of Arizona and Michigan, well-funded nominees—outpacing Harris by just enough to hold on if Trump very narrowly carries their states. But Trump appears on track to at least replicate his 2020 showing in Florida and Texas, which we feel should be enough to insulate Cruz and Scott. Texas remains the better offensive option for Democrats.
In what might be the cycle’s most unconventional race, Dan Osborn has remained competitive, at least in polling, in Nebraska as pro-Fischer forces have begun to spend more heavily. Earlier this week, an Osborn internal poll showed him leading Fischer 50%-44%. One of the reasons we moved this race onto the board originally as Likely Republican was that Osborn’s polling suggested he was moving beyond the low-40s. There has been a frustrating lack of independent polling, and if a quality, independent pollster had shown him hitting 50%, we’d be more tempted to make the race even more competitive. That said, Fischer recently relayed her own internal poll, where she led 48%-42%–not an especially strong showing for her, considering Trump was up by 20 points in the same poll, and fellow Sen. Pete Ricketts (R), who is also up this year, has typically performed more like a “generic” Republican. This week, Fischer released another internal poll showing her up a similar amount.
So, what might a potential Osborn upset look like? Map 1 applies a uniform swing to the 2020 presidential two-party result to show what a potential 50-50 result would look like in the state.
Map 2: Hypothetical tied race in Nebraska
Considering both Districts 1 and 2 are already markedly less Republican than the rest of the state, Osborn would probably have to carry the former by mid-single digits and poll around 60% in the former. Osborn would also probably need to take about one-third of the vote in the geographically vast 3rd District, which is one of the most Republican districts in the nation, although the Osborn campaign has emphasized to us that they’ve kept a presence in western Nebraska. Still, these are high benchmarks for Osborn to hit.