Ed Note: On Sunday, we published our breakdown of the four core Senate battleground states that are likeliest to decide the majority. Today, we look at the five states closer to the edges of our board: Wisconsin, North Carolina, Ohio, New Hampshire, and Colorado.
This was one of the initial races we moved towards Democrats In the wake of the summer Democratic electoral resurgence, and we had shifted it from Leans Republican to Tossup in late August. At the time, Democratic Lieutenant Governor Mandela Barnes enjoyed a healthy lead over incumbent senator Ron Johnson, with his blank-slate candidacy allowing him to be viewed as a generic Democrat against an extremely unpopular incumbent with underwater favorables. We had noted, however, that Barnes’ candidacy and public image were relatively untested, and that a good 40% of the electorate had yet to form an opinion of him, giving Republicans an opening to change the dynamics of the race.
Over the last two months, that lane has been seized by the GOP to maximum effect, if polling is any indication, with the Republicans launching a barrage of attack ads against Barnes centered around his relatively liberal stance on criminal justice. This has resulted in Barnes’ favorables cratering, to the point where they are now underwater. In such a matchup, we would expect the incumbent Republican to win, especially in an environment that is more Republican than in 2020, when Biden won Wisconsin by a mere 0.6%.
Given the way the race has unfolded over the last few weeks, it is no surprise that Democrats have seen their standing in Wisconsin erode in the recent polling. Johnson has seen his margin grow in each of Marquette University’s last 3 statewide polls — in August, he was down 7, but was up 1 by September and by 6 in the October survey of likely voters. Barnes has seen his favorables go from 37 favorable, 22 unfavorable in August to 39 favorable, 40 unfavorable. Meanwhile, Johnson’s favorability splits are at 41/45.
It, however, would be a huge mistake to say that Barnes is out of this race just yet. A CBS/YouGov poll conducted in early October showed the race essentially tied, with Johnson up 50 to 49. Our understanding is that private polling on both sides continues to show a very tight and competitive race, and we would not count the Democratic candidate out here, especially given that Johnson’s favorables are still atrocious and are not necessarily immune from further erosion. In further indications of a tightly contested race, Democrats continue to spend heavily here as well, with the Senate Majority PAC adding $2M more recently to bring the combined Senate Democratic PAC spending in Wisconsin to $16M so far. The state is also heavily pro-choice, and if special elections in demographically similar areas are any indication, Democrats could outperform pre-election projections here.
With that said, we cannot comfortably state that this is a race where Democrats can make a credible case to be favored or at even odds against Johnson, and so we shifted this race from TOSSUP to LEANS REPUBLICAN. This is the type of seat that might actually have been Democratic-leaning in a bluer cycle. However, every indicator we see points to an environment that is less Democratic than the one in 2020, and the fundamentals do not favor the president’s party here.
No Senate race has flown under the radar as much as the matchup between Cheri Beasley and Ted Budd. Beasley, the former Chief Justice of the state Supreme Court, has been running a formidable campaign on the airwaves and has outraised Budd significantly, spending $26M herself as compared to Budd’s $9M. Perhaps for this reason, she continues to poll within the margin of error against the Republican congressman in polls, necessitating significant outside spending against her as a result; according to OpenSecrets, outside groups have poured in $32M against Beasley.
Until a week ago, there was little sign of this spending causing any poll movement, with the two being virtually tied in every poll. In the most recent East Carolina University October poll, however, Budd led 50-44, hitting a majority for the first time. While North Carolina has no runoff rule, this is significant solely because both candidates had previously been stuck in the low 40s in most polls, and there was no clear indication on how the undecideds would break. Our understanding is that private surveys of the race indicate a very tight election, with Beasley actually having some room to grow with undecideds. But Budd hitting 50 in the ECU poll might imply that the Republican position in the race is now a bit better than the R+2 polling average suggests, and it will be interesting to track how this changes.
North Carolina is ultimately a state in which Democrats have faced heartburn year after year; their biggest Achilles heel happens to be rural Black voter turnout, and this tends to plunge in midterms. However, running Black candidates does appear to help significantly with Black voter turnout, and so there is certainly a lane for Beasley here, especially if she manages to keep the Black share of the electorate high.
Dobbs is not likely to move the needle as strongly in North Carolina, which is more racially polarized and more religious than a lot of midwestern white states that Democrats may be poised to see gains in. While she did outrun Biden in her Chief Justice election in 2020 and likely did put up better numbers with whites, that was not enough to carry her to victory, and it is somewhat difficult to see her improving on those numbers in 2022 to the point where she would beat Budd. Her best hope may be via Libertarian Shannon Bray siphoning away enough from Budd to let Beasley win with a plurality. There is a clear lane through which Beasley can win, but it would involve three factors (a better vote share with whites, high Black turnout, and a third party vote helping Democrats) combining at once. For this reason, we rate it at LEANS REPUBLICAN.
Perhaps the biggest shock of this cycle comes in Ohio, where Democratic congressman Tim Ryan has kept the race surprisingly close against businessman JD Vance, defying virtually every indicator suggesting the state was a layup for Republicans. Poll after poll shows a competitive race, and although Ohio polls have traditionally overestimated Democrats of late, there are several flags here that point to the race actually being tight at the moment. For one, several polls do weight for recalled 2020 vote, and even with Trump +10 electorates, Ryan and Vance generally continue to poll within the margin of error of each other. Secondly, several of the same polls that show Ryan and Vance in a virtual dead heat also show DeWine up by over 20 points. While it is possible that these polls are over-representing college-educated or highly-engaged Republicans, our understanding is that high-quality private surveys continue to show a close race as well, albeit one where Vance is ahead by slightly more.
Vance has done himself no favors in this race and has struggled with anemic fundraising, resulting in Republicans needing to pour in millions of dollars from SuperPACs to close the gap between him and Ryan (FEC data aggregated by Rob Pyers suggests that Republican outside groups have outspent Democrats $24M to $3M).
Part of the problem for Ryan, however, is that he continues to poll in the mid 40s, with most surveys placing him at around 46% of the vote. The undecideds in virtually every poll skew heavily Republican and are generally non-degree holding whites, meaning they are more likely than not to break against the Democratic candidate. Vance’s favorables are not phenomenal, but they have rebounded from atrocious to respectable — a Suffolk poll that had him up 47-45 also showed his favorables four points above water, and while this is less than Ryan’s +9 rating, it is enough for us to conclude that Ryan has a tougher time than not with the undecideds.
With that said, our previous rating for this race was LIKELY REPUBLICAN, and we do not think that is a fair assessment in a race where both candidates often poll within the margin of error against each other three weeks before the election. Vance should win, but there is a small lane for Ryan, and this election could be closer than we would have expected a few months ago. We now rate this race as LEANS REPUBLICAN.
Maggie Hassan continues to swamp Don Bolduc in fundraising and advertising time, and she leads Don Bolduc by seven points in the FiveThirtyEight averages. Bolduc has put up the worst set of fundraising numbers we have seen this cycle in virtually any remotely-competitive state, and has made a series of controversial comments on issues ranging from abortion to the 2020 election, none of which are likely to play well in a Biden +7, heavily pro-choice state. Neither public nor private polling suggests this race is especially close, and so we have shifted it from LEANS DEMOCRATIC to LIKELY DEMOCRATIC.
Although Republicans continue to heavily play up their chances in the Centennial State, unseating an incumbent in the Biden +13.5 state may simply be too tough of a lift for them this year. Incumbent Democratic senator Michael Bennet continues to poll healthily ahead of businessman Joe O’Dea in every public survey, and there has been no serious evidence to suggest that this socially liberal and highly educated state is shaping up to be extremely close just yet. O’Dea may have a lane still if the environment degrades further for Democrats, and his vocally anti-Trump stance may help him with some swing voters in the Denver suburbs, but he needs to flip a very large number of votes in order to do so. In a year that is likely to be much closer to R+0 than it is to R+7, we think Bennet is strongly favored to keep his seat, and it remains on the very edge of the board at LIKELY DEMOCRATIC.
I’m a software engineer and a computer scientist (UC Berkeley class of 2019 BA, class of 2020 MS) who has an interest in machine learning, politics, and electoral data. I’m a partner at Split Ticket, handle our Senate races, and make many kinds of electoral models.