If it's in the News, it's in our Polls. Public opinion polling since 2003.


Elections 2023: Democrats Enjoy a Strong Night

A Commentary By Kyle Kondik and J. Miles Coleman

They score victories in five of the six key races we were watching, although off-year elections do not necessarily predict the following year.


— Democrats won five of the six key races we were watching in Tuesday night’s elections, turning in a strong showing just a couple of days after a series of bad polls for President Joe Biden left some Democrats shook as the presidential race looms next year.

— The 2023 elections have limited predictive value, in large part because of the key differences between the Tuesday night results and what we should expect next year in the states we were watching.

Sifting through last night’s results

Political news over the weekend was dominated by the release of New York Times/Siena College polling that showed President Joe Biden trailing former President Donald Trump in several key 2024 battlegrounds. The polls contained some eye-popping and frankly hard-to-believe crosstabs showing Biden doing very poorly for a Democrat among both young and nonwhite voters — although it is worth noting that such findings are hardly limited to those polls.

A couple of days later, the United States went to the polls for an off-year election. There were a half-dozen key races we were watching: the Kentucky and Mississippi gubernatorial races; an abortion rights ballot issue in Ohio; a Pennsylvania state Supreme Court race; and both chambers of the Virginia state legislature. Democrats won five of these six races, losing only the Mississippi gubernatorial race.

So what gives?

As analysts, it is easier when indicators all point in the same direction. So had Republicans done markedly better in the big races on Tuesday, one could point to both the polls and those results and notice a negative trend for the White House. But even then, that wouldn’t necessarily have been predictive of next year — two years ago, the Republicans did do well in the midst of emerging weaknesses for Biden: They swept Virginia, made inroads in New Jersey and performed competitively in the gubernatorial race, and won a state Supreme Court race in Pennsylvania. But the midterm a year later featured much more mixed results, as Republicans flipped the House in underwhelming fashion and Democrats netted a Senate seat and made gains in state governments. And just like that hypothetical Republican overperformance wouldn’t have forecasted the future, neither does this actual Democratic overperformance have to forecast it, either. Off-year elections feature smaller electorates and don’t feature presidential candidates at the top of the ballot.

We can also see how next year’s presidential results are going to obviously differ from this year’s election results in many of the key states we were watching last night.

For instance, in Kentucky, Gov. Andy Beshear (D) won without too much drama, beating state Attorney General Daniel Cameron (R) by 5 points and outrunning Biden’s 2020 26-point statewide loss by 31 points. That of course means nothing for the state’s presidential trajectory — Beshear’s more modest win in 2019, after all, was bookended in 2016 and 2020 by landslide Trump wins. Across the Ohio River in the Buckeye State, Ohio voters enshrined robust reproductive rights protections into the state constitution by a margin that it appears will be almost identical to an earlier August vote, when voters rejected an effort to make constitutional amendments harder to pass. After that August vote, one of us argued in Politico Magazine that such ballot issues are not really reflective of where Ohio is in partisan races; that analysis still stands. In Mississippi, Gov. Tate Reeves (R) won reelection by a margin that appears to be very similar to his 5-point win in 2019; that state of course is more Republican in presidential races.

Meanwhile, and as we’ll address in more detail below, Virginia is a state where the off-year electorate is more Republican than what one would expect in a presidential year, to the point where even a GOP victory in both chambers would not have really changed our belief that Virginia is clearly a Democratic-leaning state in presidential elections. The Democratic wins in the state House and Senate obviously are more in line with where the state is federally.

Pennsylvania’s state Supreme Court race, which was the one key race happening in a core 2024 battleground, did appear to have a more federalized voting pattern two years after a state Supreme Court race won by Republicans featured some results that conformed a little less to federal voting patterns. Last night, Daniel McCaffery (D) won by what is currently a 6-point margin, bolstered by a strong performance in blue-trending southeast Pennsylvania (this came after the region produced smaller Democratic margins in the 2021 race — this may very well be a Dobbs effect). However, one has to remember that these non-presidential year races don’t predict the Keystone State — the Republicans’ close 2021 court victory had no predictive value for the Democrats’ smashing state victories a year later, for instance.

(It is also worth noting that the Democratic side had an overall money edge in the contests they won; Mississippi was relatively balanced in spending, with a small Republican advantage.)

Last night’s results have given Democrats a shot in the arm and have confounded the recent narrative about Democrats being in deep trouble next year. But it’s also true that these races in many respects differ from the election coming up next year. It may be the case that President Biden is in fact uniquely vulnerable, and that even former President Trump — himself dragged down by plenty of vulnerabilities that likely are not getting the kind of attention now that they will if he is renominated — could beat Biden. It may also be the case that polling a year out from an election is not predictive (and it often is not). Maybe the Democrats do just have an advantage now in smaller turnout, off-year elections as their base has absorbed many higher-turnout, college-educated voters while shedding lower-turnout voters who don’t have a four-year degree. Maybe the presidential year turnout will bring out more Trump voters and give the Republicans a clearer shot. About all we feel comfortable saying is that we should continue to expect the presidential race to be close and competitive — a boring statement, we know, but probably true.

One other thing before we take a quick look at some more granular results: In case it wasn’t already blindingly obvious before, the abortion issue in a post-Dobbs political environment continues to be a significant advantage for Democrats. Another abortion-related state ballot issue triumphed in Ohio; the abortion issue, we suspect, helped bring southeast Pennsylvania more in line with presidential partisanship, powering the Democratic victory there; even in Kentucky, Beshear was able to run against the strong anti-abortion stance of Cameron as part of his campaign; and in Virginia, abortion rights were a huge factor in the campaign, and Democrats came out on top (albeit narrowly). Abortion is not a one-size-fits-all automatic decisive winner for Democrats in every race, but it’s clear that the Democrats are just closer to where average Americans are on the issue than Republicans, and they have used the issue to great effect in many parts of the country.

Democrats lag behind Biden by just enough in Virginia

Earlier this year, when trying to come up with a baseline for Virginia’s legislative elections, we considered a January special election in the Virginia Beach area. At the time, now-state Sen. Aaron Rouse, a Democrat, flipped the seat that now-Rep. Jen Kiggans (R, VA-2) vacated as she moved on to Congress. Rouse won that special election by just under 2 points but ran about 8.5 points behind Joe Biden’s 2020 two-party margin in the seat. When we applied that swing to the entire legislature, we found that, if Democrats replicated Rouse’s performance everywhere, they would win 22 Senate seats and 52 in the House of Delegates. That is very close to what panned out last night in the Old Dominion.

In our Biden-to-Rouse swing scenario, Democrats would have carried SD-31, in Loudoun County, by about 5 points and flipped SD-16, a Republican-held but blue-trending seat in Henrico County, by about 8 points. Last night, Democrat Russet Perry won SD-31 by 5 points and in the 16th, Del. Schuyler VanValkenburg (D) ousted Sen. Siobhan Dunnavant (R) by just under 9 points — these victories gave Democrats 21 seats, the bare minimum for an outright majority in the chamber. Keeping to our hypothetical Biden-to-Rouse swing scenario, Democrats’ 22nd seat would have been the Hampton Roads area’s SD-24, although it would have been basically tied (we would have expected the Democrat to prevail by a fraction of a percentage point). Although it seems likely some outstanding provisional ballots will narrow the margin, Sen. Monty Mason (D) is currently down by a point and change to Danny Diggs (R), a former local sheriff.

In the House of Delegates, Democrats have flipped the chamber with at least 51 seats. According to the Virginia Public Access Project, the sole uncalled race as of this writing is HD-82 in the Petersburg area. In our election preview last week, we flagged this seat as a race to watch, noting that it seemed more of a must-win district for Republicans — currently, Del. Kim Taylor (R) leads her Democratic challenger Kimberly Pope Adams by less than 200 votes.

While the night was clearly a win for Virginia Democrats, any vacancies that arise over the next few years could be crucial for control of either chamber, assuming that the Democratic edge in each chamber remains just a single seat. This is especially true for the state Senate, which will not be up again until 2027, and where Lt. Gov. Winsome Earle-Sears (R) will be able to break any potential ties in favor of the GOP until 2025 (there is no tiebreaker in the House, so 50-50 as an actual tie is possible at some point down the line).

Ohio and Kentucky nuggets

— One thing that is easy to overlook when studying election maps is that metro areas often cross state borders. For instance, the combined southwest Ohio/northern Kentucky region around Cincinnati crosses into both states (and actually into Indiana, too, but that state did not have key statewide elections last night). There are three suburban Cincinnati counties in Ohio (Butler, Warren, and Clermont) and another three in northern Kentucky (Boone, Kenton, and Campbell). These counties are all blood red at the presidential level, all voting for Trump by about 20 points or more in 2020. But Democrats, be it Beshear himself in a partisan race or Issue 1 in a nonpartisan race on the other side of the Ohio River, did way better than Biden in all of them, as shown in Map 1. These suburban/exurban counties are not close to voting Democratic for president anytime soon, but they were flexible in both of these races.

Map 1: Greater Cincinnati, 2020 vs. 2023

— At the topline level, there was hardly any difference between Issue 1 in Ohio, the reproductive rights vote, and Issue 2, a ballot issue that legalized marijuana sales and use. As of Wednesday morning, Issue 1 was leading 56.6%-43.4%, while Issue 2 was doing slightly better, 57.0%-43.0%. However, the actual county-level voting patterns were notably different in some places, as the marijuana issue did less well in Northeast Ohio and some other places than the abortion rights issue but a lot better in many other parts of the state, winning several Appalachian counties ranging from east of Cincinnati to Athens, home to Ohio University (alma mater of one of the authors). We were not surprised that Athens turned in the highest “yes” vote in the state on the marijuana issue; we also could not conceal a chuckle when we saw that sparsely-populated Meigs County, Athens’s southern neighbor and home to the legend of “Meigs Gold,” narrowly backed the marijuana issue despite strongly opposing the abortion issue. Undergraduate laughs aside, we suspect that for some rural Ohio counties, backing marijuana may have been an economic issue as much as anything else. Map 2 shows the differences between the two issues at the county level.

Map 2: Issue 1 (abortion) versus Issue 2 (marijuana) in Ohio

Kyle Kondik is a Political Analyst at the Center for Politics at the University of Virginia and the Managing Editor of Sabato's Crystal Ball.

J. Miles Coleman is an elections analyst for Decision Desk HQ and a political cartographer. Follow him on Twitter @jmilescoleman.

See Other Political Commentary by Kyle Kondik.

See Other Political Commentary by J. Miles Coleman.

See Other Political Commentary.

Views expressed in this column are those of the author, not those of Rasmussen Reports. Comments about this content should be directed to the author or syndicate.

Rasmussen Reports is a media company specializing in the collection, publication and distribution of public opinion information.

We conduct public opinion polls on a variety of topics to inform our audience on events in the news and other topics of interest. To ensure editorial control and independence, we pay for the polls ourselves and generate revenue through the sale of subscriptions, sponsorships, and advertising. Nightly polling on politics, business and lifestyle topics provides the content to update the Rasmussen Reports web site many times each day. If it's in the news, it's in our polls. Additionally, the data drives a daily update newsletter and various media outlets across the country.

Some information, including the Rasmussen Reports daily Presidential Tracking Poll and commentaries are available for free to the general public. Subscriptions are available for $4.95 a month or 34.95 a year that provide subscribers with exclusive access to more than 20 stories per week on upcoming elections, consumer confidence, and issues that affect us all. For those who are really into the numbers, Platinum Members can review demographic crosstabs and a full history of our data.

To learn more about our methodology, click here.